HK40042230A

HK40042230A - System and method for determining blood pressure of subject

Info

Publication number: HK40042230A
Application number: HK62021031919.5A
Authority: HK
Inventors: Ziming DENG; Chuanmin WEI; Ying Lu; Zijian HUANG; Jiwei Zhao; Kezheng MA; Mengcheng HU
Original assignee: Vita-Course Technologies Co., Ltd.
Filing date: 2018-04-13
Publication date: 2021-08-27

Description

System and method for determining blood pressure of a subject

Technical Field

The present application relates generally to systems and methods in the healthcare-related field. More particularly, the present application relates to systems and methods for determining blood pressure of a subject.

Background

Blood pressure measurement can be divided into invasive blood pressure measurement and non-invasive blood pressure measurement. Invasive blood pressure measurements are commonly used in medical procedures or medical studies and need to be performed by medical professionals. Non-invasive blood pressure measurement is an indirect blood pressure measurement method. A blood pressure meter is a popular non-invasive blood pressure measuring device. It comprises an inflatable cuff which collapses and then releases the underlying artery in a controlled manner, and a mercury or mechanical manometer for measuring pressure. However, frequent measurements with a sphygmomanometer can be uncomfortable for a subject due to the frequent inflation compressing the subject's blood vessels. In addition, the size of the cuff, the elastic effect, and the posture of the subject can affect the accuracy of the measured blood pressure during the measurement.

Another non-invasive blood pressure measurement system may collect a plurality of physiological characteristic data of a plurality of subjects (e.g., patients or persons) as sample data to build a model that predicts the subject's blood pressure. Because the amount of physiological characteristic data for multiple subjects is large, the dimensionality of the characteristics is high, and includes outlier sample data, it is difficult to build a model for predicting blood pressure in a subject. Therefore, there is a need to provide an accurate, efficient and personalized blood pressure prediction model by reducing the high dimension of features without deleting outlier sample data.

Disclosure of Invention

According to one aspect of the present application, a system for determining blood pressure is provided. The system may include at least one storage medium comprising a set of instructions, a communication platform connected to a network; at least one processor in communication with the at least one storage medium. The at least one processor, when executing the set of instructions, may be configured to receive a request from a terminal to determine a blood pressure of a first subject. The at least one processor may acquire data related to the first subject, wherein the data related to the first subject may include data related to cardiac activity of the first subject and personal information related to the first subject. The at least one processor may extract a target feature associated with the first subject from the data associated with the first subject. The at least one processor may determine an initial blood pressure of the first subject using a predictive model based on the target feature associated with the first subject. The at least one processor may determine a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure. The at least one processor may also transmit the predicted blood pressure of the first subject to the terminal in response to the request.

According to another aspect of the present application, a method implemented on a computing device having at least one processor, a memory, and a communication platform connected to a network is provided for determining blood pressure. The method may include receiving a request from a terminal to determine a blood pressure of a first subject. The method may include obtaining data related to the first subject, wherein the data related to the first subject may include data related to cardiac activity of the first subject and personal information related to the first subject. The method may include extracting a target feature associated with the first subject from data associated with the first subject. The method may include determining an initial blood pressure of the first subject using a predictive model based on the target feature associated with the first subject. The method may include determining a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure. The method may further comprise transmitting the predicted blood pressure of the first subject to the terminal in response to the request.

According to another aspect of the present application, a non-transitory computer-readable medium is provided that includes at least one set of instructions for determining blood pressure. The at least one set of instructions, when executed by the at least one processor, may cause the at least one processor to receive a request from a terminal to determine a blood pressure of a first subject. The at least one processor may also acquire data related to the first subject, wherein the data related to the first subject may include data related to cardiac activity of the first subject and personal information related to the first subject. The at least one set of instructions may cause the at least one processor to extract a target feature associated with the first subject from data associated with the first subject. The at least one set of instructions may cause the at least one processor to determine an initial blood pressure of the first subject using a predictive model based on a target feature associated with the first subject. The at least one set of instructions may cause the at least one processor to determine a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure. The at least one set of instructions may also cause the at least one processor to send the predicted blood pressure of the first subject to the terminal in response to the request.

In some embodiments, to acquire data related to cardiac activity of a first subject, the at least one processor may be further directed to communicate with a device connected to the first subject, the device configured to detect cardiac activity of the first subject and generate a signal, and receive data related to cardiac activity of the first subject generated based on the signal from the device.

In some embodiments, the predictive model may be generated via a first training process. The first training process may include: obtaining historical data relating to at least two second subjects and at least two historical blood pressure measurements associated with the at least two second subjects, wherein the at least two second subjects may include the first subject, and the historical data relating to the at least two second subjects may include data relating to cardiac activity of the at least two second subjects and historical personal information relating to the at least two second subjects. The first training process may further include generating an initial predictive model based on historical data relating to the at least two second subjects and a plurality of historical blood pressure measurements associated with the at least two second subjects; and generating a first subject-related predictive model based on the initial predictive model and at least a portion of the historical data associated with the first subject.

In some embodiments, generating an initial predictive model based on historical data associated with the at least two second subjects and at least two historical blood pressure measurements associated with the at least two second subjects may include extracting a first set of features from the historical data associated with the at least two second subjects; determining a second set of features based on the first set of features, the second set of features having dimensions smaller than dimensions of the first set of features; clustering historical data associated with at least two second subjects into one or more clusters; determining historical target features from the second set of features; for each cluster in one or more clusters, determining a sub-prediction model based on historical target features of historical data in each cluster and historical blood pressure measurement values corresponding to the historical data in each cluster; one or more sub-prediction modes corresponding to one or more clusters are designated as an initial prediction model.

In some embodiments, generating the initial predictive model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects may further comprise normalizing the historical data associated with the at least two second subjects prior to extracting the first set of features from the historical data associated with the at least two second subjects.

In some embodiments, the second set of features may be determined using principal component analysis techniques.

In some embodiments, generating a predictive model related to a first subject based on an initial predictive model and at least a portion of historical data related to the first subject may include extracting historical target features from the historical data related to the first subject; determining a target cluster from one or more clusters based on the historical target features of the historical data associated with the first subject; and assigning a sub-prediction mode corresponding to the target cluster as the prediction model associated with the first subject.

In some embodiments, to determine the predicted blood pressure of the first subject using the optimization model based on the initial blood pressure, the at least one processor may be further instructed to: initializing a first optimization model; and designating the initial blood pressure as a first blood pressure; and (3) performing iteration: generating a second blood pressure based on the first blood pressure using the first optimization model; determining whether a convergence condition is satisfied based on the first blood pressure and the second blood pressure; in response to determining that a convergence condition is satisfied, designating a first optimization model as the optimization model and designating a second blood pressure produced in an iteration as a predicted blood pressure; in response to determining that a convergence condition is not satisfied, updating the first optimization model and designating the second blood pressure produced in an iteration as the first blood pressure.

In some embodiments, to initialize the first optimization model, the at least one processor may be instructed to: obtaining historical blood pressure measurements for the first subject and historical target characteristics for the first subject; and initializing the first optimization model based on the historical blood pressure measurements of the first subject and the historical target characteristics of the first subject.

In some embodiments, the predicted blood pressure of the first subject may include a systolic pressure and a diastolic pressure. An optimization model may be used to predict systolic pressure based on diastolic pressure.

Additional features of the present application will be set forth in part in the description which follows. Additional features of some aspects of the present application will be apparent to those of ordinary skill in the art in view of the following description and accompanying drawings, or in view of the production or operation of the embodiments. The features of the present application may be realized and attained by practice or use of the methods, instrumentalities and combinations of aspects of the specific embodiments described below.

Drawings

The present application will be further described by way of exemplary embodiments. These exemplary embodiments will be described in detail by means of the accompanying drawings. These embodiments are non-limiting exemplary embodiments in which like reference numerals represent similar structures throughout the several views of the drawings, and wherein:

FIG. 1 illustrates an exemplary system configuration, shown in accordance with some embodiments of the present application, in which a medical system may be deployed;

FIG. 2 is a schematic diagram of exemplary hardware and software components of a computing device shown in accordance with some embodiments of the present application;

FIG. 3 is a schematic diagram of exemplary hardware and/or software components of an exemplary mobile device on which a user terminal may be implemented, according to some embodiments of the present application;

FIG. 4 is a block diagram of an exemplary processor shown in accordance with some embodiments of the present application;

fig. 5 is a flow diagram illustrating a process and/or method for providing a first subject's blood pressure to a terminal in response to a request, according to some embodiments of the present application;

fig. 6A and 6B are block diagrams of another exemplary processor, shown in accordance with some embodiments of the present application.

Fig. 7 is a flow diagram of a process and/or method for generating a prediction model that predicts an initial blood pressure of a first subject, according to some embodiments of the present application;

FIG. 8 is a block diagram of an exemplary initial prediction model determination module shown in accordance with some embodiments of the present application;

FIG. 9 is a flow diagram illustrating a process and/or method for generating an initial predictive model according to some embodiments of the present application;

FIG. 10 is a block diagram of an exemplary predictive model determination module shown in accordance with some embodiments of the present application;

fig. 11 is a flow diagram of a process and/or method for determining a predictive model related to a first subject, according to some embodiments of the present application;

FIG. 12 is a block diagram of an exemplary predicted blood pressure determination module shown in accordance with some embodiments of the present application; and

fig. 13 is a flow diagram of a process and/or method for determining a predicted blood pressure of a first subject using an optimization model, according to some embodiments of the present application.

Detailed Description

The following description is presented to enable one of ordinary skill in the art to make and use the invention and is provided in the context of a particular application and its requirements. It will be apparent to those skilled in the art that various modifications to the disclosed embodiments are possible, and that the general principles defined in this application may be applied to other embodiments and applications without departing from the spirit and scope of the application. Thus, the present application is not limited to the described embodiments, but should be accorded the widest scope consistent with the claims.

The terminology used in the description presented herein is for the purpose of describing particular example embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises," "comprising," and/or "including," when used in this specification, specify the presence of stated features, integers, steps, operations, components, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, and/or groups thereof.

The features and characteristics of the present application, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description of the drawings, which form a part hereof. It is to be understood, however, that the drawings are designed solely for the purposes of illustration and description and are not intended as a definition of the limits of the application. It should be understood that the drawings are not to scale.

Flow charts are used herein to illustrate operations performed by systems according to some embodiments of the present application. It should be understood that the operations in the flow diagrams may be performed out of order. Rather, various steps may be processed in reverse order or simultaneously. Also, one or more other operations may be added to the flowcharts. One or more operations may also be deleted from the flowchart.

One aspect of the present application relates to systems and methods for determining blood pressure of a first subject (also referred to herein as a target subject). The present application uses a predictive model to predict an initial blood pressure of a first subject based on data associated with the first subject. The predictive model is trained using a sample data set of blood pressure measurements associated with a large number of subjects. The data associated with the first subject may include data associated with cardiac activity of the first subject and personal information associated with the first subject. The present application further generates a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure. The optimization model utilizes the correlation between the systolic and diastolic pressures of the subject to determine the systolic and diastolic pressures during each iteration to obtain a more accurate blood pressure prediction.

Fig. 1 illustrates an exemplary system configuration, shown according to some embodiments of the present application, in which a medical system 100 may be deployed. The medical system 100 may be configured to monitor a physiological parameter of interest. Medical system 100 may include measurement device 110, server 120, external data source 130, terminal 140, and storage device 160. The various components of the medical system 100 may be connected to each other directly or indirectly via a network 150.

The measurement device 110 may be configured to detect a physiological phenomenon (e.g., cardiac activity) of the subject and generate a signal. The signal may be a cardiovascular signal. The signal may relate to or be used to determine a physiological parameter of interest. The measurement device 110 may include, for example, a clinical device, a home device, a portable device, a wearable device, or the like, or a combination thereof. As used herein, a clinical device may be a device that meets applicable requirements and specifications for use in a clinical environment, including, for example, a hospital, a doctor's office, a nursing home, and the like. The clinical device may be used by or with the assistance of medical personnel. As used herein, a home appliance may be a home appliance that meets applicable requirements and specifications to be used in a home or non-clinical environment. The home appliances may be used by professional providers or non-professional providers. The clinical device or the household device or a part thereof may be portable or wearable. Exemplary clinical devices include auscultation devices, oscillometric devices, ECG monitors, PPG monitors, and the like, or combinations thereof. Exemplary home devices include oscillometric devices, home ECG monitors, blood pressure meters, and the like, or combinations thereof. Exemplary portable devices include an oscillometric device, a portable ECG monitor, a portable PPG monitor, and the like, or combinations thereof. Exemplary wearable devices include glasses 111, shoulder straps 112, smart watches 113, foot chains 114, thigh straps 115, arm straps 116, chest straps 117, neck straps 118, and the like, or combinations thereof. The above-mentioned examples of the measurement device 110 are for illustrative purposes only and are not intended to limit the scope of the present application. The measurement device 110 may be in other forms such as a finger cuff, wrist band, brassiere, undergarment, chest strap, pulse oximeter, or a device related to the principles used in pulse oximetry, or the like, or combinations thereof.

For example only, the measurement device 110 is a wearable or portable device configured to detect and generate one or more cardiovascular signals. In some embodiments, the wearable device or portable device may process at least some of the measured signals, estimate a physiological parameter of interest based on the measured signals, present a result including the physiological parameter of interest in the form of, for example, an image, an audio alert, communicate with another device or server (e.g., server 120) or the like, wired or wireless, or the like. In some embodiments, the wearable device or portable device may communicate with another device (e.g., terminal 140) or a server (e.g., server 120). The device or server may process at least some of the measured signals, estimate a physiological parameter of interest based on the measured signals, and present a result including the physiological parameter of interest, for example in the form of an image, an audio alert, or a combination thereof.

In some embodiments, the operations of processing the generated signals, estimating physiological parameters, displaying results, or performing wired or wireless communication may be performed by an integrated device or a separate device connected or communicating thereby. Such integrated devices may be portable or wearable. In some embodiments, at least some of the standalone devices may be portable or wearable, or located in proximity to a subject, which measures signals or estimates or monitors physiological parameters of interest from the subject. By way of example only, the subject wears a measurement device 110 configured to detect and generate one or more cardiovascular signals; the generated one or more cardiovascular signals are transmitted to a smartphone configured to determine a physiological parameter of interest from the measured signals. In some embodiments, at least some of the standalone devices are located at a location remote from the subject. By way of example only, the subject wears a measurement device 110 configured to detect and generate one or more cardiovascular signals; the generated one or more cardiovascular signals are further sent to a server 120, the server 120 being configured to determine a physiological parameter of interest based on the measured signals; and the determined physiological parameter of interest may be communicated back to the subject or a user other than the subject (e.g., a physician, a caregiver, a family member related to the subject, etc., or a combination thereof).

In some embodiments, the measurement device 110 may include various types of sensors, such as electrode sensors, optical sensors, photoelectric sensors, pressure sensors, accelerometers, gravity sensors, temperature sensors, humidity sensors, and the like, or combinations thereof. The measurement device 110 may be configured to monitor and/or detect one or more types of variables including, for example, temperature, humidity, user or subject input, and the like, or combinations thereof. The gravity sensor may detect a posture of the subject under test. The posture may include a lying posture, a sitting posture, a standing posture, and the like. The temperature sensor may detect the temperature of a location near the measurement device 110. The humidity sensor may detect humidity in the area near the measurement device 110. The measurement device 110 may also include a positioning system, such as a GPS receiver or position sensor, and the position information may be transmitted over the network 150 to the server 120, external data sources 130, terminals 140, and the like, or combinations thereof. The location information and the measured signal may be transmitted simultaneously or sequentially. In some embodiments, the measurement device 110 may comprise one or more computer chips on which the functionality of the server 120 described below may be implemented.

The server 120 may be a cloud server. For example only, the server 120 may be implemented in a cloud server that may provide storage capacity, computing capacity, or the like, or a combination thereof. Server 120 may include a storage device configured to collect or store data. The data may include personal data, non-personal data, or both. The data may include static data, dynamic data, or both. In some embodiments, the static data may include various information about the subject, including identity, contact information, date of birth, health history (e.g., information on whether the subject has a history of smoking, information on previous surgery, food allergies, drug allergies, medical history, genetic medical history, family health history, etc., or combinations thereof), gender, nationality, height, weight, occupation, habits (e.g., habits related to health, such as exercise habits), educational background, hobbies, marital status, religious beliefs, and the like, or combinations thereof. In some embodiments, the dynamic data may include the current health of the subject, the medication being taken by the subject, the diet, physiological signals or parameters associated with the subject at a plurality of time points or over a period of time (e.g., pulse wave transit time (PTT), Systolic Blood Pressure (SBP), Diastolic Blood Pressure (DBP), etc.), or combinations thereof.

As used herein, a subject may refer to a human or animal whose signals or information are acquired and physiological parameters are determined or monitored. For example only, the subject may be a patient for whom cardiovascular signals are acquired and blood pressure is determined or monitored based on the acquired cardiovascular signals.

In some embodiments, the server 120 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system). In some embodiments, the server 120 may be local or remote. For example, server 120 may access information and/or data stored in terminals 140 and/or storage device 160 via network 150. As another example, server 120 may connect terminal 140 and/or storage device 160 to access stored information and/or data. In some embodiments, the server 120 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof. In some embodiments, the server 120 may be implemented on a computing device 200, the computing device 200 having one or more components shown in FIG. 2 in the present application.

In some embodiments, the server 120 may include a processing engine 122. Processing engine 122 may process information and/or data to perform one or more functions described in this disclosure. For example, the processing engine 122 may determine the blood pressure of the subject based on one or more personalized models, data related to the generated signals, and/or information related to the subject. In some embodiments, processing engine 122 may include one or more processing engines (e.g., a single core processing engine or a multi-core processor). By way of example only, the processing engine 122 may include one or more hardware processors, such as a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an application specific instruction set processor (ASIP), an image processing unit (GPU), a physical arithmetic processing unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller unit, a Reduced Instruction Set Computer (RISC), a microprocessor, or the like, or any combination thereof.

The external data source 130 may include various organizations, systems, devices, and the like, or combinations thereof. In some embodiments, the data sources 130 may include medical institutions, research institutions, legacy devices, peripheral devices, and the like, or combinations thereof. The medical or research institution may provide, for example, individual cases, clinical test results, experimental study results, theoretical or mathematical study results, algorithms suitable for processing data, and the like, or combinations thereof. Conventional devices may include cardiovascular signal measuring devices, such as mercury sphygmomanometers. The peripheral device may monitor and/or detect one or more types of variables including, for example, temperature, humidity, user or subject input, and the like, or combinations thereof. The above examples of external data sources 130 and data types are provided for illustrative purposes and are not intended to limit the scope of the present disclosure. For example, the external data sources 130 may include other sources and other types of data, such as genetic information related to the subject or his family.

The terminal 140 in the medical system 100 may be configured to process at least some of the generated signals, determine a physiological parameter of interest based on the generated cardiovascular signals, display a result including the physiological parameter of interest, e.g., in the form of an image, store data, control access to the medical system 100 or a portion thereof (e.g., access to personal data stored in the medical system 100 or accessible from the medical system 100), manage input/output from or related to the subject, and the like or combinations thereof.

The terminal 140 may include a mobile device 140-1, a tablet computer 140-2, a laptop computer 140-3, a built-in device 140-4 in a motor vehicle, the like, or any combination thereof. In some embodiments, mobile device 140-1 may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, and the like, or any combination thereof. In some embodiments, the smart home devices may include smart lighting devices, smart appliance control devices, smart monitoring devices, smart televisions, smart cameras, interphones, and the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, smart footwear, smart glasses, smart helmet, smart watch, smart garment, smart backpack, smart accessory, or the like, or any combination thereof. In some embodiments, the smart mobile device may include a smart phone, a Personal Digital Assistant (PDA), a gaming device, a navigation device, a point of sale (POS), or the like, or any combination thereof. In some embodiments, a virtual machineThe real device and/or augmented reality device may include a virtual reality helmet, virtual reality glasses, virtual reality eyeshields, augmented reality helmets, augmented reality glasses, augmented reality eyeshields, and the like, or any combination thereof. For example, the virtual reality device and/or the augmented reality device may include a google glass^TM、RiftCon^TM、Fragments^TM、GearVR^TMAnd the like. In some embodiments, the built-in device 140-4 in the motor vehicle may include an on-board computer, an on-board television, or the like.

The network 150 may facilitate the exchange of information and/or data. In some embodiments, one or more components of the medical system 100 (e.g., the measurement device 110, the server 120, the external data source 130, the terminal 140, and the storage device 160) may send information and/or data to other components in the medical system 100 via the network 150. For example, the server 120 may receive a request from the terminal 140 via the network 150 to determine the blood pressure of the subject. In some embodiments, the network 150 may be any form of wired or wireless network, or any combination thereof. By way of example only, network 150 may include a cable network, a wired network, a fiber optic network, a telecommunications network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), the Public Switched Telephone Network (PSTN), a bluetooth network, a ZigBee network, a Near Field Communication (NFC) network, the like, or any combination thereof. In some embodiments, the network 150 may include one or more network access points. For example, network 150 may include wired or wireless network access points, such as base stations and/or internet exchange points 150-1, 150-2, …, through which one or more components of medical system 100 may connect to network 150 to exchange data and/or information therebetween.

Storage device 160 may store data and/or instructions. In some embodiments, the storage device 160 may store data obtained from the terminal 140. In some embodiments, storage device 160 may store data and/or instructions that server 120 uses to perform or use to perform the exemplary methods described in this application. In some embodiments, storage device 160 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), the like, or any combination thereof. Exemplary mass storage devices may include magnetic disks, optical disks, solid state drives, and the like. Exemplary removable memories may include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tape, and the like. Exemplary volatile read and write memory can include Random Access Memory (RAM). Exemplary RAM may include Dynamic Random Access Memory (DRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Static Random Access Memory (SRAM), thyristor random access memory (T-RAM), zero capacitance random access memory (Z-RAM), and the like. Exemplary ROMs may include mask ROM (mrom), programmable ROM (prom), erasable programmable ROM (eprom), electrically erasable programmable ROM (eeprom), compact disk ROM (CD-ROM), digital versatile disk ROM, and the like in some embodiments, storage device 160 may execute on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-tiered cloud, and the like, or any combination thereof.

In some embodiments, a storage device 160 may be connected to the network 150 to communicate with one or more components of the medical system 100 (e.g., the measurement device 110, the server 120, the external data source 130, the terminal 140, and the storage device 160). One or more components of medical system 100 may access data or instructions stored in storage device 160 through network 150. In some embodiments, storage device 160 may be directly connected to or in communication with one or more components in medical system 100 (e.g., measurement device 110, server 120, external data source 130, terminal 140, and storage device 160).

In some embodiments, one or more components of the hospitalization system 100 (e.g., the measurement device 110, the server 120, the external data source 130, the terminal 140, and the storage device 160) may access the storage device 160. For example, server 120 may read and/or modify information for one or more users upon requesting a prediction of a subject's blood pressure.

In some embodiments, various components of medical system 100 may include storage devices for storing intermediate data and/or information. Such components may include, for example, measurement devices 110, servers 120, external data sources 130, terminals 140, and the like, or combinations thereof.

In some embodiments, the external data source 130 may receive data from the measurement device 110, the server 120, the terminal 140, etc., or any combination via the network 150. For example only, the external data source 130 (e.g., a medical facility or smart home system, etc.) may receive information related to the subject (e.g., location information, data from a cloud server or terminal, etc., or a combination thereof) based on data received from the measurement device 110 or terminal 140. In some other embodiments, the measurement device 110 may receive data from the server 120, the external data source 130, or the like, or any combination thereof, via the network 150. For example only, the measurement device 110 may receive information related to the subject (e.g., a current/historical health condition of the subject, a medication being taken by the subject, a medical being taken by the subject, a current/historical diet, a current emotional state, historical physiological parameters related to the subject (e.g., PTT, SBP, DBP, etc.), or a combination thereof). Further, the terminal 140 can receive data from the measurement device 110, the server 120, the external data source 130, the like, or a combination thereof. In some embodiments, the server 120 may store one or more personalized models for predicting blood pressure, and may send the personalized models to the measurement device 110 and the terminal 140.

It should be noted that the description regarding the configuration of medical system 100 is not intended to limit the scope of the present application. In some embodiments, the server 120 may be omitted, with all of its functionality migrated to the terminal 140. In some embodiments, both the server 120 and the terminal 140 may be omitted, migrating all of their functionality to the measurement device 110. In different embodiments, the system may include various devices or combinations of devices.

FIG. 2 is a schematic diagram illustrating exemplary hardware and software components of a computing device 200 on which the measurement device 110, server 120, external data source 130, terminal 140, and storage device 160 may be implemented according to some embodiments of the present application. For example, the processing engine 122 may be implemented on the computing device 200 and configured to perform the functions of the processing engine 122 disclosed herein.

Computing device 200 may be a general purpose computer or a special purpose computer, both of which may be used to implement the system of the present application. Computing device 200 may be used to implement any of the components described herein. For example, the processing engine 122 may be implemented on the computing device 200 by its hardware, software programs, firmware, or a combination thereof. Although only one such computer is shown for convenience, the computer functions associated with the medical system 100 described herein may be implemented in a distributed manner across a plurality of similar platforms to spread the processing load.

For example, computing device 200 may include a communication port 250 to connect to a network to enable data communication. Computing device 200 may also include a processor (e.g., processor 220) in the form of one or more processors for executing program instructions. An exemplary computing device may include an internal communication bus 210, various forms of program storage and data storage including, for example, a disk 270, Read Only Memory (ROM)230 or Random Access Memory (RAM)240 for storing various data files processed and/or transmitted by the computing device. In some embodiments, the computing device may also include program instructions stored in ROM230, RAM240, and/or other types of non-transitory storage media that are executable by processor 220. The methods and/or processes of the present application may be embodied in the form of program instructions. Computing device 200 also includes input/output component 260 to support input/output between the computer and other components. Computing device 200 may also receive programming and data via network communications.

For illustrative purposes only, only one CPU and/or processor is shown in FIG. 2. The use of multiple CPUs and/or processors is also contemplated. Thus, operations and/or method steps described in this disclosure as being performed by one CPU and/or processor may also be performed collectively or separately by multiple CPUs and/or processors. For example, if in the present application the CPUs and/or processors of computing device 200 perform steps a and B simultaneously, it should be understood that steps a and B may also be performed by two different CPUs and/or processors, either collectively or independently (e.g., a first processor performing step a, a second processor performing step B, or a first processor and a second processor performing steps a and B collectively).

Fig. 3 is a schematic diagram illustrating exemplary hardware and/or software components of an exemplary mobile device 300 on which a user terminal may be implemented according to some embodiments of the present application. As shown in fig. 3, mobile device 300 may include a communication platform 310, a display 320, a Graphics Processing Unit (GPU)330, a Central Processing Unit (CPU)340, input/output components (I/O)350, memory 360, and storage 390. In some embodiments, any other suitable component, including but not limited to a system bus or a controller (not shown), may also be included in mobile device 300. In some embodiments, the operating system 370 is mobile (e.g., iOS)^TM、Android^TM、WindowsPhone^TMEtc.) and one or more application programs 380 may be loaded from storage 390 into memory 360 for execution by CPU 340. The application 380 may include a browser or any other suitable mobile application for receiving and presenting information related to image processing or other information in the processing engine 122. Interaction of the information flow with the user may be accomplished through I/O350 and provided to processing engine 122 and/or other components of medical system 100 via network 150.

To implement the various modules, units, and their functions described herein, a computer hardware platform may be used as the hardware platform for one or more of the components described herein. A computer with user interface elements may be used to implement a Personal Computer (PC) or any other type of workstation or terminal device. The computer may also function as a server if appropriately programmed.

FIG. 4 is a block diagram of an exemplary processor shown in accordance with some embodiments of the present application. The processor 400 may include a data acquisition module 402, a feature extraction module 404, an initial blood pressure determination module 406, a predicted blood pressure determination module 408, and a communication module 410. Each module may be a hardware circuit designed to perform the following operations, a set of instructions stored in one or more storage media, and/or any combination of a hardware circuit and one or more storage media. The modules in processor 400 may be connected or in communication with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. In some embodiments, any two of the modules may be combined into a single module, and any one of the modules may be divided into two or more units.

The data acquisition module 402 can acquire data related to a first subject. The data associated with the first subject may include data associated with cardiac activity of the first subject and personal information associated with the first subject. By way of example only, the signal may be a physiological signal including, but not limited to, an Electrocardiogram (ECG) signal, a signal related to pulse waves (e.g., photoplethysmography (PPG)), an electrocardiogram (PCG) signal, an Impedance Cardiogram (ICG) signal, or the like, or any combination thereof. The personal information associated with the first subject may include the gender of the first subject, the age of the subject, the height of the first subject, the weight of the first subject, the posture of the first subject at the time the signal was obtained, whether the first subject has hypertension, whether the first subject is receiving treatment with at least one drug, information associated with the at least one drug, the name and date of birth of a family member, the phone and address of a family residence, emergency contact information, a list of current medications and dosages, a list of allergies, a list of any medical devices (e.g., cardiac pacemakers), a list of current doctors and office phones, a copy of insurance cards, DNR (no cardiopulmonary resuscitation) willingbooks and forms, an authorization book (POA) form, and the like, or combinations thereof. The posture of the first subject when obtaining the signal may include a lying posture, a sitting posture, a standing posture, and the like.

The feature extraction module 404 may extract target features associated with the first subject from the data associated with the first subject acquired by the data acquisition module 402. The target feature may refer to a feature that is relevant to a prediction of blood pressure of the first subject.

The initial blood pressure determination module 406 may determine an initial blood pressure of the first subject. In some embodiments, the initial blood pressure determination module 406 may determine the initial blood pressure of the first subject using a predictive model based on a target feature associated with the first subject. The target features may be used as inputs to a predictive model. The initial blood pressure determination module 406 may then assign the output of the predictive model as the initial blood pressure of the first subject. The initial blood pressure may include systolic and diastolic blood pressure. In some embodiments, the predictive model may be trained in advance.

The predicted blood pressure determination module 408 may determine the predicted blood pressure of the first subject using an optimization model based on the initial blood pressure of the first subject to optimize the initial blood pressure. In some embodiments, the optimization model may be trained in advance, and may be stored in the server 120 (e.g., cloud server), the terminal 140, and/or the storage device 160. The predicted blood pressure determination module 408 may retrieve the optimization model from the server 120, the terminal 140, and/or the storage device 160 accordingly. In some embodiments, the optimization model may be determined by performing one or more operations described in fig. 13.

The communication module 410 may receive a request from the terminal 140 to determine the blood pressure of the first subject. The communication module 410 may also transmit the predicted blood pressure of the first subject to the terminal 140 in response to the request. The terminal 140 may correspond to a first subject. For example, the first subject may be a user of the terminal 140.

In some embodiments, all of the modules in fig. 4 may be implemented by a single processor of one component of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). In some embodiments, one or more of the modules in fig. 4 may be implemented by different processors in one component of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). In some embodiments, one or more of the modules depicted in fig. 4 may be implemented by different components of medical system 100, such as different processors in measurement device 110, server 120, or terminal 140. For example, modules 402, 406, 408, and 410 may be implemented by a processor in server 120. As another example, modules 402 and 410 may be implemented by a first processor in terminal 140, while modules 404, 406, and 408 may be implemented by a second processor in terminal 140. As yet another example, modules 402 and 410 may be implemented by a first processor in server 120, while modules 404, 406, and 408 may be implemented by a second processor in server 120. As yet another example, modules 402 and 410 may be implemented by a processor in measurement device 110, while modules 404, 406, and 408 may be implemented by a processor in server 120. As yet another example, module 402 may be implemented by a processor in measurement device 110, modules 404, 406, and 408 may be included in a processor in server 120, and module 410 may be implemented by a processor in terminal 140.

Fig. 5 is a flow diagram of a process and/or method 500 for providing a first subject's blood pressure to a terminal in response to a request, shown in accordance with some embodiments of the present application. The process and/or method 500 may be performed by the medical system 100 (e.g., terminal 140, server 120). For example, the processes and/or methods 500 may be implemented as a set of instructions (e.g., an application program) stored in storage ROM230 or RAM 240. Processor 220 (e.g., processor 400) may execute the set of instructions and be instructed to perform process and/or method 500 accordingly. The operations of the illustrative processes/methods shown below are intended to be illustrative. In some embodiments, the processes/methods may be implemented with one or more additional operations not described and/or with one or more operations described herein. Additionally, the order of the operations of the process/method illustrated in FIG. 5 and described below is not intended to be limiting. It should also be noted that process and/or method 500 may be implemented in measurement device 110, server 120, terminal 140, and/or any combination thereof.

In 502, the processor 400 (e.g., the communication module 410) may receive a request from the terminal 140 to determine the blood pressure of the first subject. In some embodiments, the user may initiate and send a request through the terminal. In some embodiments, a request may be initiated and sent by a terminal by someone other than the user (e.g., a medical worker).

At 504, processor 400 (e.g., data acquisition module 402) may acquire data related to the first subject.

In some embodiments, the data related to the first subject may include data related to cardiac activity of the first subject. The data related to cardiac activity of the first subject may include data related to a signal indicative of cardiac activity of the first subject. By way of example only, the signal may be a physiological signal including, but not limited to, an Electrocardiogram (ECG) signal, a signal related to a pulse wave (e.g., photoplethysmography (PPG)), an electrocardiogram (PCG) signal, an impedance electrocardiogram (ICG) signal, or the like, or any combination thereof.

The measurement device 110 may be attached to a first subject and detect cardiac activity of the first subject. The measurement device 110 may generate a signal indicative of heart activity by detection. Processor 400 (e.g., data acquisition module 402) may receive data related to cardiac activity of the first subject via communication module 410, the data generated based on signals derived from measurement device 110.

The signal may remain stable for a predetermined period of time. The predetermined time period may be a default setting for the medical system 100 or may be adjusted under different conditions. The predetermined time period may be any time span (e.g., 10 seconds, 15 seconds, 20 seconds, 30 seconds, etc.). The waveform representing the signal may be displayed on a user interface of the terminal (e.g., terminal 140). The signal over the predetermined time period may include at least two beats of the heart of the first subject.

For each beat of the signal, the wave may include one or more characteristic points (e.g., peaks, valleys, etc.). The data related to cardiac activity of the first subject may include first data related to a signal, such as a time value, an amplitude value, an area value, a derivative related to each of the one or more feature points, and the like. The time value and amplitude value of the feature point may be the abscissa and ordinate of the feature point, respectively. The area value may be an integral related to the time interval. The area value may be indicative of a blood volume change in a blood vessel proximate to the attached measurement device 110. The derivative may include a first derivative, a second derivative, a third derivative, a higher derivative, the like, or combinations thereof. For at least two beats, processor 400 may determine at least two time values, at least two amplitude values, at least two area values, at least two derivatives, and so on.

In some embodiments, the data related to cardiac activity of the first subject may include second data related to the signal, the second data source being based on at least two time values, at least two amplitude values, at least two area values, and/or at least two derivatives associated with one or more feature points. For example, the second data may be a variance, a standard deviation, a quartile range, a mean, a median, and/or a weighted value of the at least two time values. As another example, the second data may be a variance, a standard deviation, a quartile range, a mean, a median, and/or a weighted value of the at least two amplitude values. As another example, the second data may be a variance, a standard deviation, a quartile range, a mean, a median, and/or a weighted value of the at least two area values. As yet another example, the second data may be a variance, a standard deviation, a quartile range, a mean, a median, and/or a weighted value of the at least two derivatives.

In some embodiments, the data relating to cardiac activity of the first subject may include third data related to the signal by converting the signal from the time domain to the frequency domain. The time-frequency transform may include, but is not limited to, a fourier transform, a wavelet transform, a laplace transform, a Z transform, and the like, or any combination thereof. The Fourier transform may include, but is not limited to, prime factor FFT algorithm, Bruun's FFT algorithm, Rader's FFT algorithm, Bluestein's FFT algorithm, and the like.

In some embodiments, data acquisition module 402 may transform the signal from the time domain to the frequency domain using a fourier transform and then determine the first representation in the frequency domain. The data acquisition module 402 may determine the third data based on the first representation. For example, the first representation may be represented by equation (1):

wherein a is_l(l ═ 0, 1, 2, … …, ∞) and b_l(l ═ 1, 2, … …, infinity) respectively andcos (l · x) and sin (l · x). The processor 400 may determine the third data based on the coefficients of equation (1). For example, the third data may include a_l(l ═ 0, 1, 2, … …, ∞) and/or b_l(l＝1、2，……，∞)。

In some embodiments, data acquisition module 402 may transform a portion of the signal from the time domain to the frequency domain using a wavelet transform and then determine a second representation in the frequency domain. The portion of the signal may include at least one beat of the signal. The data acquisition module 402 may determine third data based on the second representation. For example, the second representation may be represented by equation (2):

where X (g, h) may represent a signal portion in the frequency domain, h may represent a scaling factor, and g may represent a center position of the signal portion. The processor 400 may determine the third data based on the coefficients of equation (2). For example, the third data may be associated with m and/or n.

It should be noted that equations (1) and (2) are for illustration and the application is not intended to be limiting. The representation in the frequency domain may have other forms. Thus, the third data may be in other forms.

In some embodiments, data related to the cardiac activity of the first subject may be stored in the measurement device 110, the terminal 140, and/or the storage device 160.

In some embodiments, the data associated with the first subject may include personal information associated with the first subject. The personal information related to the first subject may include registration information/login information, such as a username and password of the first subject associated with the medical system 100. The personal information associated with the first subject may further include: the gender of the first subject, the age of the subject, the height of the first subject, the weight of the first subject, the posture of the first subject at the time the signal was obtained, whether the first subject had hypertension, whether the first subject was receiving treatment with at least one medication, information related to the at least one medication, the name and date of the family member, the address and number of the family residence, emergency contact information, a list of current medications and dosages, a list of allergies, a list of any medical devices (e.g., pacemakers), a contact by the attending physician, a copy of an insurance card, DNR (do not resuscitate) willingbooks and forms, authorization book (POA) forms, and the like, or combinations thereof. The posture of the first subject at the time the signal is obtained may include a lying posture, a sitting posture, a standing posture, and the like. The information related to the medication may include the medication taken by the first subject, the dosage of the medication taken by the first subject, whether the dosage of the medication has changed, a detailed record of the change if changed, the time the first subject took the medication, the length of time the first subject took the medication, and the like. The personal information of the first subject may be pre-stored in the server 120, the terminal 140, the external data source 130, the storage device 160, and/or combinations thereof. The data acquisition module 402 may acquire personal information from the terminal 140, the external data source 130, and/or the storage device 160 via the network 150. In some embodiments, the data acquisition module 402 may acquire personal information from the external data source 130. In some embodiments, the first subject or other subjects may enter personal information of the first subject through the terminal 140 when predicting the blood pressure of the first subject.

At 506, the processor 400 (e.g., the feature extraction module 404) may extract a target feature associated with the first subject from the data associated with the first subject. The target feature may refer to a feature that is relevant to a prediction of blood pressure of the first subject. The target feature may be an input to a prediction model for determining an initial blood pressure of the first subject.

At 508, the processor 400 (e.g., the initial blood pressure determination module 406) may determine an initial blood pressure of the first subject. In some embodiments, the initial blood pressure determination module 406 may determine the initial blood pressure of the first subject using a predictive model based on a target feature associated with the first subject. The target features may be used as inputs to a predictive model. The initial blood pressure determination module 406 may then assign the output of the predictive model as the initial blood pressure of the first subject. The initial blood pressure may include systolic and diastolic blood pressure.

In some embodiments, the predictive model may be trained in advance. Also, the trained predictive model may be stored in the server 120 (e.g., cloud server), the terminal 140, and/or the storage device 160. The initial blood pressure determination module 406 may accordingly obtain the predictive model from the server 120, the terminal 140, and/or the storage device 160. In some embodiments, the predictive model may be determined by performing one or more of the operations described in fig. 7, 9, and/or 11.

In some embodiments, the processor 400 (e.g., the predicted blood pressure determination module 408) may use the initial blood pressure as a final result of the predicted blood pressure of the first subject. Since the initial blood pressure (e.g., one or both of the systolic and diastolic pressures) is determined based on information associated with a large number of subjects, it may not be accurate for a particular subject. Therefore, the processor 400 may need to further optimize the initial blood pressure to suit a particular subject. In some embodiments, an optimization model based on the correlation between systolic and diastolic pressures may be used to optimize the initial blood pressure.

Then, at 510, the processor 400 (e.g., the predicted blood pressure determination module 408) may determine a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure of the first subject to optimize the initial blood pressure. In some embodiments, the optimization model may be trained in advance, and may be stored in the server 120 (e.g., cloud server), the terminal 140, the storage device 160, and/or combinations thereof. The predicted blood pressure determination module 408 may retrieve the optimization model from the server 120, the terminal 140, and/or the storage device 160, accordingly. In some embodiments, the optimization model may be determined by performing one or more operations described in fig. 13.

At 512, the processor 400 (e.g., the communication module 410) may transmit the predicted blood pressure of the first subject to the terminal 140 in response to the request. The terminal 140 may correspond to a first subject. For example, the first subject may be a user of the terminal 140. The predicted blood pressure may be displayed on a user interface of the terminal 140.

As another example, one or more other optional steps (e.g., storage steps, pre-processing steps) may be added elsewhere in the exemplary process/method.

In some embodiments, all of the steps described in fig. 5 may be performed by a signal processor of one component of the medical system 100 (e.g., the measurement device 110, the server 120, or the terminal 140). In some embodiments, one or more of the steps described in fig. 5 may be performed by different processors in one component of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). In some embodiments, one or more of the steps described in fig. 5 may be performed by different processors in different components of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). For example, steps 502 to 512 may be performed by a processor in server 120. As another example, steps 502-508 may be performed by a first processor in server 120, and steps 510 and 512 may be performed by a second processor in server 120. As another example, steps 502 and 504 may be performed by a first processor in terminal 140 and steps 506 through 512 may be performed by a second processor in terminal 140. As another example, steps 502 and 504 may be performed by a processor in measurement device 110, while steps 506 through 512 may be performed by a processor in server 120. As yet another example, steps 502 and 504 may be performed by a processor in terminal 140, while steps 506 through 512 may be performed by a processor in server 120.

Fig. 6A and 6B are block diagrams of another exemplary processor, shown in accordance with some embodiments of the present application. As depicted in fig. 6A, processor 600 may include a historical data acquisition module 602, an initial predictive model determination module 604, and a predictive model determination module 606. As depicted in fig. 6B, processor 600 may include a historical data acquisition module 602, an initial predictive model determination module 604. And another processor may include a predictive model determination module 606. Each module may be a hardware circuit designed to perform the operations described below, a set of instructions stored in one or more storage media, and/or any combination of a hardware circuit and one or more of the storage media described. The modules in the processor 600 may be connected or in communication with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. In some embodiments, any two of the modules may be combined into one module, and any one of the modules may be divided into two or more units.

The historical data acquisition module 602 may acquire historical data related to the at least two second subjects and at least two blood pressure measurements associated with the at least two second subjects. The historical data associated with the at least two second subjects may also be referred to herein as at least two sets of historical data. Each set of historical data may be related to a historical blood pressure measurement of the second subject. Each of the at least two historical blood pressure measurements may correspond to one of at least two sets of historical data. As indicated at 504, the term "historical data" may refer to historical data relating to cardiac activity of at least two second subjects and historical personal information relating to the at least two second subjects. In some embodiments, the term "blood pressure measurement" may refer to a historical blood pressure measurement measured by a sphygmomanometer (e.g., aneroid, mercury, automatic, electronic, etc.).

The initial prediction model determination module 604 may generate an initial prediction model based on historical data related to the at least two second subjects acquired by the historical data acquisition module 602 and at least two historical blood pressure measurements associated with the at least two second subjects. In some embodiments, the initial predictive model may include one or more sub-initial models.

The predictive model determination module 606 can generate a predictive model related to the first subject based on the initial predictive model and at least a portion of the historical data related to the first subject. At least a portion of the historical data may correspond to at least a portion of the features of the relevant historical data associated with the initial model. In some embodiments, the predictive model determination module 606 may specify one of the one or more sub-predictive models as the predictive model for the first subject based on at least a portion of the historical data associated with the first subject.

In some embodiments, all of the modules in fig. 6A and 6B may be implemented by a processor of one component of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). In some embodiments, one or more of the modules in fig. 6A and 6B may be implemented by different processors in one component of medical system 100 (e.g., measurement device 110, server 120, or terminal 140). In some embodiments, one or more of the modules described in fig. 6A and 6B may be included in different processors in different components of medical system 100, e.g., measurement device 110, server 120, or terminal 140. For example, modules 602 and 604 may be implemented by a processor in server 120. As another example, modules 602 and 604 may be implemented by a first processor in server 120 and module 606 may be implemented by a second processor in server 120. As yet another example, modules 602 and 604 may be implemented by a processor in server 120, and module 606 may be implemented by a processor in terminal 140.

Fig. 7 is a flow diagram of a process and/or method for generating a prediction model for predicting an initial blood pressure of a first subject, according to some embodiments of the present application. The process and/or method 700 may be performed by the medical system 100. For example, the process and/or method 700 may be implemented as a set of instructions (e.g., an application program) stored in the storage ROM230 or RAM 240. Processor 220 (e.g., processor 600) may execute the set of instructions and may be instructed accordingly to perform process and/or method 700. The operations of the illustrative processes/methods shown below are intended to be illustrative. In some embodiments, the processes/methods may, when implemented, add one or more additional operations not described, and/or subtract one or more operations described herein. Additionally, the order of the operations of the processes/methods illustrated in FIG. 7 and described below is not intended to be limiting. It should also be noted that process and/or method 700 may be implemented in measurement device 110, server 120, and/or terminal 140.

In 702, the processor 600 (e.g., the historical data acquisition module 602) may acquire historical data related to the at least two second subjects and at least two blood pressure measurements associated with the at least two second subjects. The historical data associated with the at least two second subjects may also be referred to herein as at least two sets of historical data. Each set of historical data may be correlated with historical blood pressure measurements of the second subject. Each of the at least two historical blood pressure measurements may correspond to one of at least two sets of historical data. As described at 504, the term "historical data" may refer to historical data relating to cardiac activity of at least two second subjects and historical personal information relating to the at least two second subjects. In some embodiments, the term "blood pressure measurement" may refer to a historical blood pressure measurement measured by a sphygmomanometer (e.g., aneroid, mercury, automatic, electronic, etc.). For each second subject, one or more historical blood pressure measurements may be collected. When measuring blood pressure, at least two historical blood pressure measurements associated with each second subject may be collected for one or more poses of each second subject. The postures may include lying, sitting, standing, etc. In some embodiments, at least two historical blood pressure measurements associated with each second subject may be collected for a time of day. For example, the at least two historical blood pressure measurements may be collected at 9 am of each day for three consecutive months. In some embodiments, at least two historical blood pressure measurements associated with each second subject may be collected for age. For example, at least two historical blood pressure measurements may be collected continuously during an annual physical examination between the ages of 40 and 45. The historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects can be at any time in the past (e.g., months ago, days ago, hours ago, minutes ago, etc.). The medical system 100 may save historical data related to the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects as historical data to the storage device 160, the external data source 130, or the server 120. In some embodiments, the at least two second subjects may include a first subject.

Processor 600 (e.g., historical data acquisition module 602) may determine historical data related to cardiac activity of at least two second subjects based on at least two historical signals indicative of cardiac activity of the at least two second subjects. The measurement device 110 can detect a physiological phenomenon (e.g., cardiac activity) and generate at least two historical signals at any time in the past (e.g., months ago, days ago, hours ago, minutes ago, etc.). As depicted at 504, the historical data associated with cardiac activity may include historical first data, historical second data, and historical third data. For each beat in each of the at least two historical signals, the wave may include one or more feature points (e.g., a peak of the wave, a trough of the wave, etc.). The historical first data may include time values, amplitude values, area values, derivatives, etc. associated with each of the one or more feature points of each of the at least two historical signals. The time value and amplitude value of the feature point may be the abscissa and ordinate of the feature point, respectively. The area value may be an integral related to the time interval. The area value may refer to the change in blood volume in a blood vessel near the attached measurement device 110. The derivative may include a first derivative, a second derivative, a third derivative, a higher derivative, or the like, or combinations thereof. For at least two beats, processor 400 may determine at least two time values, at least two amplitude values, at least two area values, at least two derivatives, and so on. The historical second data may be a plurality of time values, variances, standard deviations, ranges of quartiles, averages, medians, and/or weighted values, and the like, of amplitude values and/or area values. The historical third data may be associated with coefficients representing at least two historical signals in the frequency domain.

Processor 600 (e.g., historical data acquisition module 602) may determine historical personal information related to at least two second subjects. As indicated at 504, the historical personal information associated with the at least two second subjects can include the gender of the second subject, the age of the second subject, the height of the second subject, the weight of the second subject, the posture of the second subject at the time the signal was obtained, whether the second subject has hypertension, whether the second subject is receiving at least one medication, information related to at least one medication, the name and date of birth of the family member, the family address and phone number of the family member, emergency contact information, current medication and dosage lists, allergy lists, lists of any medical devices (e.g., pacemakers), contact information of primary care physicians, insurance card copies, willingness and forms for DNR (do not cardiopulmonary resuscitation), authority (POA) forms, and the like, or combinations thereof. The information related to the medication therapy of each of the at least two second subjects may include the medication taken by the second subject, the dose of the medication taken by the second subject, whether the dose of the medication changed, if the detailed record of the change changed, the time the second subject took the medication, the length of time the second subject took the medication, and the like, or any combination thereof. In some embodiments, processor 600 (e.g., historical data acquisition module 602) may obtain historical personal information from server 120 or storage device 160 via network 150.

In some embodiments, processor 600 (e.g., historical data acquisition module 602) may combine historical data related to at least two second subjects (also referred to herein as at least two historical terms) and at least two historical blood pressure measurements associated with the at least two second subjects to generate sample data for training. The combination of one of the at least two historical blood pressure measurements and a set of historical data corresponding to the historical blood pressure measurement may be referred to as a historical sample data item. The sample data may correspond to a first predetermined number of measurements for each second subject. The first predetermined number of measurements may include a first one measurement, a first three measurements, a first five measurements, a first ten measurements, and so on. In some embodiments, the processor 600 may similarly determine calibration sample data. The calibration sample data may include a historical blood pressure measurement and a set of historical data corresponding to the historical blood pressure measurement. The calibration sample data may correspond to a second predetermined number of measurements for each second subject. The second predetermined number of measurements may include a second measurement, a second three measurements, a second five measurements, a twentieth measurement, and so on. In some embodiments, the sample data may comprise calibration sample data. In some embodiments, the sample data and calibration sample data may or may not overlap. The sample data and calibration sample data may each comprise at least two sets of data.

In some embodiments, the processor 600 (e.g., the historical data acquisition module 602) may filter the sample data, for example, by determining whether the value of each of at least two sample data items is anomalous. When it is determined that a value of a sample data item is abnormal, the sample data item associated with the abnormal value may be ignored. For example, when a sample data item's historical blood pressure measurement is identified as negative, the processor 600 may ignore the sample data item associated with an abnormal historical blood pressure measurement. In some embodiments, the processor 600 may obtain time information relating to each of at least two sample data items. The time information relating to each of the at least two sample data items may refer to the time at which the sample data item was acquired by the measurement device 110, for example at 8 am of the day, at 2 pm of the day. In some embodiments, processor 600 may divide a day into a predetermined number of groups, e.g., 4 groups, 6 groups. Each group may correspond to a time period of the same length. For example, when a day is divided into 6 groups, each of the six groups may have 4 hours. Each group may be associated with a time tag. Processor 600 may then classify the time tag.

In 704, the processor 600 (e.g., the initial prediction model determination module 604) may generate an initial prediction model based on the historical data related to the at least two subjects acquired by the historical data acquisition module 602 and the at least two historical blood pressure measurements associated with the at least two second subjects. In some embodiments, the initial predictive model may include one or more sub-initial models.

In some embodiments, processor 600 may extract features from sets of historical data. When training the initial prediction model, the processor 600 (e.g., the initial prediction model determination module 604) may first normalize the sample data using a normalization technique (e.g., min-max normalization, z-score normalization, decimal scaling normalization). The processor 600 (e.g., the initial prediction model determination module 604) may then extract features from the normalized sample data, which may include at least two features related to time values, amplitude values, area values of feature points of the historical signal, second data related to the historical signal, third data related to representations in the frequency domain, gender, age, height, weight, posture of each of the at least two second subjects, time of measurement, historical blood pressure measurements, whether each of the at least two second subjects has hypertension, whether each of the at least two subjects is receiving treatment with at least one drug, etc A detailed description of the initial prediction model can be found in fig. 9.

In 706, the predictive model determination module 606 can generate a predictive model related to the first subject based on the initial predictive model and at least a portion of the historical data related to the first subject. At least a portion of the historical data may correspond to at least a portion of the features associated with the historical data associated with the first subject. In some embodiments, processor 600 (e.g., predictive model determination module 606) may specify one sub-predictive model from among one or more sub-predictive models of the initial predictive model as the predictive model related to the first subject based on at least a portion of the historical data associated with the first subject. A detailed description of generating the initial prediction model can be seen in fig. 11.

In some embodiments, the predictive model determination module 606 may be implemented by the processor 600 as shown in fig. 6A. For example, processor 600 is a processor in measurement device 110, server 120, or terminal 140, and processor 600 may perform all operations as described in steps 702, 704, and 706 of process/method 700. For example, server 120 may obtain historical data related to at least two second subjects and at least two corresponding blood pressure measurements associated with the at least two second subjects and generate an initial predictive model. Because the at least two second subjects include the first subject, the server 120 may directly generate a predictive model related to the first subject based on the initial predictive model and historical data related to the first subject. In some embodiments, the predictive model determination module 606 may be implemented by a processor in the measurement device 110 or the terminal 140 instead of the processor 600. For example, the processor 600 is a processor in the server 120, but the predictive model determination module 606 may be implemented by a processor of the measurement device 110 or the terminal 140. Accordingly, the operations described in steps 702 and 704 may be performed in the server 120, and the operations described in step 706 may be performed in the measurement device 110 or the terminal 140. For example, server 120 may first obtain historical data relating to at least two second subjects and at least two corresponding blood pressure measurements associated with the at least two second subjects and generate an initial predictive model. Server 120 may then transmit the initial predictive model to measurement device 110 or terminal 140, and measurement device 110 or terminal 140 may then generate a first subject-related predictive model based on the initial predictive model and historical data associated with the first subject. The generated predictive model may be stored in the measurement device 110, the terminal 140, or a combination thereof.

FIG. 8 is a block diagram of an exemplary initial predictive model determination module shown in accordance with some embodiments of the present application. The initial prediction model determination module 800 may include a first group feature extraction unit 802, a second group feature extraction unit 804, a clustering unit 806, a historical target feature determination unit 808, a sub-prediction model determination unit 810, and an initial prediction model determination unit 812. Each unit may be a hardware circuit designed to perform the following operations, a set of instructions stored in one or more storage media, and/or any combination of a hardware circuit and one or more storage media. The units in the initial predictive model determination module 800 may be connected or in communication with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. In some embodiments, any two units may be combined into a single unit, and any one unit may be divided into two or more sub-units.

The first set of feature extraction unit 802 may extract a first set of features from at least two sets of historical data associated with at least two second subjects. The first set of features may be related to data such as time values, amplitude values, area values of feature points of the history signal, second data related to the history signal, third data related to a representation of the frequency domain. The gender, age, height, weight, posture, time of measurement, historical blood pressure measurements, whether each of the at least two second subjects has hypertension, whether each of the at least two second subjects is receiving at least one medication, etc., of each of the at least two second subjects. The dimension of the first set of features may be denoted as n 1. In some embodiments, the initial predictive model determination module 800 may also normalize the first set of features.

The second set of feature extraction unit 804 may determine the second set of features based on the normalized first set of features. The dimensions of the second set of features may be smaller than the dimensions of the normalized first set of features. The dimension of the second set of features may be denoted as n 2. n2 is less than n 1.

In some embodiments, the second set of feature extraction unit 804 may determine the second set of features using Principal Component Analysis (PCA) based on the normalized first set of features. One of the second set of features may be one of the normalized first set of features or a linear combination of the features.

Clustering unit 806 may cluster at least two sets of historical data (e.g., filtered at least two sets of historical data) associated with at least two second subjects into one or more clusters. In some embodiments, clustering unit 806 may cluster the historical data into one or more clusters using a clustering algorithm based on the second set of features. The clustering algorithms may include connectivity-based clustering (e.g., mean link clustering), centroid-based clustering (e.g., k-medoid clustering), distribution-based clustering (e.g., Gaussian mixture model), density-based clustering (e.g., space density based clustering with noise (DBSCAN), etc., or any combination thereof.

The historical target feature determination unit 808 may determine the historical target features based on a second set of features of the training sample data set. In some embodiments, a portion of the second set of features may be less relevant to the prediction of blood pressure. Thus, the processor 220 may select only features that are more relevant to the prediction of blood pressure. In some embodiments, the historical target feature determination unit 808 may determine the historical target feature from the second set of features using a feature selection technique and information criteria. The feature selection techniques may include stepwise regression, penalty methods, and the like. Stepwise regression may include, but is not limited to, forward selection, backward elimination, bi-directional elimination, and the like or combinations thereof. The penalty methods may include lasso algorithms, methods associated with lasso algorithms, smoothed shear absolute deviation (SCAD), and methods associated with SCAD, among others. The information criteria may include bayesian information criteria, Akaike information criteria, deviation information criteria, Hannan-Quinn information criteria, and the like, or combinations thereof. The historical target features may include fixed impact features and random impact features. The fixed influence feature may comprise a feature having a linear correlation with the historical blood pressure measurements, e.g. a feature relating to time values, amplitude values, area values, medication of feature points. The random impact feature may include a feature that does not have a linear correlation with the historical blood pressure measurements, but the random impact feature may have a random impact on the historical blood pressure measurements, such as a serial number of the feature associated with the feature point or medication.

The sub-prediction model determination unit 810 may determine a sub-prediction model for each of the one or more clusters based on historical target features of the historical data in each cluster and historical blood pressure measurements corresponding to the historical data in each cluster. The sub-prediction model for each cluster may be the sub-prediction model corresponding to the cluster determined by clustering unit 806. For example, if the clustering unit 806 generates k clusters, the sub prediction model determination unit 810 may generate k sub prediction models.

The initial prediction model determination unit 812 may designate one or more sub-prediction models corresponding to one or more clusters as an initial prediction model. The initial predictive model may include all of the one or more sub-predictive models corresponding to all of the one or more clusters. The initial predictive model may be adapted to different subjects.

FIG. 9 is a flow diagram illustrating a process and/or method for generating an initial predictive model according to some embodiments of the present application. The process and/or method 900 may be performed by the medical system 100. For example, the processes and/or methods 900 may be implemented as a set of instructions (e.g., an application program) stored in the storage ROM230 or RAM 240. Processor 220 may execute the set of instructions described above and may be instructed accordingly to perform process and/or method 900. The operations of the illustrative processes/methods shown below are intended to be illustrative. In some embodiments, the processes/methods may, when implemented, add one or more additional operations not described, and/or subtract one or more operations described herein. Additionally, the order of the operations of the processes/methods illustrated in FIG. 9 and described below is not intended to be limiting. It should also be noted that process and/or method 900 can be implemented in measurement device 110, server 120, and/or terminal 140.

At 902, the processor 220 (e.g., the first set of feature extraction unit 802 of the initial predictive model determination module 800) may extract a first set of features from at least two sets of historical data associated with at least two second subjects. The first set of features may be related to data such as time values, amplitude values, area values of feature points of the history signal, second data related to the history signal, third data related to a representation of the frequency domain. The gender, age, height, weight, posture, time of measurement, historical blood pressure measurements, whether each of the at least two second subjects has hypertension, whether each of the at least two second subjects is receiving at least one medication, etc., of each of the at least two second subjects. The dimension of the first set of features may be denoted as n 1.

In some embodiments, processor 220 may first filter at least two sets of historical data associated with at least two second subjects before processor 220 performs step 902. For example, the processor 220 may remove historical data for blood pressure value anomalies and/or historical data for no blood pressure measurements. Then, the first group feature extraction unit 802 may extract a first group feature from the filtered at least two groups of history data.

In some embodiments, the processor 220 may normalize the historical data before the first set of features is extracted by the first set of feature extraction unit 802. The normalization of the historical data may include feature scaling and zero mean normalization. Feature scaling may cause all historical data to be at the same scale. Zero-mean normalization can bring the average of each historical data to zero and the standard deviation of each historical data to one.

At 904, the processor 220 (e.g., the second set of features determination unit 804 of the initial predictive model determination module 800) may determine a second set of features based on the normalized first set of features. The dimensions of the second set of features may be smaller than the dimensions of the normalized first set of features. The dimension of the second set of features may be denoted as n 2. n2 is less than n 1.

In some embodiments, the processor 220 may determine the second set of features using Principal Component Analysis (PCA) based on the normalized first set of features. One of the second set of features may be a linear combination of one or more of the normalized first set of features. For example, a second feature Z1 may be expressed as equation (3):

Z1＝a1*X1+a2*X2+…+ai*Xi， (3)

where the ith of the first set of features normalized can be designated by Xi^thThe coefficients ai may refer to the ith of the first set of features normalized to^thEach feature corresponds to a coefficient, and i may refer to an integer greater than 0 and less than or equal to n 1.

In some embodiments, after applying principal component analysis to the normalized first set of features, the dimensions of the second set of features may be determined based on the cumulative variance contribution rate. The cumulative variance contribution rate may be a default value in the system or may be set by the system operator. In some embodiments, the cumulative variance contribution rate may be a value from 85% to 95%. In some embodiments, the cumulative variance contribution rate may be a constant. In some embodiments, the cumulative variance contribution rate may be varied based on a prediction associated with all or part of the historical data. For example, the cumulative variance contribution rate may be set to the first number and all or a portion of the historical data may be predicted to generate a prediction. The processor 220 may then compare the prediction to historical blood pressure measurements corresponding to all or part of the historical data. When the comparison result does not satisfy the condition, the processor 220 may change the value of the cumulative variance contribution rate until the comparison result satisfies the condition.

In 906, the processor 220 (e.g., the clustering unit 806 of the initial prediction model determination module 800) may cluster at least two sets of historical data (e.g., at least two sets of filtered historical data) related to at least two second subjects into one or more clusters. In some embodiments, clustering unit 806 may cluster the historical data into one or more clusters using a clustering algorithm based on the second set of features. The clustering algorithm may include connectivity-based clustering (e.g., mean link clustering), centroid-based clustering (e.g., k-medoid clustering), distribution-based clustering (e.g., Gaussian mixture model), density-based clustering (e.g., noise-based spatial density clustering (DBSCAN), etc., or any combination thereof.

In some embodiments, prior to clustering, the processor 220 (initial predictive model determination module 800) may randomly divide the sample data into a training sample data set and a test sample data set according to at least two predetermined times (e.g., 10 times, 50 times, 100 times, 150 times, etc.). Then, for each of the at least two predetermined times, the initial prediction model determination module 800 may cluster only the training sample data set to generate one or more clusters. The processor 220 may designate a cluster from one or more clusters for each of the test sample data sets.

In some embodiments, the training sample data may correspond to a portion of at least two sets of historical data associated with at least two second subjects. The test sample data set may correspond to the remainder of at least two sets of historical data associated with at least two second subjects. The training sample data set may correspond to 50% -90% (e.g., 90%, 85%, 80%, 75%, 70%, 65%, 60%, 55%, 50%) of the data of at least two sets of historical data. In some embodiments, the training sample data set and the test sample data set may correspond to a first percentage of the at least two second subjects and a second percentage of the at least two second subjects, respectively. For example, the training sample data may correspond to 10%, 20%, 30% or 40% of the at least two second subjects. The test sample data set may correspond to 90%, 80%, 70% or 60% of the at least two second subjects, respectively. The sum of the first percentage and the second percentage may be equal to or less than 1. It should be understood that the values of the first and second percentages mentioned above are for illustrative purposes and are not limiting. The first percentage and the second percentage may be any value between 0 and 1. It should be noted that each set of sample data is considered a whole during training and/or testing and is therefore inseparable.

In some embodiments, the process of dividing at least two sets of historical data into a training sample data set and a test sample data set may be processed prior to the PCA process. For example, the initial predictive model determination module 800 may divide the sample data into a training sample data set and a test sample data set prior to extracting the first set of features from the training sample data. The initial predictive model determination module 800 may then normalize the training sample data and generate one or more normalized parameters. The initial predictive model determination module 800 may then normalize the test sample data set using one or more normalization parameters. For the normalized training sample data set and the normalized test sample data set, the initial prediction model determination module 800 may extract the first set of features and perform a PCA process.

In 908, the processor 220 (e.g., the historical target feature determination unit 808 of the initial predictive model determination module 800) may determine historical target features based on a second set of features of the training sample data Akaike information criterion, deviation information criterion, Hannan-Quinn information criterion, the like, or combinations thereof. The historical target features may include fixed impact features and random impact features. The fixed influence feature may comprise a feature having a linear correlation with the historical blood pressure measurements, e.g. a feature relating to time values, amplitude values, area values, medication of feature points. The random impact feature may include a feature that does not have a linear correlation with the historical blood pressure measurements, but the random impact feature may have a random impact on the historical blood pressure measurements, such as a serial number of a feature related to a feature point or medication.

At 910, the processor 220 (e.g., the sub-prediction model determination unit 810 of the initial prediction model determination module 800) may determine a sub-prediction model for each of the one or more clusters based on the historical target features of the historical data in each cluster and the historical blood pressure measurements corresponding to the historical data in each cluster. The sub-prediction model for each cluster may be the sub-prediction model corresponding to the cluster generated in 906. For example, if k clusters are generated at 906, k sub-prediction models may be generated at 910. In some embodiments, the sub-predictive models may include generalized linear models.

In some embodiments, the generalized linear model may be represented by equation (4):

P＝q₁·f(Z1)+q₂·f(Z2)+…+q_n·f(Zn)， (4)

where P may refer to blood pressure, Zi may refer to historical target features, 1 ≦ i ≦ n, n may refer to the total number of historical target features, f (Zi) may refer to a representation of historical target features Zi, and q (Zi) may refer to a representation of historical target features Zi_iMay refer to the coefficients corresponding to the historical target features Zi. Each of the one or more sub-prediction models may be expressed in a form similar to equation (4).

In 912, the processor 220 (e.g., the initial prediction model determination unit 812 of the initial prediction model determination module 800) may designate one or more sub-prediction models corresponding to one or more clusters as the initial prediction model. The initial prediction model may include all sub-prediction models of the one or more sub-prediction models corresponding to all of the one or more clusters. The initial predictive model may be adapted to different subjects.

In some embodiments, the processor 220 may use the test sample data set to test the predictive performance of the trained initial predictive model. If the processor 220 determines that the prediction performance of the trained initial prediction model meets the predetermined criteria, the processor 220 may predict the blood pressure using the initial prediction model. If the processor 220 determines that the initial predictive model does not meet the predetermined criteria, the processor 220 (e.g., the initial predictive model determination module 800) may update the initial predictive model based on the new historical data and the blood pressure measurements corresponding to the new historical data.

FIG. 10 is a block diagram of an exemplary predictive model determination module shown in accordance with some embodiments of the present application. The prediction model determination module 1000 may include a target feature extraction unit 1002, a target cluster determination unit 1004, and a prediction model determination unit 1006. Each unit may be designed as a hardware circuit that performs the following operations, a set of instructions stored in one or more storage media, and/or any combination of a hardware circuit and one or more storage media. The units in the predictive model determination module 1000 may be connected to or in communication with each other via a wired or wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. In some embodiments, any two units may be combined into a single unit, and any one unit may be divided into two or more sub-units.

The target feature extraction unit 1002 may extract historical target features from historical data associated with the first subject. In some embodiments, the at least two second subjects may include a first subject, and the historical data associated with the at least two second subjects may include historical data associated with the first subject. The target feature extraction unit 1002 may then extract historical target features associated with the first subject directly from the historical target features determined in 908. In some embodiments, the at least two second subjects may not include the first subject, and processor 220 may first obtain at least one set of historical data and corresponding blood pressure measurements associated with the first subject. Then, the target feature extraction unit 1002 may extract the target feature from at least one set of historical data associated with the first subject.

The target cluster determination unit 1004 may determine a target cluster from the one or more clusters based on historical target features of historical data associated with the first subject. In some embodiments, the target cluster determination unit 1004 may compare historical target features of the historical data related to the first subject to the historical target features in each cluster to determine one of the one or more clusters as the target cluster corresponding to the first subject.

The prediction model determination unit 1006 may designate a sub-prediction model corresponding to the target cluster as the prediction model related to the first subject.

Fig. 11 is a flow diagram illustrating a process and/or method for determining a first subject-related predictive model, shown in accordance with some embodiments of the present application. Process and/or method 1100 may be performed by medical system 100. For example, the process and/or method 1100 may be implemented as a set of instructions (e.g., an application program) stored in the storage ROM230 or RAM 240. Processor 220 may execute the set of instructions described above and may be instructed accordingly to perform process and/or method 1100. The operations of the illustrative processes/methods shown below are intended to be illustrative. In some embodiments, the processes/methods may, when implemented, add one or more additional operations not described, and/or subtract one or more operations described herein. Additionally, the order of the operations of the process/method illustrated in FIG. 11 and described below is not intended to be limiting. It should also be noted that process and/or method 1100 can be implemented in measurement device 110, server 120, and/or terminal 140.

At 1102, the processor 220 (e.g., the target feature extraction unit 1002 of the predictive model determination module 1000) may extract historical target features from historical data associated with the first subject. And the historical data associated with the at least two second subjects may include historical data associated with the first subject, then, the target feature extraction unit 1002 may extract historical target features associated with the first subject directly from the historical target features determined in 908, hi some embodiments, the at least two second subjects may not include the first subject, and processor 220 may first obtain at least one set of historical data and corresponding blood pressure measurements associated with the first subject, then, the target feature extraction unit 1002 may extract a target feature from at least one set of historical data associated with the first subject.

In 1104, the processor 220 (e.g., the target cluster determination unit 1004 of the predictive model determination module 1000) may determine a target cluster from the one or more clusters based on historical target features of historical data associated with the first subject. In some embodiments, the processor 220 (e.g., the target cluster determination unit 1004 of the predictive model determination module 1000) may compare historical target features of the historical data related to the first subject to the historical target features in each cluster to determine one of the one or more clusters as the target cluster corresponding to the first subject.

In 1106, the processor 220 (e.g., the predictive model determination unit 1006 of the predictive model determination module 1000) may designate a sub-predictive model corresponding to the target cluster as the predictive model for the first subject.

In some embodiments, the predictive model determination module 1000 may update the predictive model based on the updated historical data associated with the first subject and the corresponding blood pressure measurements associated with the first subject. For example, the first subject may have a disease recently and may be receiving treatment with at least one drug that may have an effect on blood pressure. If processor 220 still uses historical data associated with the first subject to determine a predictive model related to the first subject, the predictive model may not produce an accurate prediction result that is adapted to the recent physical condition of the first subject. Accordingly, processor 220 may need to update the predictive model associated with the first subject using the recently obtained updated historical data associated with the first subject. For example, processor 220 (e.g., predictive model determination module 1000) may extract new target features from the updated historical data associated with the first subject and determine updated target clusters from the one or more clusters based on the updated target features associated with the first subject and assign sub-predictive models corresponding to the updated target clusters as updated predictive models associated with the first subject. The updated predictive model may be more adaptive to the recent physical condition of the first subject. Thus, the updated predictive model may more accurately predict the blood pressure of the first subject.

Fig. 12 is a block diagram of an exemplary predicted blood pressure determination module shown in accordance with some embodiments of the present application. The predicted blood pressure determination module 1200 may include a model initialization unit 1202, a first blood pressure determination unit 1204, a second blood pressure determination unit 1206, a condition determination unit 1208, an optimized model determination unit 1210, a predicted blood pressure determination unit 1212, and a model update unit 1214. Each unit may be a hardware circuit designed to perform the following operations, a set of instructions stored in one or more storage media, and/or any combination of a hardware circuit and one or more storage media. The units in the predicted blood pressure determination module 1200 may be connected or in communication with each other via a wired connection or a wireless connection. The wired connection may include a metal cable, an optical cable, a hybrid cable, and the like, or any combination thereof. The wireless connection may include a Local Area Network (LAN), a Wide Area Network (WAN), bluetooth, zigbee network, Near Field Communication (NFC), etc., or any combination thereof. In some embodiments, any two units may be combined into a single unit, and any one unit may be divided into two or more sub-units.

The model initialization unit 1202 may initialize the optimization model by initializing parameters related to the optimization model. In some embodiments, the initial values of one or more parameters related to the optimization model may be random values. In some embodiments, an initial value for one or more parameters associated with the optimization model may be determined using a predictive model associated with the first subject based on sample data associated with the first subject and an initial blood pressure of the first subject.

The first blood pressure determination unit 1204 may designate the initial blood pressure as the first blood pressure. The first blood pressure may include a first systolic pressure and a first diastolic pressure. The first blood pressure determination unit 1204 may designate the initial systolic pressure and the initial diastolic pressure as a first systolic pressure and a first diastolic pressure, respectively.

In some embodiments, for one iteration, the first blood pressure determination unit 1204 may also designate a blood pressure before the iteration as the first blood pressure.

The second blood pressure determination unit 1206 may generate a second blood pressure based on the first blood pressure using the first optimization model. The second blood pressure may include a second systolic pressure and a second diastolic pressure.

The condition determining unit 1208 may determine whether a convergence condition is satisfied. The convergence condition may be associated with the first blood pressure and the second blood pressure. In some embodiments, the condition determining unit 1208 may determine a standard deviation of the first blood pressure and the second blood pressure and determine whether the standard deviation is less than a predetermined threshold. In response to determining that the standard deviation is less than the predetermined threshold, the condition determining unit 1208 may determine that the convergence condition is satisfied. In response to determining that the standard deviation is greater than or equal to the predetermined threshold, the condition determination unit 1208 may determine that the convergence condition is not satisfied.

The optimization model determination unit 1210 may determine an optimization model based on the iteration result.

The predicted blood pressure determination unit 1212 may determine a predicted blood pressure of the first subject. For example, when the convergence condition is satisfied, the predicted blood pressure determination unit 1212 may designate the second blood pressure generated in the iteration as the predicted blood pressure.

The model updating unit 1214 may update the optimization model. For example, the model update unit 1214 may adjust one or more parameter values in the optimization model. Then, the model updating unit 1214 may generate an updated optimization model.

Fig. 13 is a flow diagram of a process and/or method for determining a predicted blood pressure of a first subject using an optimization model, according to some embodiments of the present application. The process and/or method 1300 may be performed by the medical system 100. For example, the process and/or method 1300 may be implemented as a set of instructions (e.g., an application program) stored in the storage ROM230 or RAM 240. Processor 220 may execute the set of instructions described above and may be instructed accordingly to perform process and/or method 1300. The operations of the illustrative processes/methods shown below are intended to be illustrative. In some embodiments, the processes/methods may, when implemented, add one or more additional operations not described, and/or subtract one or more operations described herein. Additionally, the order of the operations of the process/method illustrated in FIG. 13 and described below is not intended to be limiting. It should also be noted that process and/or method 1300 may be implemented in measurement device 110, server 120, and/or terminal 140.

Since the initial blood pressure (e.g., one or both of the systolic and diastolic pressures) is determined based on information associated with a large number of subjects, it may not be accurate for a particular subject. Thus, the processor 220 may need to further optimize the initial blood pressure to suit a particular subject. In some embodiments, an optimization model based on the correlation between blood pressure and diastolic pressure may be used to optimize the initial blood pressure. To determine the predicted blood pressure, processor 220 may perform one or more iterations to calculate the blood pressure until a convergence condition is satisfied. The first blood pressure may include a first systolic pressure and a first diastolic pressure. The second blood pressure may include a second systolic pressure and a second diastolic pressure.

In 1302, the processor 220 (e.g., the model initialization unit 1202 of the predicted blood pressure determination module 1200) may initialize an optimization model.

In some embodiments, the optimization model may be represented by equations (5) and (6):

Sbp2＝α1*Dbp1+β1*Y+1， (5)

Dbp2＝α2*Dbp1+β2*Y+2， (6)

where Sbp2 may refer to post-iteration systolic blood pressure, Dbp2 may refer to post-iteration diastolic blood pressure, Dbp1 may refer to pre-iteration diastolic blood pressure, Y may refer to a target feature associated with the first subject, α 1 and α 2 may refer to coefficients corresponding to diastolic blood pressure Dbp1, β 1 and β 2 may refer to parameters corresponding to target feature Y, and 1 and 2 may refer to error values. The model initialization unit 1202 may determine initial values of parameters (e.g., α 1, α 2, β 1, β 2, 1, and 2) related to the optimization model. When determining α 1, α 2, β 1, β 2, 1, and 2, a first optimization model may be initialized.

In some embodiments, the model initialization unit 1202 of the predicted blood pressure determination module 1200 may initialize β 1, β 2, 1, 2 based on sample data associated with the first subject.

In some embodiments, the model initialization unit 1202 of the predicted blood pressure determination module 1200 may initialize β 1, β 2, 1, and 2 using equations (7) and (8) below based on the sample data related to the first subject determined above and the initial blood pressure of the first subject.

Sbp1＝β1*Y+1， (7)

Dbp1＝β2*Y+2， (8)

Wherein Sbp1 may refer to an initial systolic blood pressure determined using a predictive model, Dbp1 may refer to an initial diastolic blood pressure determined using a predictive model, Y may refer to a target feature associated with the first subject, β 1 may refer to a coefficient of the target feature Y corresponding to the initial systolic blood pressure, β 2 may refer to a coefficient of the target feature Y corresponding to the initial diastolic blood pressure, 1 may refer to an error value corresponding to the initial systolic blood pressure, and 2 may refer to an error value corresponding to the initial diastolic blood pressure. Based on target features extracted from sample data related to the first subject and an initial blood pressure (including an initial systolic pressure and an initial diastolic pressure), the model initialization unit 1202 may determine values of parameters including β 1, β 2, 1, and 2. The determined values of the parameters including β 1, β 2, 1, and 2 may be initial values of the parameters including β 1, β 2, 1, and 2.

In some embodiments, model initialization unit 1202 may assign random values to α 1 and α 2.

For illustration only, the initialized optimization model may also be referred to as a first optimization model.

At 1304, the processor 220 (e.g., the first blood pressure determination unit 1204 of the predicted blood pressure determination module 1200) may designate the initial blood pressure as the first blood pressure. The first blood pressure may include a first systolic pressure and a first diastolic pressure. The first blood pressure determination unit 1204 may designate the initial systolic pressure and the initial diastolic pressure as a first systolic pressure and a first diastolic pressure, respectively.

In 1306, the processor 220 (e.g., the second blood pressure determination unit 1206 of the predicted blood pressure determination module 1200) may generate a second blood pressure based on the first blood pressure using the first optimization model described in equations (5) and (6). Sbp2 and Dbp2 may refer to the second systolic and diastolic pressures, respectively, and Dbp1 may refer to the first diastolic pressure.

In 1308, the processor 220 (e.g., the condition determination unit 1208 of the predicted blood pressure determination module 1200) may determine whether a convergence condition is satisfied. The convergence condition may be associated with the first blood pressure and the second blood pressure. In some embodiments, the condition determining unit 1208 may determine a standard deviation of the first blood pressure and the second blood pressure and determine whether the standard deviation is less than a predetermined threshold. The predetermined threshold may be 0.001 to 0.1, e.g. 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09. In response to determining that the standard deviation is less than the predetermined threshold, the condition determination unit 1208 may determine that the convergence condition is satisfied, and the process 1300 may then proceed to 1310. In response to determining that the standard deviation is greater than or equal to the predetermined threshold, the condition determining unit 1208 may determine that the convergence condition is not satisfied, and the process 1300 may then proceed to 1314.

In some embodiments, the convergence condition may include a standard deviation of Sbp1 and Sbp2 (also referred to as systolic blood pressure standard deviation), and/or a standard deviation of Dbp1 and Dbp2 (also referred to as diastolic blood pressure standard deviation). In some embodiments, the condition determining unit 1208 may compare with a predetermined threshold using only one of the systolic standard deviation and the diastolic standard deviation to determine whether the convergence condition is satisfied. For example, the condition determining unit 1208 may determine that the convergence condition is satisfied if the systolic pressure standard deviation or the diastolic pressure standard deviation is smaller than a predetermined threshold.

In some embodiments, the condition determining unit 1208 may compare the systolic pressure standard deviation and the diastolic pressure standard deviation with a predetermined threshold value at the same time to determine whether the convergence condition is satisfied. For example, the condition determining unit 1208 may determine that the convergence condition is satisfied only when both the systolic pressure standard deviation and the diastolic pressure standard deviation are smaller than a predetermined threshold.

In some embodiments, the condition determining unit 1208 may compare the systolic blood pressure standard deviation with a first predetermined threshold and the diastolic blood pressure standard deviation with a second predetermined threshold to determine whether a convergence condition is satisfied. The condition determining unit 1208 may determine that the convergence condition is satisfied when the systolic standard deviation is smaller than a first predetermined threshold and the diastolic standard deviation is smaller than a second predetermined threshold. The first predetermined threshold may be different from the second predetermined threshold. The first predetermined threshold and/or the second predetermined threshold may be 0.001 to 0.1, e.g. 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09.

In 1310, the processor 220 (e.g., the optimization model determination unit 1210 of the predictive blood pressure determination module 1200) may designate the first optimization model as the optimization model. Processor 220 may generate a predicted blood pressure related to the first subject using the optimization model.

In 1312, the processor 220 (e.g., the predicted blood pressure determination unit 1212 of the predicted blood pressure determination module 1200) may designate the second blood pressure generated in the iteration as the predicted blood pressure. The predicted blood pressure may be displayed on a user interface of the terminal 140 associated with the first subject.

In 1314, the processor 220 (e.g., the model update unit 1214 of the predicted blood pressure determination module 1200) may update the first optimization model. For example, the model update unit 1214 may adjust the values of one or more of α 1, α 2, β 1, β 2, 1, and 2 in the first optimization model. The processor 220 (e.g., the model update unit 1214 of the predicted blood pressure determination module 1200) may then generate an updated first optimization model.

At 1316, the processor 220 (e.g., the first blood pressure determination unit 1204 of the predicted blood pressure determination module 1200) may designate the second blood pressure generated in the iteration as the first blood pressure. Process 1300 may then return to 1306 to generate a new second blood pressure based on the first blood pressure and the updated first optimization model.

Having thus described the basic concepts, it will be apparent to those of ordinary skill in the art having read this application that the foregoing disclosure is to be construed as illustrative only and is not limiting of the application. Various modifications, improvements and adaptations of the present application may occur to those skilled in the art, although they are not explicitly described herein. Such modifications, improvements and adaptations are proposed in the present application and thus fall within the spirit and scope of the exemplary embodiments of the present application.

Also, this application uses specific language to describe embodiments of the application. For example, "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic described in connection with at least one embodiment of the application. Therefore, it is emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the application may be combined as appropriate.

Moreover, those of ordinary skill in the art will understand that aspects of the present application may be illustrated and described in terms of several patentable species or situations, including any new and useful combination of processes, machines, articles, or materials, or any new and useful improvement thereof. Accordingly, various aspects of the present application may be embodied entirely in hardware, entirely in software (including firmware, resident software, micro-code, etc.) or in a combination of hardware and software. The above hardware or software may be referred to as "block," module, "" device, "" unit, "" component, "or" system. Furthermore, aspects of the present application may take the form of a computer program product embodied in one or more computer-readable media, with computer-readable program code embodied therein.

A computer readable signal medium may contain a propagated data signal with computer program code embodied therewith, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including electro-magnetic, optical, and the like, or any suitable combination. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code on a computer readable signal medium may be propagated over any suitable medium, including radio, cable, fiber optic cable, RF, etc., or any combination of the preceding.

Computer program code required for the operation of various portions of the present application may be written in any one or more of a variety of programming languages, including an object oriented programming language such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C + +, C #, VB.NET, Pvthon, etc., a conventional programming language such as C programming language, VisualBasic, Fortran1703, Perl, COBOL1702, PHP, ABAP, a dynamic programming language such as Python, Ruby, and Groovy, or other programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the use of a network service provider's network) or provided in a cloud computing environment or as a service, such as a software service (SaaS).

Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments of the invention have been discussed in the foregoing disclosure by way of example, it is to be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.

Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more embodiments of the invention. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed subject matter requires more features than are expressly recited in each claim. Indeed, claimed subject matter may require less than all features of a single disclosed embodiment.

Claims

1. A system for determining blood pressure, comprising:

at least one storage medium, the memory comprising a set of instructions;

a communication platform connected to a network; and

at least one processor in communication with the at least one storage medium, wherein the at least one processor, when executing the set of instructions, is configured to:

receiving a request from a terminal to determine a blood pressure of a first subject;

obtaining data related to the first subject, the data related to the first subject including data related to cardiac activity of the first subject and personal information related to the first subject;

extracting a target feature associated with the first subject from the data associated with the first subject;

determining an initial blood pressure of the first subject using a predictive model based on the target feature associated with the first subject;

determining a predicted blood pressure of the first subject using an optimization model based on the initial blood pressure; and

transmitting the predicted blood pressure of the first subject to the terminal in response to the request.

2. The system of claim 1, wherein to acquire the data related to the cardiac activity of the first subject, the at least one processor is further configured to:

communicating with a device connected to the first subject, the device configured to detect the cardiac activity of the first subject and generate a signal; and

receiving, from the device, the data related to the cardiac activity of the first subject generated based on the signal.

3. The system of claim 1, wherein the predictive model is generated by a first training process comprising:

obtaining historical data relating to at least two second subjects and at least two historical blood pressure measurements associated with the at least two second subjects, wherein the at least two second subjects include the first subject and the historical data relating to the at least two second subjects includes data relating to cardiac activity of the at least two second subjects and historical personal information relating to the at least two second subjects;

generating an initial prediction model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects; and

generating a predictive model related to the first subject based on the initial predictive model and at least a portion of the historical data related to the first subject.

4. The system of claim 3, wherein generating the initial predictive model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects comprises:

extracting a first set of features from the historical data associated with the at least two second subjects;

determining a second set of features based on the first set of features, the second set of features having dimensions smaller than dimensions of the first set of features;

clustering the historical data associated with the at least two second subjects into one or more clusters;

determining a historical target feature based on the second set of features;

for each cluster of the one or more clusters, determining a sub-prediction model based on the historical target features of the historical data in the each cluster and the historical blood pressure measurements corresponding to the historical data in the each cluster; and

designating the one or more sub-prediction modes corresponding to the one or more clusters as the initial prediction model.

5. The system of claim 4, wherein generating the initial predictive model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects further comprises:

normalizing the historical data associated with the at least two second subjects prior to extracting the first set of features from the historical data associated with the at least two second subjects.

6. The system of claim 4, wherein the second set of features is determined using a principal component analysis technique.

7. The system of claim 4, wherein generating the predictive model related to the first subject based on the initial predictive model and the at least a portion of the historical data related to the first subject comprises:

extracting historical target features from the historical data associated with the first subject;

determining a target cluster from the one or more clusters based on the historical target features of the historical data associated with the first subject; and

designating the sub-prediction mode corresponding to the target cluster as the prediction model related to the first subject.

8. The system of claim 1, wherein to determine the predicted blood pressure of the first subject using the optimization model based on the initial blood pressure, the at least one processor is further configured to:

initializing a first optimization model;

designating the initial blood pressure as a first blood pressure; and

and (3) performing iteration:

generating a second blood pressure based on the first blood pressure using the first optimization model;

determining whether a convergence condition is satisfied based on the first blood pressure and the second blood pressure;

in response to determining that the convergence condition is satisfied,

designating the first optimization model as the optimization model and designating the second blood pressure produced in an iteration as the predicted blood pressure; and

in response to determining that the convergence condition is not satisfied,

updating the first optimization model and designating the second blood pressure generated in an iteration as the first blood pressure.

9. The system according to claim 8, wherein to initialize the first optimization model, the at least one processor is configured to:

obtaining historical blood pressure measurements for the first subject and historical target characteristics for the first subject; and

initializing the first optimization model based on the historical blood pressure measurements of the first subject and the historical target features of the first subject.

10. The system of claim 1, wherein the predicted blood pressure of the first subject comprises a systolic pressure and a diastolic pressure, and the systolic pressure is predicted based on the diastolic pressure using the optimization model.

11. A method implemented on a computing device having at least one processor, memory, and a communication platform connected to a network for determining blood pressure, the method comprising:

12. The method of claim 11, wherein acquiring the data related to the cardiac activity of the first subject comprises:

13. The method of claim 11, wherein the predictive model is generated by a first training process comprising:

obtaining historical data related to the at least two second subjects and at least two historical blood pressure measurements associated with the at least two second subjects, wherein the at least two second subjects include the first subject and the historical data related to the at least two second subjects includes data related to cardiac activity of the at least two second subjects and historical personal information related to the at least two second subjects;

generating the predictive model related to the first subject based on the initial predictive model and at least a portion of the historical data related to the first subject.

14. The method of claim 13, wherein generating the initial predictive model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects comprises:

determining historical target features from the second set of features;

15. The method of claim 14, wherein generating the initial predictive model based on the historical data associated with the at least two second subjects and the at least two historical blood pressure measurements associated with the at least two second subjects further comprises:

16. The method of claim 14, wherein the second set of features is determined using a principal component analysis technique.

17. The method of claim 14, wherein generating the predictive model related to the first subject based on the initial predictive model and the at least a portion of the historical data related to the first subject comprises:

assigning a sub-prediction mode corresponding to the target cluster as the prediction model associated with the first subject.

18. The method of claim 11, wherein determining the predicted blood pressure of the first subject using the optimization model based on the initial blood pressure comprises:

initializing a first optimization model;

designating the initial blood pressure as a first blood pressure; and

and (3) performing iteration:

in response to determining that the convergence condition is satisfied,

in response to determining that the convergence condition is not satisfied,

19. The method of claim 18, wherein initializing the first optimization model comprises:

20. A non-transitory computer readable medium comprising at least one set of instructions for determining blood pressure, wherein the at least one set of instructions, when executed by at least one processor, cause the at least one processor to: