HK1248006B - Health tracking device - Google Patents

Description

Health tracking devices

Technical Field

The present invention relates to health tracking devices.

Background Art

A wearable fitness tracker may perform measurements of user activity based on sensor data received from a sensor. For example, a wearable fitness tracker may include an accelerometer that provides sensor data to estimate user activity during an exercise session. The wearable fitness tracker may provide information associated with the user activity during the exercise session for display. For example, the wearable fitness tracker may estimate one or more metrics (such as an estimated distance traveled, an estimated calorie expenditure metric, an estimated metabolic equivalent (MET) metric, etc.) based on the sensor data from the accelerometer and may provide the one or more metrics for display.

Summary of the Invention

According to some possible embodiments, a device may include one or more processors. The one or more processors may receive sensor data related to a user from a plurality of sensors. The plurality of sensors may include multiple types of sensors, the multiple types of sensors including one or more of an accelerometer, a heart rate sensor, a blood pressure sensor, a blood glucose sensor, a sweat sensor, a skin conductivity sensor, or an imaging sensor, and a spectrometer. The one or more processors may process the sensor data to determine the health status of the user, the sensor data being from the multiple types of sensors and being related to the user. The one or more processors may provide information identifying the health status of the user via a user interface based on processing the sensor data, the sensor data being from the multiple types of sensors and being related to the user.

According to certain possible embodiments, a non-transitory computer-readable medium may store one or more instructions that, when executed by one or more processors, may cause the one or more processors to receive a first spectral classification model. The first spectral classification model may be associated with identifying a health condition based on chemometric features. The first spectral classification model may be generated based on a calibration performed on a group of objects using a spectrometer. When executed by one or more processors, the one or more instructions may cause the one or more processors to obtain a set of attributes about a user. The set of attributes includes first sensor data about the user. When executed by one or more processors, the one or more instructions may cause the one or more processors to generate a second spectral classification model based on the first spectral classification model and the set of attributes about the user. The second spectral classification model may allow the characteristics of the user or food item to be determined. When executed by one or more processors, the one or more instructions may cause the one or more processors to periodically update the second spectral classification model based on second sensor data about the user.

According to certain possible embodiments, a method may include determining, by a device, an activity level of a user based on first sensor data, the first sensor data being related to the activity level of the user and coming from a first set of sensors. The method may include determining, by the device, a nutritional content of a group of foods for consumption by the user based on second sensor data, the second sensor data being related to the nutritional content of the group of foods and coming from a second set of sensors. The second sensor data may be obtained from a spectrometer. The method may include obtaining, by the device, a stored diet and exercise plan for the user. The stored diet and exercise plan may include goals related to the activity level of the user and the nutritional content of the group of foods for consumption by the user. The method may include determining, by the device, the user's compliance with the stored diet and exercise plan based on the user's activity level and the nutritional content of the group of foods. The method may include providing, by the device, recommendations related to the user's compliance with the stored diet and exercise plan based on the determination of the user's compliance with the stored diet and exercise plan.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a diagram outlining an exemplary embodiment described herein;

FIG2 is a diagram of an exemplary environment in which the systems and/or methods described herein may be implemented;

FIG3 is a diagram of exemplary components of one or more devices of FIG2;

4 is a flow chart of an exemplary process for providing health information based on sensor data associated with a plurality of sensors;

5A to 5D are diagrams of exemplary embodiments related to the exemplary process shown in FIG. 4 ;

FIG6 is a flow chart of an exemplary process for dynamically updating a classification model for spectral analysis; and

7A and 7B are diagrams of an exemplary embodiment related to the exemplary process shown in FIG. 6 .

DETAILED DESCRIPTION

The following detailed description of exemplary embodiments refers to the accompanying drawings, in which the same reference numerals in different drawings may represent the same or similar elements.

A wearable fitness tracker may utilize a set of accelerometers to obtain sensor data regarding a user's activity. For example, the wearable fitness tracker may determine an estimated distance traveled by the user based on the sensor data identifying movement of the wearable fitness tracker. The wearable fitness tracker may include information that can be used to estimate one or more metrics related to the user's activity. For example, the wearable fitness tracker may estimate calorie consumption based on the estimated distance traveled and general information relating travel distance to calorie consumption for an average person.

However, utilizing general correlations to determine one or more metrics may result in inaccurate calculations for a particular user. Additionally, a wearable fitness tracker may not take into account factors other than exercise that affect a user's health (nutrition, mood, illness, etc.), thereby limiting the usefulness of the wearable fitness tracker. The embodiments described herein may utilize sensor data associated with multiple sensors to provide health information associated with a user. In this manner, a single user device may determine the user's health information, thereby eliminating the need for the user to utilize multiple different devices, thereby reducing costs, power consumption, etc. relative to utilizing multiple different devices for health tracking. Additionally, the user device may utilize sensor data from multiple sensors to facilitate calibration of a model for spectral analysis, thereby improving the accuracy of the model relative to utilizing a model that has not been calibrated based on sensor data from other sensors.

Figure 1 is a diagram that provides an overview of an exemplary embodiment 100 described herein. As shown in Figure 1, exemplary embodiment 100 includes a user device, a healthcare provider application server, and a spectral calibration application server.

As further shown in FIG. 1 , the user device may receive sensor data from multiple sensors (e.g., receiving first sensor data from a first sensor, receiving second sensor data from a second sensor, ..., receiving nth sensor data from an nth sensor). For example, the user device 210 may utilize a set of integrated sensors to obtain sensor data (such as utilizing an integrated accelerometer sensor to obtain user activity data, utilizing an integrated heart rate sensor to obtain heart rate data, utilizing an integrated temperature sensor to obtain temperature data, etc.). Additionally or alternatively, the user device may utilize a camera to obtain sensor data. For example, the user device may utilize an integrated camera to capture images, such as images of a user's face (e.g., for facial recognition analysis), images of a user's skin (e.g., for skin condition analysis), a set of food images (e.g., a set of images for performing volumetric analysis on the food), etc. Additionally or alternatively, the user device may utilize a spectral sensor to obtain sensor data. For example, the user device may utilize a spectral sensor to determine a chemometric signature of an object (e.g., a user or food) and may classify the object based on the chemometric signature and a classification model. In another example, the user device can communicate with one or more sensors to obtain sensor data. For example, the user device can utilize a connection to a medical device to obtain sensor data recorded by the medical device.

As further shown in FIG1 , the user device may combine sensor data from multiple sensors to generate health information about the user. For example, based on sensor data about the user's activity level, the user's heart rate, the user's body temperature, etc., and stored information about the user's height, the user's weight, etc., the user device may determine the user's calorie consumption, which is associated with an exercise schedule. By also utilizing information other than accelerometer data (e.g., user's heart rate data and/or user's body temperature data), the user device may obtain a more accurate determination of calorie consumption than would be possible based on utilizing only accelerometer data.

Similarly, based on the image sensor data and the spectral sensor data, the user device can determine the volume of the food and the nutritional content of the food, respectively. Based on the use of spectral sensor data that identifies the ingredients of the food and sensor data that identifies the volume of the food, the user device can obtain a more accurate determination of the user's caloric intake than based on the use of user estimates of nutritional value. Based on the calorie expenditure determination and the calorie intake determination, the user device can determine the user's net calorie expenditure and can generate recommendations associated with the net calorie expenditure, such as nutritional recommendations, exercise recommendations, etc., to improve the user's health and/or the user's compliance with a diet and exercise plan. In this way, the user device performs more accurate determinations related to the user's health than determinations performed by a wearable fitness tracker, thereby allowing for improved fitness tracking.

As another example, a user device may process sensor data to determine diagnostic information about the user. For example, based on pattern recognition analysis performed on an image of the user, the user device may detect facial redness and pimples associated with the condition rosacea. Similarly, based on pattern recognition analysis performed on an image of the user, the user device may determine that the user is in a state of well-being and may correlate this state of well-being with other sensor data, such as sensor data indicating that the user has previously eaten a particular food. In this case, the user device may periodically provide suggestions related to altering the user's mood, such as suggesting that the user eat a particular food associated with the state of well-being.

As another example, the user device may process spectral sensor data to calibrate a classification model for spectroscopy. For example, the user device may utilize a first model to perform a first classification on the first spectral sensor data, such as classifying an observed chemometric feature as being associated with a specific person having a specific blood pressure and after eating a specific food. In this case, the user device may calibrate the first model based on the first spectral sensor data and other sensor data (e.g., sensor data identifying a specific blood pressure and a specific food) to generate a second model, and may utilize the second model to perform another classification on the second spectral sensor data. For example, the user device may utilize the second classification model to distinguish between a specific person when associated with a specific blood pressure and a specific person when associated with another blood pressure. In this way, the user device improves the classification model to perform spectroscopy with improved accuracy relative to a classification model that is not improved based on other sensor data.

As further shown in FIG1 , the user device may provide information based on processing sensor data from multiple sensors, such as health information related to the user, diagnostic information related to the user's health condition, and recommendations for improving the user's health and/or adherence to a diet and exercise plan. Adherence to a diet and exercise plan may be associated with reducing the severity of a health condition (e.g., a heart condition or an obesity condition), managing symptoms of a health condition (e.g., a diabetic health condition or a stress condition), reducing the likelihood of a health condition worsening (e.g., a degenerative disease), and achieving a desired health condition (e.g., improving diet, increasing whole grain intake, increasing muscle mass, or improving athletic performance). For example, the user device may provide real-time health updates for display via a user interface, such as information indicating net calorie expenditure determined based on calorie consumption and calorie intake. Additionally or alternatively, the user device may provide diagnostic information indicating that a particular condition has been detected for the user, and the data may be automatically sent to a specialist to establish an appointment between the user and the specialist. Additionally or alternatively, the user device may generate recommendations (such as suggestions for exercise regimens that allow the user to adhere to the diet and exercise plan) and may provide updates regarding the user's adherence to the diet and exercise plan for display via another device (e.g., a user device utilized by a personal trainer or nutritionist). In this way, the user device provides information customized to the user rather than generic information.

As further shown in Figure 1, the user device may provide health information to a healthcare provider application server for inclusion in a patient file associated with the user. In this way, the user device improves physician-patient consultations by improving data accuracy relative to a physician's reliance on manual patient reports of diet, exercise, etc., and reducing utilization of processing resources and/or power consumption relative to providing one or more user interfaces for receiving input for and/or modifying patient reports. In addition, or alternatively, the user device may provide calibration data to a spectral calibration application server. For example, the user device may provide calibration information determined from a spectral sensor and/or other sensor for use in calibrating and/or improving a classification model provided to one or more other user devices, thereby improving the spectral accuracy for the one or more other user devices relative to the one or more other user devices receiving a model that is not calibrated based on the calibration information.

In this way, the user device provides health information determined based on sensor data obtained from multiple sensors, thereby reducing costs and power consumption relative to the user utilizing multiple different devices. In addition, based on the integration of sensor data collection and processing into a single user device, the user device allows for the determination of health information that cannot be obtained via a single sensor (such as nutritional information obtained based on both volumetric analysis and spectral analysis, etc.). In addition, relative to a single initial calibration performed via a single device, the user device performs model calibration and model improvement based on both spectral data of an object for spectral analysis and other sensor data, thereby improving the calibration of the classification model used for spectral analysis.

As indicated above, FIG1 is provided merely as an example. Other examples are possible and may differ from the example that has been described with respect to FIG1 .

Figure 2 is a diagram of an exemplary environment 200 in which the systems and/or methods described herein may be implemented. As shown in Figure 2, environment 200 may include user devices 210, application servers 220, and network 230. The devices of environment 200 may be interconnected via wired connections, wireless connections, or a combination of wired and wireless connections.

The user device 210 includes one or more devices capable of receiving, generating, storing, processing, and/or providing health information. For example, the user device 210 may include a communication and/or computing device, such as a mobile phone (e.g., a smartphone, a wireless phone, etc.), a laptop computer, a tablet computer, a handheld computer, a gaming device, a wearable device (e.g., a smartwatch, a pair of smart glasses, a smart wristband, etc.), a medical device, a spectroscopic device (e.g., a wearable spectrometer device that performs near-infrared (NIR) spectroscopy, mid-infrared (mid-IR) spectroscopy, or Raman spectroscopy), or a similar type of device. In some embodiments, the spectroscopic device may include a hyperspectrometer (e.g., a hyperspectral imaging sensor). In some embodiments, the user device 210 may receive information from another device in the environment 200 and/or send information to another device in the environment 200.

The application server 220 includes one or more devices capable of storing, processing, and/or routing information (such as health information, calibration information, etc.). For example, the application server 220 may include a server that utilizes health information and/or information associated with the user device 210. In some embodiments, the application server 220 may include a calibration application server 220 that receives information associated with calibration of a spectral model based on a set of measurements performed by one or more user devices 210. Additionally or alternatively, the application server 220 may include a healthcare provider application server 220 (such as a hospital server, etc.) associated with routing information for healthcare providers. In some embodiments, the application server 220 may receive information from another device in the environment 200 and/or send information to another device in the environment 200.

The network 230 includes one or more wired and/or wireless networks. For example, the network 230 may include a cellular network (e.g., a Long Term Evolution (LTE) network, a 3G network, a Code Division Multiple Access (CDMA) network, etc.), a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., a public switched telephone network (PSTN)), a private network, an ad hoc network, an intranet, the Internet, a fiber-optic-based network, a cloud computing network, etc., and/or a combination of these or other types of networks.

The number and arrangement of the devices and networks shown in Figure 2 are provided as examples. In fact, there may be additional devices and/or networks, fewer devices and/or networks, different devices and/or networks, or devices and/or networks of different arrangements compared to the devices and networks shown in Figure 2. In addition, two or more devices shown in Figure 2 may be implemented within a single device, or the single device shown in Figure 2 may be implemented as multiple distributed devices. Additionally or alternatively, a group of devices (e.g., one or more devices) of environment 200 may perform one or more functions, which are described as being performed by another group of devices of environment 200.

FIG3 is a diagram of exemplary components of a device 300. Device 300 may correspond to user device 210 and/or application server 220. In some embodiments, user device 210 and/or application server 220 may include one or more devices 300 and/or one or more components of device 300. As shown in FIG3 , device 300 may include a bus 310, a processor 320, a memory 330, a storage component 340, an input component 350, an output component 360, and a communication interface 370.

The bus 310 includes components that allow communication between components of the device 300. The processor 320 is implemented in hardware, firmware, or a combination of hardware and software. The processor 320 may include a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), etc.), a microprocessor, and/or any processing component that interprets and/or executes instructions (e.g., a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), etc.). In certain embodiments, the processor 320 may include one or more processors that can be programmed to perform functions. The memory 330 includes a random access memory (RAM), a read-only memory (ROM), and/or another type of dynamic or static storage device (e.g., flash memory, magnetic storage, optical storage, etc.) that stores information and/or instructions for use by the processor 320.

Storage component 340 stores information and/or software related to the operation and use of device 300. For example, storage component 340 may include a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optical disk, a solid-state disk, etc.), a compact disk (CD), a digital versatile disk (DVD), a floppy disk, a cassette, a magnetic tape, and/or another type of non-transitory computer-readable medium, along with a corresponding drive.

Input components 350 include components that allow device 300 to receive information, such as via user input (e.g., a touch screen display, a keyboard, a keypad, a mouse, buttons, switches, a microphone, etc.). Additionally or alternatively, input components 350 may include sensors for sensing information (e.g., a global positioning system (GPS) component, an accelerometer, a gyroscope, an actuator, etc.). Output components 360 include components that provide output information from device 300 (e.g., a display, a speaker, one or more light emitting diodes (LEDs), etc.).

The communication interface 370 includes transceiver-type components (e.g., a transceiver, a separate receiver and transmitter, etc.) that enable the device 300 to communicate with other devices, such as via a wired connection, a wireless connection, or a combination of a wired connection and a wireless connection. The communication interface 370 can allow the device 300 to receive information from another device and/or provide information to another device. For example, the communication interface 370 can include an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi interface, a cellular network interface, etc.

Device 300 may perform one or more of the processes described herein. Device 300 may perform these processes in response to processor 320 executing software instructions stored by a non-transitory computer-readable medium, such as memory 330 and/or storage component 340. A computer-readable medium is defined herein as a non-transitory memory device. A memory device includes memory space within a single physical storage device or memory space distributed across multiple physical storage devices.

The software instructions may be read from another computer-readable medium or from another device into the memory 330 and/or storage component 340 via the communication interface 370. The software instructions stored in the memory 330 and/or storage component 340, when executed, may cause the processor 320 to perform one or more of the processes described herein. Additionally or alternatively, hard-wired circuitry may be used in place of or in combination with the software instructions to perform one or more of the processes described herein. Thus, the embodiments described herein are not limited to any specific combination of hardware circuitry and software.

The number and arrangement of components shown in FIG3 are provided as examples. In practice, device 300 may include additional components, fewer components, different components, or components arranged differently than those shown in FIG3. Additionally or alternatively, a group of components (e.g., one or more components) of device 300 may perform one or more functions that are described as being performed by another group of components of device 300.

FIG4 is a flow chart of an exemplary process 400 for providing health information based on sensor data associated with a plurality of sensors. In some embodiments, one or more processing blocks of FIG4 may be performed by user device 210. In some embodiments, one or more processing blocks of FIG4 may be performed by another device or group of devices (such as application server 220, etc.) that is separate from or includes user device 210.

As shown in FIG4 , process 400 may include obtaining sensor data from a set of sensors (block 410). For example, user device 210 may obtain sensor data from the set of sensors of user device 210. In some embodiments, user device 210 may obtain sensor data based on communicating with the set of sensors. For example, user device 210 may cause a heartbeat sensor of user device 210 to activate and record sensor data regarding the user's heartbeat. Similarly, user device 210 may cause an accelerometer sensor of user device 210 to activate and record accelerometer sensor data regarding movement of user device 210 (e.g., activity of a user using user device 210).

In certain embodiments, user device 210 may obtain sensor data based on prompts provided via a user interface. For example, user device 210 may provide a prompt to cause the user to utilize user device 210 to capture an image of the user (e.g., a photograph of the user's face or a photograph of a portion of the user's skin). Additionally or alternatively, user device 210 may provide a prompt to cause the user to utilize user device 210 to capture an image of food. For example, user device 210 may be configured to capture an image of a meal and may process the image of the meal to identify the nutritional content of the meal. Additionally or alternatively, user device 210 may capture the image automatically. For example, user device 210 may determine that a meal is ready, such as based on the time of day, location data indicating that user device 210 is at a restaurant, social media information indicating that user device 210 is at a restaurant, etc., and may automatically activate an imaging sensor to capture an image. Similarly, user device 210 may determine that an imaging sensor of user device 210 is pointed at the user based on an accelerometer, a touch sensor, etc., and may cause the imaging sensor to capture an image of the user.

In certain embodiments, user device 210 may monitor an imaging sensor to obtain sensor data. For example, user device 210 may perform object recognition techniques on a set of images captured via the imaging sensor to determine whether the captured set of images includes information that can be used to provide a health report, such as determining that a particular image includes food, determining that a particular image includes a user, etc. In this case, user device 210 may select a particular image for processing (e.g., a volume analysis processing technique, an emotion analysis processing technique, or a skin condition processing technique).

In some embodiments, the user device 210 may obtain sensor data based on a trigger. For example, the user device 210 may detect a user's interaction with a user interface and may be caused to obtain sensor data. Additionally or alternatively, the user device 210 may periodically obtain sensor data. For example, based on determining that a threshold time period (e.g., one hour, one day, or one week) has passed, the user device 210 may obtain sensor data. Similarly, at a specific time of day, the user device 210 may obtain sensor data. In some embodiments, the user device 210 may obtain the sensor data based on other sensor data. For example, the user device 210 may detect that the user has completed an exercise program based on monitoring a heart rate sensor (e.g., based on heart rate sensor data), and may obtain other sensor data based on determining that the user has completed the exercise program. Similarly, the user device 210 may identify the user based on chemometric signature sensor data from a spectrometer sensor, and may obtain other sensor data based on identifying the user.

As further shown in FIG4 , process 400 may include processing sensor data to obtain health information (block 420). For example, user device 210 may process sensor data to obtain health information. In some embodiments, user device 210 may process sensor data to identify one or more metrics related to the user's health. For example, user device 210 may determine an activity level metric based on sensor data associated with various types of sensors, such as an accelerometer, a heart rate sensor, a skin temperature sensor, a blood pressure sensor, and the like. Similarly, user device 210 may determine a nutritional metric based on sensor data associated with a blood pressure sensor, a blood glucose sensor, a sweat sensor, a skin conductivity sensor, and the like.

In certain embodiments, the user device 210 may utilize specific processing techniques to process the sensor data. For example, the user device 210 may utilize a classification technique associated with a classification model to identify the content of a food based on spectral measurements (e.g., chemometric features), and may determine health information related to food consumption based on the identified content of the food. In certain embodiments, the user device 210 may perform classification based on sensor data received from a hyperspectrometer, which may be user-difficult and have errors related to the distance from the object compared to utilizing another type of sensor. Additionally or alternatively, the user device 210 may apply pattern detection techniques, facial recognition techniques, three-dimensional depth sensing techniques, and the like to an image of the user to determine the user's mood, fatigue level, stress level, migraine symptoms, skin condition, and the like. Additionally or alternatively, the user device 210 may utilize color classification techniques to determine that the color of urine in an image corresponds to a specific health condition, such as insufficient or excessive consumption of a specific vitamin (e.g., B vitamin deficiency), and the like.

Additionally or alternatively, the user device 210 may utilize volumetric analysis techniques to process sensor data. For example, the user device 210 may utilize a set of images of the food (e.g., a set of images captured from different directions and/or positions) to determine the volume of the food. Additionally or alternatively, the user device 210 may utilize sensor data captured by a depth sensing module, a gesture recognition module, etc. to perform volumetric analysis. In certain embodiments, the user device 210 may determine the quality of the food based on the volumetric analysis. For example, the user device 210 may utilize information that identifies the density of the food (e.g., sensor data, image recognition of the food, or the user's selection of the food and the corresponding stored food density data), and may determine the quality of the food based on the density (e.g., which may be used to determine the nutritional content of the food based on spectral analysis that indicates the nutritional content on a per-mass basis). Additionally or alternatively, the user device 210 may utilize a set of images of the user to determine the volume of a part of the user, such as the volume of the shoulder, a skin lump, etc. In this manner, user device 210 identifies the volume of an object for comparison with the volume of the object at another time to determine the nutritional content of the object (e.g., by identifying other sensor data as the type of food for the object), etc.

In some embodiments, the user device 210 may process the sensor data using a comparison technique (such as comparing first sensor data recorded at a first time with second sensor data recorded at a second time). For example, the user device 210 may compare a first three-dimensional image of the user at a first time period with a second three-dimensional image of the user at a second time period to identify changes in the user's appearance, such as mole growth (e.g., corresponding to tumor growth), changes in breast shape (e.g., corresponding to cyst growth), changes in body shape (e.g., corresponding to weight gain), changes in stoichiometric characteristics (e.g., corresponding to changes in blood composition associated with a diabetic-type disorder), etc.

As further shown in FIG4 , process 400 may include providing health information (block 430). For example, user device 210 may provide health information. In some embodiments, user device 210 may provide a health report via a user interface for the user to review. For example, user device 210 may provide a report including health information identifying the user's net calorie expenditure (e.g., a comparison of calorie intake to calorie expenditure). Additionally or alternatively, user device 210 may provide information identifying a portion of sensor data processed by user device 210, such as a set of vital statistics determined for the user (e.g., blood pressure, body temperature, heart rate, or sweat level).

In some embodiments, the user device 210 may provide recommendations based on sensor data. For example, the user device 210 may determine a set of health recommendations (e.g., improving obesity conditions, managing diabetes conditions, preventing degenerative diseases, improving athletic performance, or meeting nutritional goals), may select a specific health recommendation from the set of health recommendations (e.g., based on the user's metabolic rate, the user consumes a specific amount of calories during a specific meal), and may provide health information including the specific health recommendation for display via a user interface. Additionally or alternatively, the user device 210 may select another health recommendation (e.g., predicting the user's improved mood level based on consuming specific foods or participating in a specific exercise regimen), and may provide other health recommendations for display via another user device 210 (e.g., a user device utilized by a physician, life coach, trainer, or personal chef).

In certain embodiments, user device 210 may provide alerts based on sensor data. For example, based on sensor data identifying a user's health condition, user device 210 may provide a specific alert identifying the health condition for display via a user interface. In this case, the alert may identify the health condition, the severity of the health condition, and so on. Additionally or alternatively, user device 210 may provide an alert for display via another user device 210. For example, user device 210 may identify a medical specialist associated with the health condition and send an alert for display via another user device 210 utilized by the medical specialist. Similarly, user device 210 may send an alert to dispatch emergency response personnel to the user. For example, when user device 210 determines that the severity of the health condition meets a threshold severity, user device 210 may utilize location-determination technology to determine the user's location and send an alert to an ambulance dispatch system, a hospital, or the like to dispatch emergency response personnel to the user's location. In certain embodiments, user device 210 may provide a health report along with the alert. For example, user device 210 may provide information identifying the user's blood pressure, the user's heart rate, the user's body temperature, etc. for use by doctors, emergency management personnel, etc.

In some embodiments, user device 210 may provide calibration data. For example, when user device 210 sends spectral data about a user for use in calibrating a model, user device 210 may also provide health information about the user for use in calibrating the model. In this case, the health information may be used to calibrate the model to a specific health condition (e.g., a first chemometric characteristic may be determined to correspond to a first sweat level, and a second chemometric characteristic may be determined to correspond to a second sweat level).

In certain embodiments, user device 210 may provide an augmented reality image for display based on sensor data. For example, when user device 210 determines the calorie content of a food item based on an image of the food item, user device 210 may determine a portion of the food item that, when consumed, would correspond to meeting a nutritional goal. In this case, user device 210 may provide an image of the food item (with the portion of the food item highlighted), an augmented reality display of the food item (with the portion of the food item highlighted), and the like for display. In this manner, the user may be provided with information indicating how much food item needs to be consumed to meet a nutritional goal.

Although Figure 4 illustrates example blocks of process 400, in some implementations, process 400 may include additional blocks, fewer blocks, different blocks, or a different arrangement of blocks than those depicted in Figure 4. Additionally or alternatively, two or more blocks of process 400 may be performed in parallel.

Figures 5A through 5D are diagrams of an example implementation 500 related to the example process 400 shown in Figure 4. Figures 5A through 5D illustrate an example of providing health information based on sensor data associated with a plurality of sensors.

5A and shown by reference numeral 510, a wearable user device 210 (e.g., a smartwatch including a set of sensors) may store a set of diet and exercise goals received from a healthcare provider application server 220. For example, the set of diet and exercise goals may be generated based on a user's consultation with a healthcare provider (e.g., a doctor, a personal trainer, or a nutritionist) and may be sent to the wearable user device 210 to allow the wearable user device 210 to monitor the user's diet and exercise to determine and improve compliance with the set of diet and exercise goals. Assume that the set of diet and exercise goals includes a diet goal associated with the user having a threshold amount of carbohydrate intake during the day, an exercise goal associated with the user meeting a threshold level of physical activity during the day, and a combined diet and exercise goal associated with the user meeting a threshold net calorie expenditure (e.g., calorie consumption is greater than calorie intake during the day).

5B and shown by reference numeral 515, the wearable user device 210 uses data from a spectral sensor to perform spectral analysis on the food to determine the content of the food, such as protein content, fat content, or carbohydrate content. As shown by reference numeral 520, the wearable user device 210 uses a set of images from an imaging sensor to perform image analysis on the food to determine the volume of the food. It is assumed that the wearable user device 210 determines the user's carbohydrate intake and calorie intake based on the content of the food (e.g., a spectral analysis indicating the food content) and the volume of the food (e.g., a volumetric analysis indicating the food content). It is assumed that based on the carbohydrate intake, the wearable user device 210 provides information indicating compliance with the diet goal via the user interface, such as a warning indicating the amount of carbohydrates the user is allowed to consume during the remainder of the day.

As shown in FIG5C and by reference numeral 525, the wearable user device 210 performs activity analysis on the user's exercise regimen using data from a heart rate sensor, an accelerometer sensor, and the like. Assume that the wearable user device 210 determines calorie consumption and a physical activity level based on the activity analysis performed on the user's exercise regimen. Assume that based on the physical activity level, the wearable user device 210 provides information indicating compliance with the exercise goal via the user interface, such as information indicating that the user has met a threshold physical activity level for the day.

As shown in Figure 5D and by reference numeral 530, wearable user device 210 determines compliance with the combined diet and exercise goal and generates an activity plan for the user, the activity plan being associated with causing the user to perform a specific activity (e.g., a specific number of steps) that is associated with corresponding changes to sensor data associated with the user (e.g., causing the sensor data to indicate that the user's calorie consumption for the day has increased). Assume that wearable user device 210 determines that the user has failed to meet the combined diet and exercise goal based on the user's calorie intake exceeding the user's calorie consumption. Assume also that wearable user device 210 determines that the user's walking a specific number of steps before the end of the day corresponds to achieving the combined diet and exercise goal based on the user's previous activity. As shown by reference numeral 535, wearable user device 210 provides a warning indicating that the user is to walk a specific number of steps before the end of the day, and continues to monitor user activity and provides warnings to cause the user to walk a specific number of steps. In this way, the wearable user device 210 can improve compliance with combined diet and exercise goals relative to a pedometer device that only provides information indicating the number of steps taken and estimated calorie consumption without including other data and without comparing estimated calorie consumption with calorie intake.

As indicated above, Figures 5A to 5D are provided merely as examples. Other examples are possible and may differ from the examples that have been described with respect to Figures 5A to 5D.

FIG6 is a flow chart of an exemplary process 600 for dynamically updating a classification model for spectral analysis. In some embodiments, one or more processing blocks of FIG6 may be performed by user device 210. In some embodiments, one or more processing blocks of FIG6 may be performed by another device or group of devices, such as application server 220, that is separate from or includes user device 210.

As shown in FIG6 , process 600 may include obtaining a first classification model for spectral analysis (block 610). For example, user device 210 may obtain a first classification model for spectral analysis. A classification model (e.g., a spectral classification model) may refer to a model that can be used to identify an object or a characteristic of the object based on chemometric features obtained for the object being spectrally analyzed. For example, a classification model may include information associated with a set of chemometric features for a set of samples (e.g., a set of people or a set of foods), and user device 210 may utilize the classification model to determine whether the chemometric features of a person correspond to a specific characteristic of the person (e.g., blood sugar level). Similarly, user device 210 may utilize another classification model to determine whether the chemometric features of a food correspond to a specific nutritional content of the food.

In certain embodiments, the user device 210 may obtain a first classification model from a specific application server 220, which is associated with calibrating the first classification model. For example, the specific application server 220 may perform spectral analysis on a calibration set (e.g., a set of identified subjects) using a spectrometer and may utilize processing techniques (e.g., optimization techniques for distinguishing between various chemometric characteristics of the calibration set) to generate the first classification model. In this case, the user device 210 may receive the first classification model based on the calibration model optimized by the application server 220. In certain embodiments, the user device 210 may obtain a specific classification model that was calibrated based on spectral analysis performed on a group of individuals. For example, the application server 220 may perform a first spectral analysis on a first individual and a second spectral analysis on a second individual, and may generate the first classification model to account for differences (e.g., related to different body compositions) in the chemometric characteristics (e.g., associated with blood glucose levels) of the first and second individuals.

In some embodiments, the user device 210 may obtain the first classification model based on requesting the first classification model. For example, the user device 210 may send a request for the first classification model and may receive the first classification model from the application server 220 based on sending the request. Additionally or alternatively, the user device 210 may obtain the first classification model from a data structure of the user device 210. For example, the user device 210 may include the first classification model stored via the data structure. In some embodiments, the user device 210 may generate the first classification model. For example, the user device 210 may receive a set of chemometric features associated with a set of objects and may generate the first classification model based on the set of chemometric features. Additionally or alternatively, the user device 210 may perform a set of spectral measurements on a set of known objects to obtain the chemometric features of the set of known objects and may generate the first classification model based on the chemometric features of the set of known objects.

As further shown in FIG6 , process 600 may include obtaining a set of attributes of an object for spectral analysis (block 620). For example, user device 210 may obtain the set of attributes of the object for spectral analysis. In some embodiments, user device 210 may determine the set of attributes based on sensor data from one or more sensors. For example, user device 210 may utilize sensors of user device 210 to determine blood glucose level, body temperature, etc., about the user based on sensor data recorded by the sensors. Additionally or alternatively, user device 210 may communicate with a sensor (e.g., a Bluetooth-enabled sensor) (e.g., via network 230) to receive sensor data associated with attributes of an object (e.g., a user) for spectral analysis.

In some embodiments, the user device 210 may obtain one or more attributes of an object, which are associated with classifying the object. For example, the user device 210 may obtain information identifying the gender of an object, the age of an object, the race of an object, etc. In this case, the user device 210 may obtain information from a data structure stored by the user device 210. Additionally or alternatively, the user device 210 may obtain information from an application server 220 (e.g., an application server associated with a health care provider and storing information about the user). In some embodiments, the user device 210 may obtain information via a user interface. For example, the user device 210 may generate a user interface and provide a set of prompts for display via the user interface, and may detect an interaction of the user with the user interface, wherein the interaction is associated with providing a set of responses to the set of prompts.

In certain embodiments, user device 210 may obtain properties of a subject for spectral analysis, the properties being related to a raw absorption spectrum of the subject for spectral analysis. For example, user device 210 may utilize an integrated spectral sensor to perform spectral measurements on a user and determine a chemometric signature of the user (e.g., a raw absorption spectrum). Additionally or alternatively, user device 210 may communicate with the spectral sensor to cause the spectral sensor to determine a chemometric signature associated with the user, and user device 210 may receive the chemometric signature from the spectral sensor based on the communication with the spectral sensor.

As further shown in FIG6 , the process 600 may include generating a second classification model for spectral analysis based on the set of attributes and the first classification model (block 630). For example, the user device 210 may generate the second classification model for spectral analysis based on the set of attributes and the first classification model. In some embodiments, the user device 210 may optimize the first classification (or quantitative) model using a model optimization technique (such as a support vector machine classifier (SVM) optimization technique or a support vector regression (SVR) optimization technique, etc.) and generate the second classification (or quantitative) model. For example, the user device 210 may optimize the first classification model (e.g., generated based on a group of people) to generate a second classification model for performing classification related to the user of the user device (e.g., determining the user's blood glucose level). In this way, the user device 210 takes into account differences between people (e.g., differences in body composition).

Additionally or alternatively, the user device 210 may optimize a first classification model (e.g., generated based on a first spectrometer associated with the application server 220) to generate a second classification model for use with a second spectrometer associated with the user device 210. For example, the user device 210 may generate a second classification model for classifying food products based on spectral sensor data obtained via the second spectrometer. In this manner, the user device 210 accounts for differences between spectrometers, thereby improving spectral accuracy relative to utilizing a single classification model generated by the application server 220 for each user device 210 utilized by each user.

As further shown in FIG6 , process 600 may include performing spectral analysis using a second classification model (block 640). For example, user device 210 may perform spectral analysis using the second classification model. In some embodiments, user device 210 may determine a metric associated with a user (e.g., the subject of the spectral analysis) using the second classification model. For example, user device 210 may identify a characteristic (e.g., blood sugar level, triglyceride level, ketone level, insulin level, skin condition, or personal identity) based on spectral sensor data associated with the user and the second classification model (e.g., a first chemometric feature may correspond to a first triglyceride level of the user when classified based on the second classification model, and a second chemometric feature may correspond to a second triglyceride level when classified based on the second classification model). In this case, user device 210 may identify a health condition, such as a change in skin thickness, a change in skin density, a change in skin collagen level, a change in capillary density, etc., based on detecting the characteristic.

In certain embodiments, user device 210 may utilize other sensor data about the user to perform spectral analysis. For example, user device 210 may determine that a user is associated with a particular skin condition based on both the user's chemometric characteristics and skin conductivity sensor data about the user. Similarly, user device 210 may determine the nutritional content (e.g., calorie content, carbohydrate content, protein content, or fat content) of a food based on both the food's chemometric characteristics and a volumetric analysis of the food. In this way, user device 210 may combine sensor data from multiple sensors to determine and provide health information associated with the user.

In some embodiments, user device 210 may provide information identifying a health condition based on performing spectral analysis. For example, user device 210 may generate a user interface and may provide information identifying a health condition for display via the user interface. Additionally or alternatively, user device 210 may provide information identifying a health condition to application server 220 (e.g., for inclusion in a patient medical record associated with the user). Additionally or alternatively, user device 210 may provide information identifying the nutritional content of a food. In some embodiments, user device 210 may provide the results of performing spectral analysis for further processing. For example, user device 210 may perform spectral analysis and may include data identifying the spectral analysis in a data set that is processed to determine and provide health information, as described herein with respect to FIG. 4 .

As further shown in FIG6 , process 600 may include periodically updating the second classification model (block 650). For example, user device 210 may periodically update the second classification model. In certain embodiments, user device 210 may optimize the second classification model. For example, when user device 210 performs spectral analysis on a user, user device 210 may utilize the spectral analysis and/or other sensor data to improve the second model using an optimization technique (such as an SVM optimization technique). In this case, user device 210 may utilize the improved second model for subsequent spectral analysis. In this manner, user device 210 improves the accuracy of the second classification model relative to utilizing a second classification model that does not utilize the spectral analysis and/or other sensor data for the improvement process. Furthermore, over time, the model continues to become more accurate and/or more specific to the user of the model, thereby allowing user device 210 to more accurately identify the user's characteristics, perform more accurate diagnosis of the user's condition, and so on. Furthermore, utilizing the SVM optimization technique reduces the utilization of processor and/or memory resources and significantly reduces the time required to generate a new classification model. Additionally, the incremental training approach may allow user device 210 to develop accurate models as user device 210 receives sensor data, rather than requiring sufficient processing and/or memory resources to generate an initial model based on a large amount of sensor data associated with an increasing number of sensors associated with the Internet of Things.

As further shown in FIG6 , the process 600 may include periodically providing feedback information associated with causing the first classification model to be updated (block 660). For example, the user device 210 may provide feedback information associated with causing the first classification model to be updated. In some embodiments, the user device 210 may provide information identifying a second classification model to cause the first classification model to be updated. For example, the user device 210 may provide information identifying the second classification model to the application server 220 to cause the application server 220 to update the first classification model.

In some embodiments, the user device 210 may provide information identifying sensor data obtained by the user device 210 to enable updating of the first classification model. For example, when the user device 210 obtains sensor data indicating a specific body temperature and performs spectroscopic measurements on a user having the specific body temperature to obtain a chemometric signature of the user at the specific body temperature, the user device 210 may provide information identifying the specific body temperature and the chemometric signature to be included in the first classification model (e.g., for identifying the specific body temperature based on similar chemometric signatures).

Additionally or alternatively, the user device 210 may provide demographic information to update the first classification model. For example, when the first classification model is generated based on the chemometric characteristics of the blood glucose levels of a group of men, the user device 210 may determine the chemometric characteristics of the blood glucose levels of women. In this case, the user device 210 may transmit information identifying the chemometric characteristics of women to optimize the first classification model, thereby performing distributed calibration on the model to improve the accuracy of the model relative to a classification model generated using static calibration based on a limited sample group (e.g., a group of men).

Although Figure 6 illustrates example blocks of process 600, in some implementations, process 600 may include additional blocks, fewer blocks, different blocks, or a different arrangement of blocks than those depicted in Figure 6. Additionally or alternatively, two or more blocks of process 600 may be performed in parallel.

Figures 7A and 7B are diagrams of an exemplary embodiment 700 related to the exemplary process 600 shown in Figure 6. Figures 7A and 7B illustrate an example of dynamically updating a classification model for spectral analysis.

As shown in FIG7A and by reference numeral 705, the calibration application server 220 may generate a general model (e.g., a general classification model) based on sample data obtained via a spectral sensor associated with the calibration application server 220 and related to a sample group of people. As shown by reference numeral 710, the wearable user device 210 receives the general model from the calibration application server 220. As shown by reference numeral 715, the wearable user device 210 may obtain chemometric feature data related to the user, such as chemometric features of a portion of the user's body. As shown by reference numeral 720, the wearable user device 210 may utilize support vector machine (SVM) optimization techniques to generate a local model (e.g., another classification model) for the user based on the chemometric feature data, other data (e.g., sensor data (e.g., the user's heart rate, the user's skin conductivity, or the user's blood glucose level), demographic data (e.g., the user's age, the user's body mass index, or the user's gender), etc.). It is assumed that the wearable user device 210 stores the local classification model for classifying one or more other chemometric features obtained about the user (e.g., for diagnosing a condition of the user).

As shown in FIG7B and by reference numeral 725, the wearable user device 210 provides information associated with the local model (such as chemometric characteristic data, one or more other chemometric characteristics, data about the user, etc.) to update the general model. As shown by reference numeral 730, the calibration application server 220 performs an update on the general model based on the received information associated with the local model.

As indicated above, Figures 7A and 7B are provided merely as examples. Other examples are possible and may differ from the examples that have been described with respect to Figures 7A and 7B.

In this manner, the user device 210 utilizes sensor data from multiple sensors to provide health information about the user, improve the calibration of one or more sensors (e.g., calibration of a classification model associated with a spectral sensor), and the like. Furthermore, by utilizing sensor data from multiple sensors, the user device 210 reduces the cost and power consumption associated with requiring a device for each sensor. Furthermore, by utilizing optimization techniques to calibrate the spectral sensor, the user device 210 reduces the utilization of processing and/or memory resources relative to generating a new classification model each time additional sample data is available, and reduces costs relative to generating a classification model based on sufficient sample data to perform all unimproved classifications.

The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Modifications and variations are possible in light of the above disclosure or may be acquired from practice of the embodiments.

As used herein, the term component is intended to be broadly interpreted as hardware, firmware, and/or a combination of hardware and software.

Certain embodiments are described herein in conjunction with threshold values. As used herein, satisfying a threshold value may refer to a value being greater than a threshold value, more than a threshold value, above a threshold value, greater than or equal to a threshold value, less than a threshold value, less than a threshold value, below a threshold value, less than or equal to a threshold value, equal to a threshold value, and the like.

Certain user interfaces have been described herein and/or shown in the figures. User interfaces may include graphical user interfaces, non-graphical user interfaces, text-based user interfaces, and the like. A user interface may provide information for display. In some embodiments, a user may interact with the information, such as by providing input via an input component of a device that provides the user interface for display. In some embodiments, the user interface may be configurable by the device and/or the user (e.g., the user may change the size of the user interface, the information provided via the user interface, the location of the information provided via the user interface, etc.). Additionally or alternatively, the user interface may be preconfigured to a standard configuration, a specific configuration based on the type of device that displays the user interface, and/or a set of configurations based on the capabilities and/or specifications associated with the device that displays the user interface.

It will be apparent that the systems and/or methods described herein can be implemented in various forms of hardware, firmware, or a combination of hardware and software. The actual dedicated control hardware or software code used to implement these systems and/or methods does not limit the implementation. Therefore, the operation and behavior of the systems and/or methods are not described herein with reference to specific software code - it should be understood that software and hardware can be designed to implement the systems and/or methods based on the description herein.

Even though particular combinations of features are recited in the claims and/or disclosed in the specification, these combinations are not intended to limit the disclosure of possible implementations. In fact, many of these features can be combined in ways not specifically recited in the claims and/or disclosed in the specification. Although each dependent claim listed below may directly depend on only one claim, the disclosure of possible implementations includes each dependent claim in combination with every other claim in the claim group.

As used herein, the elements, actions or instructions should not be interpreted as being essential or necessary unless clearly described in this way. In addition, as used herein, the articles "a" and "an" are intended to include one or more items and can be used interchangeably with "one or more". In addition, as used herein, the term "group" is intended to include one or more items (e.g., related items, unrelated items, combinations of related items and unrelated items, etc.), and can be used interchangeably with "one or more". When intended to only one item, the term "one" or similar language is used. In addition, as used herein, the terms "has", "have", "having" etc. are intended to be open terms. In addition, unless expressly stated otherwise, the term "based on" is intended to mean "at least partially based on".

Claims (20)

1. A non-transitory computer-readable medium storing instructions, the instructions comprising: One or more instructions, which, when executed by one or more processors, cause the one or more processors to: The user's activity level is determined based on first sensor data related to the user's activity level, the first sensor data including accelerometer sensor data and heart rate sensor data from a first set of sensors; Second sensor data related to the nutritional content of a set of food products is received from a second set of sensors, the second sensor data being obtained from at least the first spectrometer; Obtain a set of attributes about the user. The set of attributes includes first sensor data; Based on the set of attributes and the first classification model, the first classification model is calibrated to generate a second classification model used by the first spectrometer; The nutritional content of a set of foods consumed by the user is determined by using the second classification model based on the second sensor data. Obtain the user's stored diet and exercise plans. The stored diet and exercise plans include goals related to the user's activity level and the nutritional content of the set of foods consumed by the user; The user's adherence to the stored diet and exercise plan is determined based on the user's activity level and the nutritional content of the food group; and Based on determining the user's adherence to the stored diet and exercise plans, suggestions are provided related to the user's adherence to the stored diet and exercise plans. 2. The non-transitory computer-readable medium of claim 1, wherein the one or more instructions that cause the one or more processors to generate the second classification model cause the one or more processors to: Perform support vector machine classifier optimization or support vector regression classifier optimization on the first classification model to generate the second classification model; and Specifically, the one or more processors periodically update the one or more instructions of the second classification model: The support vector machine classifier optimization technique or the support vector regression classifier optimization technique is applied to the second classification model to update the second classification model. 3. The non-transitory computer-readable medium of claim 1, wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to: Obtain stoichiometric characteristics, The second sensor data includes information identifying the stoichiometric characteristics, and The stoichiometric characteristics are determined based on sensor data from another spectrometer; The classification of the stoichiometric features is performed based on the second classification model. The classification is associated with identifying health status; and Information is provided to identify the health status based on the classification of the performed chemometric characteristics. 4. The non-transitory computer-readable medium of claim 1, wherein the one or more instructions that cause the one or more processors to obtain the set of attributes of the user cause the one or more processors to: The data from the first set of sensors is obtained. The first set of sensors also includes at least one of the following: Another spectrometer, Blood pressure sensor Blood glucose sensor Sweat sensor Skin conductivity sensor, or Imaging sensor. 5. The non-transitory computer-readable medium of claim 1, wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to: Information related to the second classification model or the second sensor data is periodically sent to update the first classification model. 6. The non-transitory computer-readable medium of claim 1, wherein the one or more instructions, when executed by the one or more processors, cause the one or more processors to: Receive stoichiometric features from another spectrometer. The other spectrometer is different from the first spectrometer; The health status is determined based on the second classification model. The health status refers to the user's blood glucose level; and Provides information to identify the user's blood glucose level. 7. The non-transitory computer-readable medium of claim 6, wherein the one or more instructions, when executed by the one or more processors, further: Receive another chemometric feature from the user from the first spectrometer; Based on the second classification model, a threshold difference is detected between the stated health condition and another health condition related to the other chemometric characteristic; and A warning is issued based on the detected threshold difference. 8. A health tracking device, comprising: Memory; and One or more processors, said one or more processors being used for: The user's activity level is determined based on first sensor data related to the user's activity level, the first sensor data including accelerometer sensor data and heart rate sensor data from a first set of sensors; Second sensor data related to the nutritional content of a set of food products is received from a second set of sensors, the second sensor data being obtained from at least the first spectrometer; Obtain a set of attributes about the user, the set of attributes including first sensor data; The first classification model is calibrated based on the set of attributes and the first classification model to generate a second classification model for use by the first spectrometer; By utilizing the second classification model, the nutritional content of the group of food consumed by the user is determined based on the second sensor data; and Obtain the user's stored diet and exercise plans. The stored diet and exercise plans include goals related to the user's activity level and the nutritional content of the set of foods consumed by the user; The user's adherence to the stored diet and exercise plan is determined based on the user's activity level and the nutritional content of the food group; and Based on determining the user's adherence to the stored diet and exercise plans, suggestions are provided related to the user's adherence to the stored diet and exercise plans. 9. The device of claim 8, wherein, when the one or more processors generate the second classification model: Perform support vector machine classifier optimization or support vector regression classifier optimization on the first classification model to generate the second classification model; and Wherein, when the one or more processors periodically update the second classification model: The support vector machine classifier optimization technique or the support vector regression classifier optimization technique is applied to the second classification model to update the second classification model. 10. The device of claim 8, wherein the one or more processors further: Obtain stoichiometric characteristics, The second sensor data includes information identifying stoichiometric characteristics, and The stoichiometric characteristics are determined based on sensor data from another spectrometer; The classification of the stoichiometric features is performed based on the second classification model. The classification is associated with identifying health status; and Information is provided to identify the health status based on the classification of the performed chemometric characteristics. 11. The device of claim 8, wherein the one or more processors, upon obtaining the set of attributes concerning the user: The data from the first set of sensors is obtained. The first set of sensors also includes at least one of the following: Blood pressure sensor, Blood glucose sensor Sweat sensor Skin conductivity sensor, or Imaging sensor. 12. The device of claim 8, wherein the one or more processors further: Information related to the second classification model or the second sensor data is periodically sent to update the first classification model. 13. The device of claim 8, wherein the one or more processors further: Receive stoichiometric features from another spectrometer. The other spectrometer is different from the first spectrometer; The health status is determined based on the second classification model. The health status refers to the user's blood glucose level; and Provides information to identify the user's blood glucose level. 14. The device of claim 8, wherein the one or more processors further: Receive another chemometric feature from the user from the first spectrometer; Based on the second classification model, a threshold difference is detected between a health status and another health status related to the other chemometric characteristic; and A warning is issued based on the detected threshold difference. 15. A health tracking method, comprising: The device determines the user's activity level based on first sensor data related to the user's activity level, the first sensor data including accelerometer sensor data and heart rate sensor data from a first set of sensors; Second sensor data related to the nutritional content of a set of food products is received from a second set of sensors, the second sensor data being obtained from at least the first spectrometer; The device obtains a set of attributes about the user, the set of attributes including first sensor data; The device calibrates the first classification model based on the set of attributes and the first classification model to generate a second classification model for use by the first spectrometer; The device determines the nutritional content of a group of foods consumed by the user based on the second sensor data by utilizing the second classification model. The device obtains the user's stored diet and exercise plans. The stored diet and exercise plans include goals related to the user's activity level and the nutritional content of the set of foods consumed by the user; The device determines the user's adherence to a stored diet and exercise plan based on the user's activity level and the nutritional content of the set of food items; and Based on determining the user's adherence to the stored diet and exercise plans, the device provides recommendations related to the user's adherence to the stored diet and exercise plans. 16. The method of claim 15, wherein generating the second classification model comprises: The first classification model is optimized using either a support vector machine classifier or a support vector regression classifier to generate the second classification model. And the periodic updating of the second classification model includes: The support vector machine classifier optimization technique or the support vector regression classifier optimization technique is applied to the second classification model to update the second classification model. 17. The method of claim 15, further comprising: Obtain stoichiometric characteristics, The second sensor data includes information identifying the stoichiometric characteristics, and The stoichiometric characteristics are determined based on sensor data from another spectrometer; The classification of the stoichiometric features is performed based on the second classification model. The classification is associated with identifying health status; and Information is provided to identify the health status based on the classification of the performed chemometric characteristics. 18. The method of claim 15, wherein obtaining the set of attributes about the user comprises: The data from the first set of sensors is obtained. The first set of sensors also includes at least one of the following: Another spectrometer, Blood pressure sensor, Blood glucose sensor Sweat sensor Skin conductivity sensor, or Imaging sensor. 19. The method of claim 15, further comprising: Information related to the second classification model or the second sensor data is periodically sent to update the first classification model. 20. The method of claim 15, further comprising: Receive stoichiometric features from another spectrometer. The other spectrometer is different from the first spectrometer; The health status is determined based on the second classification model. The health status refers to the user's blood glucose level; and Provides information to identify the user's blood glucose level.