TWI861469B - Smart health management system, method and computer readable medium thereof - Google Patents

Abstract

The present invention is a smart health management system and a method thereof. An application is installed on a mobile device to capture medication information images of a medication bag. A health management unit of A health management server performs image recognition on a medication information image to obtain name of the medical institution contained in the medication bag. An image model corresponding to the name of the medical institution is selected from the plural image models stored in a database unit of the health management server, and a medication information of the medicine bag is obtained through analysis. Based on the medication information of the medicine bag, corresponding information about the medication bag can be provided and a reminder notification can be issued to the mobile device to avoid the user's medication errors or omissions. The present invention also provides a computer-readable medium for executing the method of the present invention.

Description

Intelligent health management system, method and computer-readable medium thereof

The present invention relates to health management technology, and in particular to an intelligent health management system, method and computer-readable medium that can provide medication management and notification.

After the user goes to a medical institution for treatment, he will receive medicines for the treatment of his condition. Generally speaking, large medical institutions often provide medicines to users in a one-drug-one-bag method, that is, the medical institution puts each medicine in a different medicine bag, and the user will receive many medicine bags; secondly, each medicine bag has the medication information of the corresponding medicine, and the user must check one by one when receiving it to confirm whether it is consistent with the medicine received; in addition, after receiving it, the user must also take the medicine according to the information recorded in the medication information. In other words, not only is the collection process quite cumbersome, but there are also many things to pay attention to during the medication process, which will cause trouble for the user in taking medicine.

In view of the above problems, how to provide a smart medication management technology, especially the management and integration of medicine bag information, by unifying the management of various medicines and medicine bags received, and further providing relevant medicine information and medication reminders, will become the goal that people in this technology field are eager to pursue.

In order to solve the above-mentioned problems of the prior art, the present invention discloses an intelligent health management system, which includes: a mobile device having an application for capturing medication information images from a medicine bag; and a health management server having a health management unit and a database unit, wherein the health management unit includes an image recognition unit for performing image recognition on the medication information image from the mobile device to obtain the name of the medical institution, and the database unit includes an image model database for storing a plurality of image models, wherein the The image recognition module first performs block analysis on the medication information image, selects a block from the medication information image to perform image recognition, and obtains the name of the medical institution. The health management unit obtains the corresponding image model from the multiple image models based on the name of the medical institution, and uses the obtained corresponding image model to analyze the image recognition to obtain the medication information corresponding to the medicine bag, so as to provide the mobile device with the corresponding information about the medicine bag and issue a reminder notification based on the medication information.

In one embodiment, the application includes a medication capture module for capturing the medication information image.

In another embodiment, the medication information includes drug information corresponding to the medicine bag, so that the health management unit can search according to the drug information to obtain drug details corresponding to the drug information, and provide the drug details to the mobile device.

In another embodiment, the mobile device includes a display element, the database unit includes a pharmacopoeia database, and the health management unit includes a text analysis module, wherein the text analysis module analyzes the text obtained after the image recognition module performs image recognition on the medication information image to obtain the drug information including the drug name, and the health management unit searches the pharmacopoeia database according to the drug name to obtain the drug details, and the display element displays the drug details.

In another embodiment, the medication information further includes the name of the recipient of the medication corresponding to the medication bag, and the application stores personal information including the individual's name, and the application determines the correctness of the name of the recipient of the medication by comparing the name of the recipient of the medication with the individual's name.

In another embodiment, the medication information includes the total number of medications collected, recording the number of medication bags.

In another embodiment, the health management unit includes a recognition learning module for marking the name of the medical institution, the name of the person who received the medicine, the total number of medicines received, and the drug information of the medicine bag, so as to train the multiple image models or add new image models.

In another embodiment, the application includes a storage module for storing medication reminders and a notification module for providing medication notifications based on the medication reminders, and the database unit includes a personal medication database, so that the data processing module of the health management unit records the medication reminders in the personal medication database.

In another embodiment, the database unit further includes a health education database having health education information, so that the health education recommendation module of the health management unit can obtain the health education information from the health education database according to the indication of the medication information, so as to push the health education information to the mobile device.

The present invention further discloses an intelligent health management method, which is executed by a computer device, and includes the following steps: using an application on a mobile device to capture a medication information image from a medicine bag; allowing an image recognition module of a health management unit of a health management server to receive the medication information image and perform image recognition, first performing block analysis on the medication information image, and selecting a block from the medication information image to perform image recognition, so as to obtain the name of a medical institution; The health management unit obtains a corresponding image model from an image model database having multiple image models in a database unit of the health management server according to the name of the medical institution; the health management unit uses the obtained corresponding image model to analyze the image recognition to obtain medication information corresponding to the medicine bag; and the health management unit provides the mobile device with corresponding information about the medicine bag and issues a reminder notification according to the medication information.

In another embodiment, after the health management unit performs image recognition to obtain the medication information, the following steps are further included: the text analysis module of the health management unit analyzes the text obtained after the image recognition module performs image recognition on the medication information image to obtain drug information including the drug name; the health management unit searches the pharmacopoeia database of the database unit according to the drug name to obtain the drug details; and the display element of the mobile device displays the drug details.

In another embodiment, the present invention further includes the following steps: the database unit is provided with a health education database for storing health education information; and the health education recommendation module of the health management unit obtains the health education information from the health education database according to the indication of the medication information, so as to push the health education information to the mobile device.

The present invention further discloses a computer-readable medium, which is applied to a computing device or a computer and stores instructions for executing the aforementioned intelligent health management method.

As can be seen from the above, the intelligent health management system, method and computer-readable medium of the present invention allow users to capture medication information images of medicine bags through the application installed on the mobile device, and use the health management server to identify and analyze the medication information images, thereby achieving the effects of drug inspection, medication notification, medication history, and health education recommendation.

1:Medicine bag

11: Medication information

111: Name of medical institution

112: Name of the person who received the medicine

113: Total number of medicines received

114: Drug information

2: Mobile devices

21: Applications

211: Medication capture module

212: Storage module

213: Notification module

22: Display component

3: Health management server

31: Health management unit

311: Image recognition module

312: Identification Learning Module

313: Text parsing module

314: Data processing module

315: Health education recommendation module

32: Database unit

321: Image model database

322:Pharmacopoeia Database

323: Personal medication database

324: Health Education Database

S401~S405: Steps

S501~S508: Steps

Figure 1 is a system architecture diagram of the intelligent health management system of the present invention.

Figure 2 is a schematic diagram of the medication information of the medicine bag of the present invention.

Figure 3 is a system architecture diagram of the smart health management system of the present invention in a specific embodiment.

Figure 4 is a flow chart of the intelligent health management method of the present invention.

Figure 5 is a flow chart of another embodiment of the smart health management method of the present invention.

The following describes the technical content of the present invention through a specific concrete implementation form. People familiar with this technology can easily understand the advantages and effects of the present invention from the content disclosed in this manual. However, the present invention can also be implemented or applied through other different specific implementation forms.

Figure 1 is a system architecture diagram of the smart health management system of the present invention, and Figure 2 is a schematic diagram of the medication information of the medicine bag of the present invention, please refer to them together.

As shown in the figure, the smart health management system of the present invention includes a mobile device 2 and a health management server 3. The mobile device 2 captures the medication information image from the medicine bag 1, and the health management server 3 performs image recognition on the medication information image to generate medication information 11 of the medicine bag 1 and provide it to the mobile device 2. The health management server 3 can also issue a reminder notification for medication based on the medication information 11. The specific description of the smart health management system of the present invention is as follows.

The mobile device 2 captures medication information images from the medicine bag 1 through the installed application 21. Specifically, the mobile device 2 usually has an image capture element such as a camera, which can take a picture of the medicine bag 1 to obtain an image of the medicine bag 1, that is, the user captures the area of the medication information 11 of the medicine bag 1 through the application 21 to obtain the medication information image. In a specific embodiment, the application 21 may include a medication capture module 211 for capturing medication information images, that is, the application 21 captures the medication bag 1 through the medication capture module 211 to obtain the medication information image.

The health management server 3 is connected to the mobile device 2 in communication, for example, by a wireless network or the Internet, and can receive medication information images from the mobile device 2 for image recognition. Specifically, the health management server 3 includes a health management unit 31 having an image recognition module 311 and a database unit 32 having an image model database 321. The health management unit 31 uses the image recognition module 311 to perform image recognition on the medication information image, thereby converting the text image of the medication information 11 into text characters. Among them, the medical institution name 111 of the medicine bag 1 of the medical institution is usually placed at the top and in the center of the medicine bag 1, so the health management unit 31 sets the medical institution name 111 as an important recognition indicator of the medicine bag 1. During image recognition, the name of the medical institution 111 is first obtained from the medication information 11 of the medicine bag 1, and then the health management unit 31 uses the image recognition module 311 to perform image recognition on the medication information image, including the image recognition module 311 first performing block analysis on the medication information image, selecting a block from the medication information image to perform image recognition, and then obtaining the name of the medical institution 111. , in order to facilitate the subsequent identification of the medicine bag 1, the block analysis mentioned above refers to dividing the medication information image into multiple blocks, such as the common 9-grid format. Based on the fact that the medical institution name 111 is often at the top or in the center of the medicine bag 1, the top three blocks of the 9-grid can be selected for identification. This can reduce the amount of data to be identified, not only can the identification be faster, but also it is conducive to improving the identification accuracy. In addition, in order to avoid the situation where the medical institution name 111 cannot be found in the selected block, a mechanism for selecting other blocks can also be designed. For example, if the medical institution name 111 cannot be found in the top three blocks, the middle three blocks are further identified, and so on. The above description shows that when performing image recognition, the image can be divided into blocks to facilitate the recognition. However, the block location, quantity, number of blocks to be selected, and the order of selecting blocks can all be determined according to the design.

In another embodiment, in order to enhance the recognition accuracy, in addition to finding the medical institution name 111, the second feature setting can also be used to assist in image recognition. For example, the medical institution name 111 in Chinese and English is usually on the medicine bag 1, so if the English institution name is added, it will be helpful to enhance the recognition process; furthermore, if the Chinese institution name is affected by the image shooting or the medicine bag 1 is squeezed and the Chinese characters are incomplete, the English institution name can be used to assist in recognition. The Chinese and English recognition mechanisms are similar, only the text type is different, so it will not be repeated. In summary, through the setting of the second feature, the recognition result of the image recognition module 311 can be made more accurate.

The image model database 321 of the database unit 32 stores a plurality of image models. After the health management unit 31 obtains the medical institution name 111, the health management unit 31 obtains the corresponding image model from the plurality of image models in the image model database 321 according to the medical institution name 111, that is, selects the image model corresponding to the medical institution name 111 from the plurality of image models, and analyzes the image recognition result by the selected image model, that is, analyzes the text characters generated after the image recognition, and obtains the medication information 11 including the medical institution name 111 based on it, thereby achieving the purpose of ensuring the accuracy of the medication information captured from the medicine bag. Thereafter, the health management unit 31 can provide a medication reminder notification, such as a medication notification, to the mobile device 2 according to the medication information 11.

As shown in FIG2 , the medication information 11 of the medicine bag 1 specifically includes the above-mentioned medical institution name 111, the name of the person who received the medicine 112, the total number of medicines received 113, and the drug information 114. The description of the medication information 11 is as follows.

The user's name on the medicine bag is the name of the recipient 112, and on the application 21, the application 21 stores personal information including the personal name corresponding to the name of the user (i.e., the recipient), based on which the health management unit 31 can compare the name of the recipient 112 and the personal name to determine the correctness of the name of the recipient 112, and also ensure the correctness of the medication information 11 captured from the medicine bag 1. In other words, the health management unit 31 checks the name of the recipient 112 to confirm whether the name of the recipient 112 is the same as the personal name recorded in the personal information on the mobile device 2, so as to avoid the user taking the wrong medicine bag or mistakenly grabbing someone else's medicine bag. Therefore, if after image recognition, it is confirmed that the name 112 of the person who receives the medicine is the same as the personal name, then it is confirmed that the medicine bag currently being captured is the one that the user wants to receive. In other words, when the application 21 captures the medication information image, it will check the name 112 of the person who receives the medicine in the medication information 11 of the medicine bag 1. If it is different from the personal information, a warning window will pop up.

Furthermore, since large medical institutions usually implement one bag for one medicine, if the user picks up a wide variety of medicines, the user may miss taking medicine or fail to capture the medication information 11 of all medicine bags 1. Therefore, the health management unit 31 sets the total number of medicines collected 113 used to record the number of medicine bags as an important reminder indicator, that is, the total number of medicines collected 113 is used to remind the user of the number of all medicines, and when capturing the image of the medication information of the medicine bag 1, it is confirmed whether the medication information 11 of all medicine bags 1 has been captured. That is, after capturing the image of the medicine bag 1, the application 21 prompts according to the total number of medicines collected 113 of the medicine bag 1 to avoid missing some medicine bags.

In addition, the drug information 114 of the medicine bag 1 is provided for the health management unit 31 to search according to the drug information 114, so as to obtain the drug details corresponding to the drug information 114, and the health management unit 31 can provide the drug details to the mobile device 2. In detail, the health management unit 31 will search in the database according to the drug name in the drug information 114 (for example, using fuzzy search, precise search, etc.), so as to display the drug details including usage and dosage, indications, main ingredients, drug appearance and instructions, so as to provide the user with the setting of medication notification, so as to avoid taking the wrong medicine.

Figure 3 is a system architecture diagram of the smart health management system of the present invention in a specific embodiment, please refer to Figure 2 together.

As shown in the figure, this embodiment is roughly the same as the first embodiment, except that the mobile device 2 includes a display element 22, the application 21 includes a storage module 212 and a notification module 213, the health management unit 31 includes a recognition learning module 312, a text analysis module 313, a data processing module 314 and a health education recommendation module 315, and the database unit 32 includes a pharmacopoeia database 322, a personal medication database 323 and a health education database 324. The detailed description of this embodiment is as follows.

After the mobile device 2 uses the medication capture module 211 of the installed application 21 to capture the medication information image, the health management unit 31 will analyze the medical institution name 111 of the medicine bag 1 through the image recognition module 311, and select the appropriate image model from the image model database 321 to analyze the medication information 11 of the medicine bag. Since the text layout of the medication information 11 of different medical institutions is different, the present invention uses the recognition learning module 312 of the health management unit 31 to mark and select the medical institution name 111, the name of the person who receives the medicine 112, the total number of medicines received 113, and the drug information 114 of the medicine bag 1, so as to train the corresponding image model or add a new image model. Therefore, the present invention can mark and frame the medication information 11 through the health management unit 31, and then train or expand the image model to identify the medicine bags 1 used by different medical institutions, thereby achieving the effect of improving the recognition rate of the medication information image of the medicine bag 1.

The health management unit 31 can analyze the text obtained by the image recognition module 311 through the text analysis module 313 to obtain the drug name of the drug information 114. The health management unit 31 searches the pharmacopoeia database 322 according to the drug name to obtain the drug details and provides it to the mobile device 2 so that the drug details are displayed on the display element 22 so that the user can understand the relevant information of the drug received. In one embodiment, the health management unit 31 uses fuzzy search to search in the pharmacopoeia database 322. As can be seen from the above, after the image recognition module 311 of the health management unit 31 extracts the text characters of the medication information 11, it extracts the relevant necessary drug information 114 through the text analysis module 313. After extracting the drug name, the health management unit 31 performs fuzzy search and comparison in the pharmacopoeia database 322 to ensure that the captured and analyzed drug name is correct, and obtains the appearance and instructions of the drug corresponding to the drug name from the pharmacopoeia database 322. After completing the fuzzy search and comparison, the data obtained after the comparison is provided to the mobile device 2, and the mobile device 2 can use the display element 22 to display the detailed drug information 114.

The storage module 212 of the application 21 is used to store medication reminders, and the notification module 213 is used to notify medication according to the medication reminders stored in the storage module 212. In addition, the database unit 32 includes a personal medication database 323, and the health management unit 31 has a data processing module 314, wherein the health management server 3 can automatically obtain the stored medication reminders from the application 21 of the device 2 to record the medication reminders in the personal medication database 323. In other words, the user can set a medication reminder corresponding to the medicine bag 1 through the storage module 212, and the health management unit 31 will record the medication reminder in the personal medication database 323 through the data processing module 314. After the data processing module 314 records the medication reminder in the personal medication database 323, it can serve as a basis for the user to query personal medication status in the future. In addition, the notification module 213 of the mobile device 2 will regularly notify the user to take the medication according to the set medication time.

The health education database 324 of the database unit 32 has health education information for providing health education knowledge. The health management unit 31 compares the health education database 324 according to the indication of the medication information 11 through the health education recommendation module 315 to obtain the health education information corresponding to the indication, and then pushes it to the mobile device 2. That is, the health education recommendation module 315 of the health management unit 31 will compare the indication of the user's current medication record in the health education database 324, and according to the comparison result, push the health education information to the user from time to time to enhance the user's understanding of the current symptoms, thereby achieving the purpose of preventive health care. In another embodiment, the health education recommendation module 315 of the health management unit 31 will compare the user's historical records and push health education information to the user from time to time.

In summary, the present invention uses the application 21 of the mobile device 2 to capture the medication information image of the medicine bag 1, and then analyzes it through the health management unit 31, and combines medication identity and total quantity verification, medication notification and health education information push to prevent users from taking the wrong medication or forgetting to take medication, and can also enhance users' understanding of current symptoms.

FIG4 is a flow chart of the smart health management method of the present invention. The smart health management method of the present invention can be executed by a computer device such as a personal computer or a server. Please refer to FIG2 and FIG3 at the same time. The steps of the smart health management method of the present invention are as follows.

In step S401, capture the medication information image. The present invention uses the mobile device 2 to capture the image of the medicine bag 1, that is, the medication information image of the medicine bag 1 is captured through the application 21 installed on the mobile device 2. In one embodiment, the application 21 includes a medication capture module 211 for capturing the medication information image.

In step S402, the medication information image is image-recognized. The present invention is equipped with a health management server 3, which is provided with a health management unit 31 including an image recognition module 311, so as to receive the medication information image and perform image recognition to obtain the name of the medical institution 111.

In step S403, an image model is selected. The health management server 3 of the present invention is equipped with a database unit 32 that provides various types of data, wherein the database unit 32 includes an image model database 321 having multiple image models, so that the health management unit 31 can select a corresponding image model from the multiple image models according to the medical institution name 111.

In step S404, the image model is used to analyze the image recognition result. Specifically, the present invention uses the image model to analyze the image recognition result to obtain the medication information 11 corresponding to the medicine bag 1. In detail, since the medicine bag information formats of different medical institutions are different, the present invention provides corresponding image models according to different medical institutions, so that the image model's analysis result of the medicine bag 1 is more accurate. In actual operation, the name of the medical institution 111 on the medicine bag 1 is first obtained. The name of the medical institution 111 on the medicine bag 1 is presented in the middle and top mode. The image recognition module 311 of the health management unit 31 first recognizes the name of the medical institution 111 on the medicine bag 1. Regarding the recognition mechanism of the image recognition module 311 for the medication information image, for example, selecting a local block Identification or adding a second feature to assist in identification is beneficial to improving the identification speed and accuracy. The relevant details have been explained above and will not be repeated here. The correct medical institution name 111 is conducive to selecting an appropriate image module (i.e., an image module corresponding to the medical institution name 111) to analyze the medication information 11, thereby ensuring the accuracy of the medication information 11 captured from the medicine bag 1.

In step S405, a reminder notification is issued based on the medication information. In this step, the health management unit 31 issues a reminder notification to the mobile device 2 based on the medication information 11.

In another embodiment, the medication information 11 on the medicine bag 1 includes the name of the person who receives the medicine 112, the total number of medicines received 113, and the drug information 114. Specifically, the name of the person who receives the medicine 112 of the medication information 11 is obtained from the image captured by the medicine bag 1, and the application 21 stores personal information including the personal name. The application 21 compares the name of the person who receives the medicine 112 with the personal name to determine the correctness of the name of the person who receives the medicine 112, that is, the medicine bag 1 belongs to the user. In addition, the total number of medicines received 113 records the number of medicine bags, so that when the medication information image of the medicine bag 1 is captured, the application 21 can prompt according to the total number of medicines received 113 of the medicine bag 1 during the image capture process to avoid missing some medicine bags.

Accordingly, after the image is captured and recognized, the medication information 11 of the present invention converts the above-mentioned information such as the name of the recipient 112, the total number of medications received 113, the drug information 114, the unit, and the dosage into characters that can be interpreted by the program, and performs text semantic analysis according to different formats to check the information of the name of the recipient 112 and the total number of medications received 113 to avoid medication errors or omissions.

In addition, the drug information 114 is used by the health management unit 31 to search for drug details corresponding to the drug information 114, and the drug details are provided to the mobile device 2 for the user to view.

FIG5 is a flow chart of another embodiment of the smart health management method of the present invention, please refer to FIG2 and FIG3 at the same time. As shown in the figure, steps S501-S505 of this embodiment are roughly the same as S401-S405 of the first embodiment, and the difference is that after receiving the medication information image for image recognition, this embodiment repeats the following steps.

In step S506, the name of the drug is obtained according to the image recognition result of the medication information. In this step, the health management unit 31 is provided with a text analysis module 313 to analyze the text obtained by the image recognition module 311 through image recognition of the medication information image, thereby forming drug information 114 including the drug name.

In step S507, detailed drug information is retrieved based on the drug name. In short, the health management unit 31 searches the pharmacopoeia database 322 of the database unit 32 based on the drug name to obtain the detailed drug information. In one embodiment, the health management unit 31 uses fuzzy search to search the pharmacopoeia database 322.

In step S508, the drug details are displayed. This step is to display the retrieved drug details through the display element 22 of the mobile device 2.

In other words, the present invention converts the medication information 11 of the medicine bag 1 into text characters that can be interpreted by the application program and performs text semantic analysis, and then performs a fuzzy search on the drug name and the pharmacopoeia database 322 provided by the Food and Drug Administration, and after the search, more detailed drug information is compared for the user to confirm, such as indications, drug leaflets, drug appearance, etc. In addition to ensuring the accuracy of image recognition and text analysis, it can also provide users with more detailed medication information.

In one embodiment, the present invention further includes training of image models. Specifically, the health management unit 31 is provided with a recognition learning module 312 for marking the medical institution name 111, the name of the recipient 112, the total number of medications 113, and the drug information 114 of the medicine bag 1, and can be used to train multiple image models or add new image models. That is, in order to identify medicine bags 1 from different medical institutions, the present invention allows users to upload screenshots of medication information 11 of medicine bags 1 through the recognition learning module 312, and trains and expands the image model of the system by marking and selecting important information such as the name of the medical institution 111, the name of the person who receives the medicine 112, the total number of medicines received 113, and the drug information 114, thereby improving the recognition rate of the captured medication information 11 of the medicine bag 1.

The smart health management method of the present invention further includes a step of recommending personalized health education. Specifically, the database unit 32 is provided with a health education database 324 storing health education information, and the health management unit 31 has a health education recommendation module 315, so that the health education recommendation module 315 compares the indications of the medication information 11 in the health education database 324, obtains health education information based on the comparison, and pushes the health education information to the mobile device 2. That is, the health management server 3 uses the image of the medicine bag 1 captured by the current user and the indications of the corresponding medicine recorded in the health education database 324 to push videos or pictures related to health education information from time to time through the health education recommendation module 315 to enhance the user's understanding of the current symptoms and achieve the purpose of preventive health care.

In addition, the smart health management method of the present invention further includes the step of recommending personalized medication time notification. In short, the application 21 includes a storage module 212 for storing medication reminders and a notification module 213 for notifying medication according to the medication reminders, and the database unit 32 includes a personal medication database 323 and the health management unit 31 includes a data processing module 314 for recording the medication reminder in the personal medication database 323. In this way, after the application 21 captures the medication information 11 of the medicine bag 1, the user can display the detailed medication information through the display element 22 of the mobile device 2 for the user to confirm, and provide the user with the ability to set the medication time notification according to personalized needs to avoid forgetting to take medication.

In summary, the smart health management system and method of the present invention mainly allows the user to capture the medication information 11 of the medicine bag 1 through the medication capture module 211 of the application 21 of the mobile device 2, and then, through the image recognition module 311 of the health management unit 31 of the health management server 3, the medication information image is recognized to first determine the name of the medical institution 111 of the medicine bag 1 Then, the image model database 321 of the database unit 32 of the health management server 3 selects the appropriate image model for analysis according to the name of the medical institution 111, and then the text analysis module 313 of the health management unit 31 interprets the text characters. Since most of the large medical institutions now implement one medicine per bag, if there are many types of medicines, medication errors or omissions may occur. Therefore, the name of the person who receives the medicine 112 and the total number of medicines received 113 will be checked. After the check, the data processing module 314 of the health management unit 31 obtains the name of the medicine, performs a fuzzy search in the pharmacopoeia database 322, and obtains the detailed information of the medicine, and then allows the user to confirm the name of the person who receives the medicine 112 and the total number of medicines received in the medicine bag 1 through the display element 22 of the mobile device 2. The information of number 113 is then used to perform drug verification based on the fuzzy search of the drug name. After the search, the drug details are displayed on the mobile device 2 for the user to confirm, and a personalized medication reminder time setting is provided. That is, the present invention can provide confirmation of medication information and provide users with personalized medication time notifications based on their needs to avoid medication errors or omissions.

Furthermore, since the formats of the medication information 11 of the medicine bags 1 of different medical institutions are not the same, the recognition learning module 312 of the health management unit 31 of the present invention allows users to upload screenshot images of the medication information 11 of the medicine bag 1, and train and expand the image module by marking and selecting important information such as the name of the person who received the medicine 112, the total number of medicines received 113, and the drug information 114, so as to improve the recognition rate of the medication information captured from different medicine bags.

In addition, the health education recommendation module 315 of the health management unit 31 of the present invention pushes relevant video or picture health education information from time to time according to the current indications of the user's medication. That is, the health education recommendation module 315 finds appropriate health education information according to the aforementioned indications and pushes it to the user, thereby enhancing the user's understanding of the current symptoms and achieving the purpose of preventive health care.

In addition, the present invention also discloses a computer-readable medium, which is applied to a computing device or computer having a processor (e.g., CPU, GPU, etc.) and/or a memory, and stores instructions, and the computing device or computer can execute the computer-readable medium through the processor and/or memory to execute the above-mentioned method and each step when executing the computer-readable medium.

In summary, the smart health management system, method and computer-readable medium of the present invention, through the combination of the application 21 of the mobile device 2 and the health management server 3, allows the user to capture the medication information 11 of the medicine bag 1 through the application 21, and achieve the functions of drug inspection, medication notification, medication history, health education recommendation, etc. Therefore, the smart health management system and method of the present invention have the following features.

First, the smart health management system of the present invention can use the characteristics of the medical institution name on the medicine bag being centered and topped, as the information format of the medicine bag of different medical institutions is different, to first identify the medical institution of the medicine bag through the image recognition module, and then analyze the medication information according to the appropriate image module to ensure the accuracy of the captured medication information on the medicine bag.

Second, the smart health management system of the present invention uses a recognition learning module to upload medication information images on medicine bags, and trains and expands the image module by marking and selecting important information such as the name of the recipient, the total number of medications received, and drug information, thereby improving the recognition rate of medication information capture on medicine bags.

Third, the intelligent health management system of the present invention can recognize the medication information on the medicine bag through the text analysis module, and convert the information on the medicine bag, such as the name of the recipient, the total number of medicines received, the name of the medicine, the unit, and the dosage, into characters that can be interpreted by the program, and check the information of the name of the recipient and the total number of medicines received to avoid medication errors or omissions.

Fourth, the intelligent health management system of the present invention uses a data processing module to perform a fuzzy search with the pharmacopoeia database after obtaining the drug name, and compares more detailed drug information for users to confirm after the search, such as indications, drug instructions, drug appearance, etc.

Fifth, the smart health management system of the present invention can display detailed medication information for users to confirm after capturing medication information from the medicine bag through the mobile device application, and can also set medication time notifications according to personalized needs.

Sixth, the smart health management system of the present invention can push medications according to the current user's indications through the health education recommendation module from time to time, enhance the user's understanding of the current symptoms, and achieve the purpose of preventive health care.

The modules, units, devices, etc. of the present invention include a microprocessor and a memory, and algorithms, data, programs, etc. are stored in the memory or chip. The microprocessor can load data or algorithms or programs from the memory to perform data analysis or calculation, etc., which will not be elaborated here. In other words, the smart health management system and method of the present invention can be executed on electronic devices, such as general computers, tablets or servers, and perform data analysis and calculations after receiving image data. Therefore, the program performed by the smart health management system and method can be designed and constructed on electronic devices with components such as processors and memories through software to run on various types of electronic devices; in addition, each module or unit of the smart health management system and method can be composed of independent components, such as a calculator, memory, storage or firmware with a processing unit, which can all become components for realizing the present invention, and related models such as image models can also be presented in software programs, hardware or firmware architectures.

The above embodiments are only illustrative and not intended to limit the present invention. Anyone familiar with this technology may modify and change the above embodiments without violating the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is defined by the scope of the patent application attached to the present invention. As long as it does not affect the effect and implementation purpose of the present invention, it should be covered by this public technical content.

1:Medicine bag

11: Medication information

2: Mobile devices

21: Applications

211: Medication capture module

3: Health management server

31: Health management unit

311: Image recognition module

32: Database unit

321: Image model database

Claims (12)

A smart health management system includes: a mobile device having an application for capturing medication information images from a medicine bag, and the application stores personal information including personal name; and a health management server having a health management unit and a database unit, the health management unit including a computer for performing image recognition on the medication information images from the mobile device to obtain the name of a medical institution. The image recognition module is called an image recognition module. The database unit includes an image model database for storing multiple image models. The image recognition module first performs block analysis on the medication information image, selects a block from the medication information image to perform image recognition, and obtains the name of the medical institution. The health management unit obtains the corresponding image model from the multiple image models based on the name of the medical institution. The image recognition is analyzed using the corresponding image model to obtain medication information including the name of the person who receives the medicine corresponding to the medicine bag and the total number of medicines received recording the number of medicine bags, so as to provide the mobile device with corresponding information about the medicine bag and issue a reminder notification based on the medication information, wherein the application determines the accuracy of the name of the person who receives the medicine by comparing the name of the person who receives the medicine with the personal name, wherein, The block parsing refers to dividing the medication information image into multiple blocks. If the name of the medical institution cannot be found in the top multiple blocks, the multiple blocks in the middle are further identified. If the name of the medical institution cannot be found in the multiple blocks in the middle, the subsequent multiple blocks are further identified, and the second feature setting is used to assist in the image recognition, among which the Chinese and English medical institution names are identified to assist in the recognition. As described in claim 1, the medication information includes drug information corresponding to the medicine bag, so that the health management unit can search according to the drug information to obtain drug details corresponding to the drug information, so as to provide the drug details to the mobile device. The smart health management system as described in claim 2, wherein the mobile device includes a display element, the database unit includes a pharmacopoeia database, and the health management unit includes a text analysis module, wherein the text analysis module analyzes the text obtained after the image recognition module performs image recognition on the medication information image to obtain the drug information including the drug name, and the health management unit searches the pharmacopoeia database according to the drug name to obtain the drug details, and the display element displays the drug details. The smart health management system as described in claim 1, wherein the health management unit includes a recognition learning module for marking the name of the medical institution, the name of the person who received the medicine, the total number of medicines received, and the drug information of the medicine bag, so as to train the multiple image models or add new image models. The smart health management system as described in claim 1, wherein the application includes a storage module for storing medication reminders and a notification module for notifying medication based on the medication reminders, and the database unit includes a personal medication database, so that the data processing module of the health management unit records the medication reminder in the personal medication database. As described in claim 1, the database unit further includes a health education database having health education information, so that the health education recommendation module of the health management unit can obtain the health education information from the health education database according to the indication of the medication information, so as to push the health education information to the mobile device. A smart health management method is implemented by a computer device, and the method includes the following steps: using an application on a mobile device to capture a medication information image from a medicine bag, wherein the application stores personal information including a personal name; allowing an image recognition module of a health management unit of a health management server to receive the medication information image and perform image recognition, first performing block analysis on the medication information image, and selecting a block from the medication information image to perform image recognition to obtain the name of a medical institution; allowing the health management unit to obtain a corresponding corresponding image model from an image model database having multiple image models in a database unit of the health management server according to the name of the medical institution; allowing the health management unit to analyze the image using the obtained corresponding image model Identify to obtain medication information including the name of the person who receives the medicine corresponding to the medicine bag and the total number of medicines received recording the number of medicine bags, wherein the application determines the correctness of the name of the person who receives the medicine by comparing the name of the person who receives the medicine with the personal name; and the health management unit provides the mobile device with corresponding information about the medicine bag and issues a reminder notification based on the medication information, wherein the block solution Analysis means dividing the medication information image into multiple blocks, and if the name of the medical institution cannot be found in the top blocks, then further identifying the middle blocks, and if the name of the medical institution cannot be found in the middle blocks, then further identifying the subsequent blocks, and assisting the image recognition through the setting of the second feature, wherein the Chinese and English names of medical institutions are identified to assist in the recognition. The smart health management method as described in claim 7, wherein after the health management unit performs image recognition to obtain the medication information, the following steps are further included: the text analysis module of the health management unit analyzes the text obtained after the image recognition module performs image recognition on the medication information image to obtain drug information including the drug name; the health management unit searches the pharmacopoeia database of the database unit according to the drug name to obtain drug details; and the display element of the mobile device displays the drug details. The intelligent health management method as described in claim 7, wherein the health management unit further includes a recognition learning module for marking the name of the medical institution, the name of the recipient, the total number of medications received, and the drug information of the medicine bag, so as to train the multiple image models or add new image models. The smart health management method as described in claim 7, wherein the application includes a storage module for storing medication reminders and a notification module for notifying medication based on the medication reminders, and the database unit includes a personal medication database, so that the data processing module of the health management unit records the medication reminder in the personal medication database. The smart health management method as described in claim 7 further includes the following steps: the database unit is provided with a health education database for storing health education information; and the health education recommendation module of the health management unit obtains the health education information from the health education database according to the indication of the medication information, so as to push the health education information to the mobile device. A computer-readable medium, used in a computing device or a computer, stores instructions for executing the intelligent health management method as described in any one of claims 7 to 11.

Images