Physical examination data analysis system and method

Abstract

The invention relates to the field of health examination data analysis, and provides a health examination data analysis system and method.The health examination data analysis system comprises a refrigerator used for determining frequently-eaten food of a user in a first preset period; the wearable device is used for acquiring motion data of a user in a first preset period; the physical examination system is used for acquiring the gender and age of the user and first physiological data after a first preset period; the analysis system classifies the users according to the genders, ages and exercise amounts of the users, performs first clustering on the users according to frequently eaten food of the users in a first preset period, and prompts that the first physiological data of the corresponding user is abnormal when a deviation average value of the first physiological data of the users in each first cluster exceeds a first threshold value; and if the first physiological data average value of a certain cluster deviates from the first physiological data average values of all the users and exceeds a second threshold value, prompting that the first physiological data of the corresponding cluster user is abnormal. According to the scheme, the accuracy of physical examination data analysis can be improved.

Description

Technical Field

[0001] The present invention relates to the field of health examination data analysis, and in particular to a health examination data analysis system and method. Background Art

[0002] Physical examination data is one of the important bases for evaluating the health status of an individual. In the prior art, whether the subject is healthy is usually determined by collecting physiological data from the subject. However, physical examination data can only reflect the physiological index data of the subject at the current time point, and the normal index range is determined by the big data mean. In order to reduce false positives, the normal index range is usually determined to be relatively large. However, physiological data is affected by age, gender, daily diet structure, exercise volume, etc., and using absolute threshold data in the prior art to determine whether there is an abnormality will result in insufficient accuracy. Summary of the invention

[0003] In order to solve the problems in the prior art, the present invention provides a health examination data analysis system and method.

[0004] In one aspect of the present invention, a health checkup data analysis system is provided, the analysis system comprising: a refrigerator, the refrigerator comprising a camera, the camera obtaining the type of food in the refrigerator within a first preset period, and determining the user's commonly eaten food within the first preset period; a wearable device, the wearable device obtaining the user's exercise data within the first preset period; a health checkup system, the health checkup system obtaining the user's gender, age and first physiological data after the first preset period; an analysis system, the analysis system being communicatively connected to the refrigerator and the health checkup system; the analysis system classifies the users according to the users' gender, age and amount of exercise; for each classified user, the analysis system performs a first clustering on the users according to the commonly eaten food within the first preset period of the users, when the first physiological data of the users in each first cluster deviates from the average value by more than a first threshold value, the corresponding user's first physiological data is prompted to be abnormal; the average value of the first physiological data of each first cluster is calculated, if the average value of the first physiological data of a cluster deviates from the average value of the first physiological data of all users by more than a second threshold value, the corresponding cluster user's first physiological data is prompted to be abnormal, and it is prompted that it may be caused by an unreasonable diet structure.

[0005] Furthermore, a camera is arranged at the door of the refrigerator, and when the user takes out food, the user is prompted to record it, and the user points the food to the camera, which captures the image of the food.

[0006] Furthermore, for packaged food, the barcode / QR code of the food is pointed at the refrigerator camera to record the food type.

[0007] Furthermore, when the user takes out food, the refrigerator display screen displays the recognition result and the estimated weight, and the user can manually adjust the estimated weight and / or the recognition result.

[0008] Furthermore, when the user puts food into the refrigerator, the refrigerator compares the images of the food taken out of the refrigerator within a preset time period. If the similarity exceeds a preset threshold, the user is prompted to select the food consumption.

[0009] Furthermore, each time the user takes out food, the recognition result is displayed to the user in the form of text or images. If the recognition is incorrect, the user manually modifies the recognition result.

[0010] Furthermore, the refrigerator can also be set with the number of diners, and the average per-person consumption can be determined based on the number of diners.

[0011] Furthermore, after the camera obtains the food types in the refrigerator within the first preset period, the refrigerator labels the food types, and the labels include first major category labels and second minor category labels under each of the first major categories.

[0012] Furthermore, when clustering, the following method is used to determine the similarity of commonly eaten foods between users:

[0013]

[0014] Where S AB Indicates the similarity of the commonly eaten foods of users A and B in the first preset period;

[0015] Count represents a calculation function that calculates the number of elements in a collection;

[0016] C A1 Indicates the major categories of food that user A often eats during the first preset period;

[0017] C A2 A subcategory set representing the food that user A often eats during the first preset period;

[0018] C B1 Indicates the major categories of food that user B often eats during the first preset period;

[0019] C B2 It represents the sub-category set of the commonly eaten food of user B in the first preset period.

[0020] The present invention also provides a health examination data analysis method, which uses the above-mentioned health examination data analysis system to perform health examination data analysis.

[0021] The present invention can produce the following beneficial effects through the above technical solution:

[0022] 1. By analyzing the user's food intake, exercise conditions and physiological indicators, the system can provide each user with personalized health management suggestions and services, and guide them to improve their health in a targeted manner.

[0023] 2. By monitoring the changes in the physiological indicators of users, compared with traditional absolute indicators, the present invention can detect the physiological indicators that deviate from a specific population at an early stage without waiting for the indicators to reach a certain absolute abnormal threshold, thereby discovering potential health problems or physiological abnormalities and reminding users to seek medical treatment or adjust their lifestyle in time to prevent further development of the disease.

[0024] 3. The present invention can analyze whether the user's diet structure is reasonable based on the user's food intake and changes in physiological indicators, and provide corresponding diet suggestions and adjustment plans to help users improve their diet structure and maintain healthy eating habits.

[0025] 4. By monitoring the user's exercise data and eating habits, the system can help users monitor their health behaviors and remind them to maintain good living habits, such as regular exercise and a healthy diet. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 It is a system diagram of the present invention. DETAILED DESCRIPTION

[0028] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0029] This embodiment solves the above problem through the following steps:

[0030] In one embodiment, reference Figure 1 , the present invention provides a health examination data analysis system.

[0031] This system can analyze and evaluate personal physical examination data to help users understand their health status and identify potential health problems and risk factors. By monitoring and analyzing physical examination data, the system can provide personalized health management suggestions to help users take preventive measures, reduce the risk of disease, and detect problems early before the disease occurs. According to the user's physical examination data and personal situation, it provides health management guidance, including suggestions on diet, exercise, lifestyle, etc., to help users improve their health status. At the same time, it provides medical institutions and doctors with patients' health data and analysis reports to assist medical decision-making and improve the accuracy and efficiency of diagnosis and treatment.

[0032] The system includes at least one refrigerator, at least one wearable device, at least one physical examination system and an analysis system. Each component will be described in detail below.

[0033] A refrigerator, comprising a camera, wherein the camera obtains the type of food in the refrigerator within a first preset period and determines the user's frequently eaten food within the first preset period.

[0034] Refrigerators equipped with cameras can provide intelligent functions, allowing users to manage food more conveniently. There is no limit to the number of cameras and their locations on the refrigerator, as long as they can obtain images of food in the refrigerator.

[0035] During a first preset period (such as one month or other time length, the first preset period can be determined according to the physical examination needs or the physiological indicators to be monitored), the camera is allowed to continuously capture images in the refrigerator to cover the food conditions in different time periods. The camera can be set to capture at a scheduled time or automatically capture when the sensor detects that the refrigerator door is opened.

[0036] The image captured by the camera is transmitted to the system for processing and analysis. The system for analyzing the image can be the refrigerator itself, or a physical server or cloud server that communicates with the refrigerator.

[0037] Image processing and machine learning techniques are used to process and analyze images and extract food information from them. First, the collected images are preprocessed, including removing noise, adjusting image brightness and contrast, etc. Image processing techniques are then used to extract image features, which can be color, texture, shape, etc. The better features include color histogram, texture features, and shape description. A labeled training data set is prepared, including images of different types of food and corresponding labels. Then, these data are used to train machine learning models, such as convolutional neural networks (CNNs) and support vector machines (SVMs), for classifying food images. The trained model is evaluated to check the performance of the model on the validation set. Model optimization is performed based on the evaluation results, including adjusting the model structure and parameter tuning, to improve the classification accuracy and generalization ability of the model. The trained model is used to classify food images in the refrigerator. The model can output the food category to which each image belongs, thereby extracting the food information contained in the image. Food in the image is identified and classified through image recognition technology to identify common food types, such as vegetables, fruits, meat, etc.

[0038] Preferably, this embodiment identifies and records the food every time the refrigerator is opened and the food is taken out of the refrigerator. A camera can be set at the door of the refrigerator. When the user takes out the food, the user is prompted to record it. The user points the food at the camera so that the camera can accurately capture the image of the food. Furthermore, for packaged food, the type can be recorded more accurately by pointing the barcode / QR code of the food at the refrigerator camera. In this embodiment, by recording the food taken out by opening the door, it is possible to more accurately determine what the user has eaten.

[0039] Furthermore, when the user takes out the food, the refrigerator display screen displays the recognition result and prompts the user to choose to eat or discard. When the user chooses to discard, the food will not be recorded. The above method can reduce the error caused by discarding.

[0040] Furthermore, when the user takes out food, the refrigerator display screen displays the recognition result and the estimated weight, and the user can manually adjust the estimated weight. Through the above means, the food consumption and food type results can be combined to improve the accuracy of subsequent analysis.

[0041] Furthermore, when the user puts food into the refrigerator, the system compares the images of food taken out of the refrigerator within a preset time period. If the similarity exceeds a preset threshold, the user is prompted to select the amount of food consumed, such as consuming 1 / 3, half, etc. Through the above means, the problem of users putting unfinished food back into the refrigerator can be solved, thereby improving the accuracy of the record.

[0042] Furthermore, the identified food type data is recorded and stored in the system database. Ensure the integrity and accuracy of the data for subsequent analysis and application. Based on the recognition results, the system can show the user the food types in the refrigerator and allow the user to confirm and correct them. The user can manually add or correct the food types identified by the system to improve the accuracy of the system. For example, each time the user takes out food, the recognition result is shown to the user in the form of text or images. If the recognition is incorrect, the user manually modifies the recognition result. After receiving the modification result, the system can adjust the system parameters to improve the accuracy of recognition.

[0043] Furthermore, the refrigerator can also set the number of people eating, and determine the average consumption per person based on the number of people eating. The above means can handle the situation where there are multiple people in the family and improve the accuracy of subsequent data.

[0044] Analyze the food type data acquired by the camera during the first preset period. Count the frequency and weight of each food. Determine the user's frequently eaten food during the first preset period based on the frequency and weight data. For example, determine the top five food types ranked by consumption frequency / weight as the user's frequently eaten food during the first preset period, or determine the food ranked in the top preset position in consumption weight as the user's frequently eaten food during the first preset period.

[0045] A wearable device, wherein the wearable device obtains the user's motion data within a first preset period.

[0046] Wearable devices can be such as smart watches, smart bracelets, heart rate belts, etc., as long as the device can accurately record the user's exercise data. Within the first preset period (such as one month). The user lives and exercises normally within the first preset period, allowing the wearable device to automatically record the user's exercise data. This data may include the number of steps, distance, calories consumed, exercise time, etc. The data in the wearable device is synchronized to the analysis system regularly.

[0047] The analysis system is used to analyze the motion data within the first preset period, such as the user's exercise volume, exercise duration, exercise intensity and other information. At the same time, the motion data analysis results can be displayed to the user (such as through a mobile phone APP, mini program, etc.). If the data is found to be not accurate enough, the user can correct the automatically recorded data.

[0048] A physical examination system, wherein the physical examination system obtains the user's gender, age, and first physiological data after a first preset period.

[0049] When a user uses the physical examination system for the first time, the system requires the user to provide basic personal information, including gender and age. This information can be filled in by the user himself or entered into the system by medical staff.

[0050] After the user completes the first preset period, such as recording a month's diet and exercise data at the request of medical staff, the system arranges for the user to undergo a physical examination. The physical examination may include measuring physiological indicators such as height, weight, blood pressure, heart rate, blood sugar, and blood lipids.

[0051] The first physiological data may be at least one data that the user or medical staff is concerned about, such as blood pressure. This embodiment does not specifically limit the first physiological data, and the user or medical staff may determine the first physiological data according to the target to be monitored.

[0052] An analysis system is communicatively connected with the refrigerator, the physical examination system, and the wearable device.

[0053] The analysis system is the core of the system of this embodiment. The refrigerator, the physical examination system, and the wearable device all need to transmit data to the analysis system for analysis and processing. In order to exchange data, any communication connection method such as wired, wireless, Bluetooth, etc. in the prior art can be used to directly or indirectly establish a connection. At the same time, in order to achieve communication between systems, suitable data interfaces and communication protocols can also be defined. Private protocols can be used, or standardized data formats and communication protocols such as JSON, XML, HTTP, TCP / IP, etc. can be used to ensure compatibility and stability between systems.

[0054] The analysis system classifies users according to their gender, age, and amount of exercise.

[0055] Since gender, age, and amount of exercise have a great impact on a person's physiological data, this embodiment first classifies users according to their gender, age, and amount of exercise to reduce errors caused by gender, age, and amount of exercise.

[0056] Specifically, users are first divided into two groups, male and female, based on gender; then users in each group are further classified by age, such as one group under 10 years old, one group between 10 and 20 years old, etc.; and then they are grouped according to the amount of exercise, such as never exercising, exercising once a week on average, exercising twice a week on average, etc. Through the above grouping method, users are divided into multiple groups, and in each group, gender, age group, and amount of exercise are at the same level, thereby eliminating the influence of gender, age group, and amount of exercise on physiological indicators.

[0057] For each classified user, the analysis system performs a first clustering of the user according to the commonly eaten foods within the first preset period of the user. In order to cluster users according to commonly eaten foods, the commonly eaten foods within the first preset period can be vectorized first. First, the food is classified into major categories and the major categories are marked with the first serial number, including vegetables, fruits, meat, fish, beverages..., respectively represented by A, B, C, D, E..., and then the minor categories are marked in each major category. For example, if vegetables include cabbage, radish, and eggplant, they are represented by A1, A2, A3... respectively. If fruits include apples, grapes, and bananas, they are represented by B1, B2, B3... respectively. By analogy, the representation method of each type of food can be obtained.

[0058] When clustering, similarity calculation is required. For food, usually when the subcategories are different but the major categories are the same, there is still a certain degree of similarity. For example, one person likes apples and another likes bananas. Although the types of fruits are different, they are both fruits and have similar effects on the human body. Therefore, in order to comprehensively consider the major categories and minor categories, this implementation uses the following method to calculate similarity:

[0059]

[0060] Where S AB Indicates the similarity of the commonly eaten foods of users A and B in the first preset period;

[0061] Count represents a calculation function that calculates the number of elements in a collection;

[0062] C A1 Indicates the major categories of food that user A often eats during the first preset period, such as vegetables, fruits, meat, etc.;

[0063] C A2 A small set of foods that user A often eats during the first preset period, such as apples, cabbage, salmon, etc.;

[0064] C B1 Indicates the major categories of food that user B often eats during the first preset period, such as vegetables, fruits, meat, etc.;

[0065] C B2 It represents a small set of foods that user B often eats during the first preset period, such as apples, cabbage, salmon, etc.

[0066] For example, user A often eats A1, B2, C3, D4, E5, and user B often eats B1, B2, C2, D4, E4. Then:

[0067] C A1 ={A,B,C,D,E}

[0068] CA2 ={A1,B2,C3,D4,E5}

[0069] C B1 ={B,C,D,E}

[0070] C A2 ={B1,B2,C2,D4,E4}

[0071]

[0072] The above method can consider the similarities of large categories and small categories at the same time, which is more conducive to accurately measuring the similarity of food between different users.

[0073] After determining the method for calculating the similarity, any means in the prior art may be used for clustering, such as k-means clustering, to cluster users with similarity greater than a certain degree into one cluster.

[0074] In each first cluster, factors such as gender, age, amount of exercise, and food are similar, so the physiological data of each person in this cluster should be similar. If the first physiological data of the user in each first cluster deviates from the average value by more than the first threshold, it means that the user may have a change in physiological data caused by a lesion, indicating that the first physiological data of the corresponding user is abnormal and requires further detailed examination. In this embodiment, the first threshold can be set according to different physiological data. For physiological data with a large variance, the first threshold can be set larger, and vice versa; the first threshold can be an absolute value, such as 20 mmHg for blood pressure, or a percentage, such as a deviation of 10%. The deviation can be too large or too small.

[0075] In this embodiment, gender, age, and amount of exercise are divided in advance, and the gender, age, and amount of exercise are the same in the same cluster group. If the overall physiological data of a cluster is different from that of other clusters (obviously larger or smaller), it means that the abnormality of the cluster may be caused by the dietary structure, and the user of the cluster can be prompted to modify the dietary results. Specifically, the average value of the first physiological data of each first cluster is calculated. If the average value of the first cluster deviates from the overall average value by more than the second threshold, the corresponding cluster user indicator is prompted to be abnormal, and it is suggested that it may be caused by an unreasonable dietary structure.

[0076] Using the above technical means, by analyzing the user's food intake, exercise conditions and physiological indicators, the system can provide each user with personalized health management advice and services, and guide them to improve their health in a targeted manner. By monitoring the changes in the user's physiological indicators, compared with traditional absolute indicators, this system can detect physiological indicators that deviate from a specific population early without waiting for the indicators to reach a certain absolute abnormal threshold, thereby discovering potential health problems or physiological abnormalities, and reminding users to seek medical treatment or adjust their lifestyle in time to prevent further development of the disease. This system can analyze whether the user's diet structure is reasonable based on the user's food intake and changes in physiological indicators, and provide corresponding dietary advice and adjustment plans to help users improve their diet structure and maintain healthy eating habits. By monitoring the user's exercise data and eating habits, the system can help users monitor their health behaviors and remind them to maintain good living habits, such as regular exercise and a reasonable diet.

[0077] Another embodiment of the present invention further provides a health examination data analysis, using the aforementioned health examination data analysis system or any combination thereof to perform health examination data analysis.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A health examination data analysis system, characterized in that The analysis system comprises: A refrigerator, the refrigerator comprising a camera, the camera acquiring the type of food in the refrigerator within a first preset period, and determining the food that the user often eats within the first preset period; A wearable device, wherein the wearable device obtains motion data of a user within a first preset period; A physical examination system, wherein the physical examination system obtains the user's gender, age, and first physiological data after a first preset period; An analysis system, the analysis system being communicatively connected with the refrigerator and the physical examination system; The analysis system classifies users according to their gender, age, and amount of exercise; For each classified user, the analysis system performs a first clustering of the users according to the foods they commonly eat within a first preset period. When the first physiological data of the users in each first cluster deviates from the average value by more than a first threshold, the system prompts that the first physiological data of the corresponding user is abnormal. The average value of the first physiological data of each first cluster is calculated. If the average value of the first physiological data of a cluster deviates from the average value of the first physiological data of all users by more than a second threshold, the system prompts that the first physiological data of the users in the corresponding cluster are abnormal, and prompts that it may be caused by an unreasonable diet structure.

2. A health examination data analysis system according to claim 1, characterized in that: A camera is arranged at the door of the refrigerator. When a user takes out food, the user is prompted to record it. The user points the food toward the camera, and the camera captures the image of the food.

3. A health examination data analysis system according to claim 2, characterized in that: For packaged food, point the barcode / QR code of the food to the refrigerator camera to record the food type.

4. A health examination data analysis system according to claim 2, characterized in that: When the user takes out food, the refrigerator display shows the recognition result and estimated weight, and the user can manually adjust the estimated weight and / or recognition result.

5. A health examination data analysis system according to claim 2, characterized in that: When the user puts food into the refrigerator, the refrigerator compares the images of the food taken out of the refrigerator within a preset time period. If the similarity exceeds a preset threshold, the user is prompted to select the amount of food consumed.

6. A health examination data analysis system according to claim 2, characterized in that: Each time the user takes out food, the recognition result is displayed to the user in the form of text or image. If the recognition is incorrect, the user manually modifies the recognition result.

7. A health examination data analysis system according to claim 2, characterized in that: The refrigerator can also be set with the number of people dining, and the average consumption per person is determined based on the number of people dining.

8. A health examination data analysis system according to claim 1, characterized in that: After the camera obtains the food types in the refrigerator within the first preset period, the refrigerator marks the food types, and the marking includes a first major category mark and a second subcategory mark under each of the first major categories.

9. A health examination data analysis system according to claim 1, characterized in that: When clustering, the following method is used to determine the similarity of commonly eaten foods between users: Where S AB Indicates the similarity of the commonly eaten foods of users A and B in the first preset period; Count represents a calculation function that calculates the number of elements in a collection; C A1 Indicates the major categories of food that user A often eats during the first preset period; C A2 A subcategory set representing the food that user A often eats during the first preset period; C B1 Indicates the major categories of food that user B often eats during the first preset period; C B2 It represents the sub-category set of the commonly eaten food of user B in the first preset period.

10. A method for analyzing health examination data, characterized in that: Use the health examination data analysis system as described in any one of claims 1 to 9 to perform health examination data analysis.

Images