Intelligent vital sign monitoring system based on improved resNet34

Abstract

The invention provides an intelligent vital sign monitoring system based on improved resNet34, and relates to the technical field of sign monitoring, the system comprises a hardware part and a software part, the hardware part comprises an ECG vital sign detection module, the ECG vital sign detection module is used for detecting physiological sign data of a user, the software part is in data connection with the hardware part, and the hardware part is connected with the software part. And the software part is used for acquiring the physiological sign data of the user, operating an artificial intelligence model to process the physiological sign data of the user, identifying a physiological sign abnormal event, and giving an alarm when the physiological sign abnormal event is identified, so that the abnormal data can be quickly and accurately identified, the alarm is given, and the user experience is improved. Therefore, potential health risks of the user can be found in time, a health assessment report can be generated, and data support and effective suggestions are provided for the user to improve the health condition.

Description

Technical Field

[0001] The present invention relates to the technical field of vital sign monitoring, and in particular to an intelligent vital sign monitoring system based on improved resNet34. Background Art

[0002] The medical and health field has always been an important field of technological innovation and development. Against the backdrop of continuous progress in medical science and information technology, vital signs monitoring technology has developed rapidly. The miniaturization and intelligence of sensors, the popularization of wireless communication technology, and the rapid development of artificial intelligence algorithms provide technical support for the implementation of intelligent vital signs monitoring systems based on the improved resNet34. At the same time, people's growing demand for personalized health management also provides a solid foundation for the market demand for this system. With the aging of the population, the increase in the number of chronic disease patients, and the increasing inequality of medical resources, the traditional medical model can no longer meet the growing demand for health management. Therefore, the development of intelligent and convenient health management systems has become one of the important directions of current research and development. At present, health management systems for individuals or families are still relatively rare on the market, and most of them can only collect and display vital signs data, and cannot further provide health advice based on vital signs data. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide an intelligent vital signs monitoring system based on an improved resNet34, which can accurately monitor the user's physiological signs, conduct health assessment for the user, and promptly discover potential health problems.

[0004] To achieve the above-mentioned purpose of the invention, the present invention provides an intelligent vital signs monitoring system based on improved resNet34, the system comprising a hardware part and a software part, the hardware part comprising an ECG vital signs detection module, the ECG vital signs detection module being used to detect the user's physiological signs data, the software part being connected to the hardware part data, the software part being used to obtain the user's physiological signs data, running an artificial intelligence model to process the user's physiological signs data, identifying abnormal physiological signs events, and issuing an alarm when an abnormal physiological signs event is identified.

[0005] Furthermore, the ECG vital sign detection module includes a digital ECG sensor and ECG electrodes, the ECG electrodes are connected to the digital ECG sensor, and the digital ECG sensor is connected to the software part signal.

[0006] Furthermore, the user's physiological sign data includes electrocardiogram (ECG) signals, pulse wave (PPG) signals, and electromyography (EMG) signals.

[0007] Furthermore, the software part includes:

[0008] A data receiving module, used to receive the user's physiological sign data sent by the ECG vital sign detection module;

[0009] A model detection module is used to run a pre-trained first deep learning model, input the user's physiological sign data into the first deep learning model for processing, and obtain a heart rate detection result;

[0010] A visualization module is used to automatically generate a corresponding electrocardiogram and visualize the electrocardiogram when new physiological sign data of the user is monitored, or to visualize the heart rate detection result when a new heart rate detection result is monitored;

[0011] The data storage module is used to realize local storage and cloud storage of user physiological signs data and heart rate detection results, data backup and regular cleaning.

[0012] Furthermore, the software part also includes:

[0013] A personalized health assessment module, which is used to run a pre-trained second deep learning model, input the user's physiological sign data and historical health assessment results within a specified period into the second deep learning model, obtain a health assessment result, fill in an assessment report template based on the health assessment result, and automatically generate a health assessment report, wherein the health assessment result includes a current health status, potential risk factors, and health recommendations;

[0014] The data storage module is also used to realize local storage and cloud storage of health assessment results and health assessment reports.

[0015] Furthermore, the software part also includes:

[0016] The sharing and exporting module is used to share the user's physiological signs data and health assessment report with other users according to user instructions, and to export the user's physiological signs data and / or health assessment report.

[0017] Furthermore, the software part also includes:

[0018] Desensitization module, used to desensitize the user's physiological sign data or health assessment report;

[0019] The encryption module is used to encrypt the user's physiological signs data or health assessment report.

[0020] Furthermore, the user's physiological sign data or health assessment report is encrypted, specifically including the following operations:

[0021] Obtaining a curve graph drawn based on the physiological sign data to be encrypted, and calculating a median value of the curvature change of the curve graph;

[0022] The median value of the curvature change is subjected to noise processing to obtain noise-contaminated data;

[0023] The noise contaminated data is used as a salt value of a key derivation function, and the key derivation function is executed to obtain an encryption key;

[0024] The user's physiological sign data or health assessment report is encrypted using an encryption key.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] The present invention provides an intelligent vital sign monitoring system based on improved resNet34. The hardware part monitors the user's physiological sign data in real time through the ECG vital sign detection module, providing reliable data support for subsequent health analysis and diagnosis. The software part processes the user's physiological sign data by adopting an artificial intelligence model, can quickly and accurately identify abnormal data, and issue an alarm, thereby timely discovering the user's potential health risks, and can also generate a health assessment report to provide data support and effective suggestions for the user to improve his health status. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 It is a schematic diagram of the overall structure of an intelligent vital signs monitoring system based on improved resNet34 provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The enumerated embodiments are only used to explain the present invention but not to limit the scope of the present invention.

[0030] Reference Figure 1 This embodiment provides an intelligent vital sign monitoring system based on an improved resNet34, the system comprising a hardware part and a software part, the hardware part comprising an ECG vital sign detection module, the ECG vital sign detection module being used to detect the user's physiological sign data, the software part being connected to the hardware part data, the software part being used to obtain the user's physiological sign data, running an artificial intelligence model to process the user's physiological sign data, identifying abnormal physiological sign events, and issuing an alarm when an abnormal physiological sign event is identified.

[0031] As an optional implementation, the ECG vital sign detection module includes a digital ECG sensor and ECG electrodes, the ECG electrodes are connected to the digital ECG sensor, and the digital ECG sensor is connected to the software part signal.

[0032] Exemplarily, the digital ECG sensor may adopt an ADS1292 digital ECG module, which adopts advanced sensing technology and signal processing algorithms and can collect the user's physiological data in real time and accurately. It is understandable that the digital ECG sensor may also adopt other types of sensors, which are not specifically limited in this embodiment.

[0033] In this embodiment, the user's physiological sign data includes but is not limited to electrocardiogram (ECG) signals, pulse wave (PPG) signals, and electromyography (EMG) signals.

[0034] The digital ECG sensor and the software can be connected via wired mode, such as UART, or wirelessly, such as BLE5.0. The system provides both Bluetooth communication protocol and serial communication protocol to ensure stable and efficient data transmission.

[0035] As an optional implementation, the software part includes:

[0036] The data receiving module is used to receive the user's physiological sign data sent by the ECG vital sign detection module.

[0037] The model detection module is used to run a pre-trained first deep learning model, input the user's physiological sign data into the first deep learning model for processing, and obtain a heart rate detection result.

[0038] Exemplarily, the first deep learning model can use the resNet34 deep learning neural network model with Spatial-Attention to quickly diagnose heart diseases such as arrhythmia, myocardial infarction, etc., and provide users with comprehensive health management support through intelligent analysis of multiple human parameters. It is understandable that the first deep learning model can also use other models, which is not specifically limited in this embodiment.

[0039] The visualization module is used to automatically generate a corresponding electrocardiogram and visualize the electrocardiogram when new user physiological sign data is monitored, or visualize the heart rate detection result when a new heart rate detection result is monitored. The system provides a real-time data display function, and the user can view the current physiological sign data at any time. The system provides an intuitive and clear interface, and displays the monitoring data in the form of charts, curves, etc., to help users intuitively understand their health status.

[0040] The data storage module is used to realize local storage and cloud storage of user physiological signs data and heart rate detection results, data backup and regular cleaning. In this embodiment, the user's relevant data can be stored on the local device or on the cloud server, depending on the user's needs and preferences. For local storage, the system provides automated data management functions, including data backup, regular cleaning, etc., to ensure the security and integrity of the data. For cloud storage, the system uses secure encryption measures to protect the privacy and security of user data.

[0041] As a further optional implementation, the software part also includes:

[0042] The personalized health assessment module is used to run a pre-trained second deep learning model, input the user's physiological sign data and historical health assessment results within a specified period into the second deep learning model, obtain health assessment results, fill in the assessment report template based on the health assessment results, and automatically generate a health assessment report, wherein the health assessment results include current health status, potential risk factors and health recommendations.

[0043] Exemplarily, the second deep learning model can be implemented using network models such as convolutional neural networks, recurrent neural networks, and long short-term memory networks, which is not specifically limited in this embodiment.

[0044] The data storage module is also used to realize local storage and cloud storage of health assessment results and health assessment reports.

[0045] As another optional implementation, the software part also includes:

[0046] The sharing and exporting module is used to share the user's physiological sign data and health assessment report with other users according to the user's instructions, and to export the user's physiological sign data and / or health assessment report. Users can choose to share the monitored physiological data with doctors or family members for better health management and support. At the same time, users can also export the data to common file formats, such as CSV or PDF, for further analysis or archiving in other platforms or software.

[0047] As another optional implementation, the software part also includes:

[0048] The desensitization module is used to perform desensitization processing on the user's physiological sign data or health assessment report. For example, masking, replacement, data perturbation and other methods can be used to desensitize the user data to protect the privacy of the user information.

[0049] The encryption module is used to encrypt the user's physiological signs data or health assessment report.

[0050] In this embodiment, the user's physiological sign data or health assessment report is desensitized or encrypted, and whether to do so can be determined based on the user's needs.

[0051] As another further optional implementation, encrypting the user's physiological sign data or health assessment report specifically includes the following operations:

[0052] S1. Obtain a curve graph drawn based on the physiological sign data to be encrypted, and calculate the median value of the curvature change of the curve graph.

[0053] S2. Noise the median value of the curvature change to obtain noise-contaminated data.

[0054] S3. Use the noise contaminated data as the salt value of the key derivation function and execute the key derivation function to obtain the encryption key.

[0055] S4. Encrypt the user's physiological sign data or health assessment report using an encryption key.

[0056] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent vital signs monitoring system based on improved resNet34, characterized in that: The system includes a hardware part and a software part. The hardware part includes an ECG vital sign detection module, which is used to detect the user's physiological sign data. The software part is connected to the hardware part data, and the software part is used to obtain the user's physiological sign data, run the artificial intelligence model to process the user's physiological sign data, identify abnormal physiological sign events, and issue an alarm when an abnormal physiological sign event is identified.

2. According to claim 1, an intelligent vital signs monitoring system based on improved resNet34 is characterized in that: The ECG vital sign detection module includes a digital ECG sensor and an ECG electrode sheet, wherein the ECG electrode sheet is connected to the digital ECG sensor, and the digital ECG sensor is connected to a software part signal.

3. According to claim 1, an intelligent vital signs monitoring system based on improved resNet34 is characterized in that: The user's physiological sign data includes electrocardiogram (ECG) signals, pulse wave (PPG) signals, and electromyography (EMG) signals.

4. According to claim 1, an intelligent vital signs monitoring system based on improved resNet34 is characterized in that: The software part includes: A data receiving module, used to receive the user's physiological sign data sent by the ECG vital sign detection module; A model detection module is used to run a pre-trained first deep learning model, input the user's physiological sign data into the first deep learning model for processing, and obtain a heart rate detection result; A visualization module is used to automatically generate a corresponding electrocardiogram and visualize the electrocardiogram when new physiological sign data of the user is monitored, or to visualize the heart rate detection result when a new heart rate detection result is monitored; The data storage module is used to realize local storage and cloud storage of user physiological signs data and heart rate detection results, data backup and regular cleaning.

5. The intelligent vital signs monitoring system based on improved resNet34 according to claim 4 is characterized in that: The software part also includes: A personalized health assessment module, which is used to run a pre-trained second deep learning model, input the user's physiological sign data and historical health assessment results within a specified period into the second deep learning model, obtain a health assessment result, fill in an assessment report template based on the health assessment result, and automatically generate a health assessment report, wherein the health assessment result includes a current health status, potential risk factors, and health recommendations; The data storage module is also used to realize local storage and cloud storage of health assessment results and health assessment reports.

6. The intelligent vital signs monitoring system based on improved resNet34 according to claim 4, characterized in that: The software part also includes: The sharing and exporting module is used to share the user's physiological signs data and health assessment report with other users according to user instructions, and to export the user's physiological signs data and / or health assessment report.

7. The intelligent vital signs monitoring system based on improved resNet34 according to claim 6, characterized in that: The software part also includes: Desensitization module, used to desensitize the user's physiological sign data or health assessment report; The encryption module is used to encrypt the user's physiological signs data or health assessment report.

8. The intelligent vital signs monitoring system based on improved resNet34 according to claim 7, characterized in that: Encrypt the user's physiological sign data or health assessment report, including the following operations: Obtaining a curve graph drawn based on the physiological sign data to be encrypted, and calculating a median value of the curvature change of the curve graph; The median value of the curvature change is subjected to noise processing to obtain noise-contaminated data; The noise contaminated data is used as a salt value of a key derivation function, and the key derivation function is executed to obtain an encryption key; The user's physiological sign data or health assessment report is encrypted using an encryption key.

Images