Search

CN-121987151-A - Real-time anesthesia depth monitoring system based on deep learning

CN121987151ACN 121987151 ACN121987151 ACN 121987151ACN-121987151-A

Abstract

The invention relates to the technical field of anesthesia depth monitoring, in particular to a real-time anesthesia depth monitoring system based on deep learning, which comprises a data collection module, a data processing module, a cloud module and an anesthesia prediction and response module. In the invention, firstly, a data collection module collects real-time physiological data and historical physiological data and sends the data to a data processing module, the data processing module performs feature extraction on the data, in addition, the historical physiological data and the real-time physiological data are respectively sent to a cloud module for model training and anesthesia prediction and response module for prediction of anesthesia depth, the anesthesia prediction and response module sends a predicted result and the real-time physiological data to the cloud module, and the cloud module updates the model by taking the predicted result and the real-time physiological data as the historical physiological data.

Inventors

  • SHEN WEI
  • ZHANG LIN
  • HE JINFENG

Assignees

  • 南通大学

Dates

Publication Date
20260508
Application Date
20260226

Claims (9)

  1. 1. The real-time anesthesia depth monitoring system based on deep learning is characterized by comprising a data collection module (100), a data processing module (200), a cloud module (300) and an anesthesia prediction and response module (400), wherein: The data collection module (100) is used for collecting real-time physiological data and historical physiological data and sending the real-time physiological data and the historical physiological data to the data processing module (200); The data processing module (200) performs feature extraction on the real-time physiological data, performs feature and label extraction on the historical physiological data, and sends the processed real-time physiological data and the processed historical physiological data to the anesthesia prediction and response module (400) and the cloud module (300) respectively; The cloud module (300) receives the historical physiological data sent by the data processing module (200), performs model training according to the historical physiological data by using a deep learning algorithm, and sends the trained model to the anesthesia prediction and response module (400); The anesthesia prediction and response module (400) predicts the anesthesia depth of the real-time physiological data sent by the data processing module (200) according to the model trained by the cloud module (300), and sends the prediction result and the real-time physiological data to the cloud module (300) as historical physiological data for model updating.
  2. 2. The depth learning based real-time anesthesia depth monitoring system according to claim 1, wherein the data collection module (100) comprises a real-time data collection unit (101) and an analog-to-digital conversion unit (102), the data collection module (100) collects real-time physiological data generated when a patient is anesthetized by using a sensor and transmits the data to the analog-to-digital conversion unit (102), and the analog-to-digital conversion unit (102) is used for converting the received analog signal data into digital signal data and transmitting the data to the data cleaning unit (201) in the data processing module (200).
  3. 3. The depth learning based real-time anesthesia depth monitoring system of claim 1 wherein the data collection module (100) includes a historical data collection unit (103), the historical data collection unit (103) collects historical physiological data using a hospital database and sends the data to a data cleansing unit (201) in the data processing module (200).
  4. 4. The deep learning-based real-time anesthesia depth monitoring system according to claim 1, wherein the data processing module (200) comprises a data cleaning unit (201), a feature and tag processing unit (202) and a training and prediction classification unit (203), the data cleaning unit (201) is used for receiving real-time physiological data and historical physiological data respectively transmitted by the analog-digital conversion unit (102) and the historical data collection unit (103), processing missing values and repeated values in the data and transmitting the data to the feature and tag processing unit (202), the feature and tag processing unit (202) is used for performing feature extraction on the real-time physiological data, performing feature and tag extraction on the historical physiological data and transmitting the processed data to the training and prediction classification unit (203), and the training and prediction classification unit (203) is used for distinguishing the real-time physiological data and the historical physiological data, and transmitting the historical physiological data to the training data receiving unit (301) in the cloud module (300) according to whether the tag column is the predicted data receiving unit (401) in the empty anesthesia prediction and response module (400).
  5. 5. The deep learning-based real-time anesthesia depth monitoring system according to claim 1, wherein the cloud module (300) comprises a training data receiving unit (301), a model updating unit (302) and a parameter updating unit (303), the training data receiving unit (301) receives the historical physiological data sent by the training and prediction classifying unit (203) and is used for carrying out standardization processing on the data and sending the processed data to the model updating unit (302), the model updating unit (302) carries out model training according to the historical physiological data sent by the training data receiving unit (301) by using a deep learning algorithm and sends the trained model and parameters to the parameter updating unit (303), and the parameter updating unit (303) is used for sending the trained model and parameters to the prediction unit (402) in the anesthesia prediction and response module (400).
  6. 6. The depth learning-based real-time anesthesia depth monitoring system according to claim 1, wherein the anesthesia prediction and response module (400) comprises a prediction data receiving unit (401), a prediction unit (402) and a response unit (403), the prediction data receiving unit (401) receives real-time physiological data sent by the training and prediction classifying unit (203) and is used for carrying out standardization processing on the data and sending the processed data to the prediction unit (402), the prediction unit (402) carries out anesthesia depth prediction on the real-time physiological data sent by the prediction data receiving unit (401) according to a model sent by the parameter updating unit (303), the prediction result and the real-time physiological data are sent to the training data receiving unit (301), the prediction result is sent to the response unit (403), and the response unit (403) carries out visual output on the prediction result sent by the prediction unit (402).
  7. 7. The deep learning-based real-time anesthesia depth monitoring system according to claim 5, wherein the training data receiving unit (301) receives the prediction result and the real-time physiological data sent by the prediction unit (402), and is configured to use the prediction result and the real-time physiological data as historical physiological data, and send the data to the model updating unit (302) for model updating.
  8. 8. The deep learning-based real-time anesthesia depth monitoring system according to claim 4, wherein the training data receiving unit (301) converts tag data in the historical physiological data into digital type data and converts tag column values into corresponding integer sizes.
  9. 9. The deep learning-based real-time anesthesia depth monitoring system of claim 5 wherein the model updating unit (302) performs model training based on historical physiological data using a deep learning algorithm, specifically comprising: Forward propagation, the neural network passing input data from the input layer to the output layer; After forward propagation, comparing a predicted result obtained by the neural network with a corresponding label value to calculate a value of a loss function; And back propagation, namely reversely transmitting the error into the network by using a loss function, calculating the contribution of each parameter to the loss, reversely calculating the gradient from the output layer to the input layer by using a chain rule, updating the value of each parameter according to the direction of the gradient, and finally updating the parameters in the neural network according to the gradient information obtained by calculation.

Description

Real-time anesthesia depth monitoring system based on deep learning Technical Field The invention relates to the technical field of anesthesia depth monitoring, in particular to a real-time anesthesia depth monitoring system based on deep learning. Background Implementation principle of real-time anesthesia monitoring systems generally evaluate the anesthesia depth and status of a patient based on experience and statistical analysis, use predefined algorithms or models, and make decisions based on features and indicators, however, conventional real-time anesthesia monitoring systems often have drawbacks. On the one hand, conventional real-time anesthesia monitoring systems typically use predefined criteria to determine the patient's anesthesia depth and status, which rely on subjective choices by an expert, which may lead to human error, affecting the accuracy of the monitoring system; On the other hand, when the traditional real-time anesthesia monitoring system updates the predefined standard or model, a great deal of human resources are needed to analyze and calculate, and the required algorithm model cannot be updated in real time. Disclosure of Invention The invention aims to provide a real-time anesthesia depth monitoring system based on deep learning so as to solve the problems in the background technology. In order to achieve the above purpose, the invention provides the following technical scheme that the real-time anesthesia depth monitoring system based on deep learning comprises a data collection module, a data processing module, a cloud module and an anesthesia prediction and response module, wherein: The data collection module is used for collecting real-time physiological data and historical physiological data and sending the real-time physiological data and the historical physiological data to the data processing module; the data processing module performs feature extraction on the real-time physiological data, performs feature and label extraction on the historical physiological data, and sends the processed real-time physiological data and the processed historical physiological data to the anesthesia prediction and response module and the cloud module respectively; The cloud module receives the historical physiological data sent by the data processing module, performs model training according to the historical physiological data by using a deep learning algorithm, and sends the trained model to the anesthesia prediction and response module; The anesthesia prediction and response module predicts the anesthesia depth of the real-time physiological data sent by the data processing module according to the model trained by the cloud module, and sends the predicted result and the real-time physiological data to the cloud module to serve as historical physiological data to update the model. As a further improvement of the technical scheme, the data collection module comprises a real-time data collection unit and an analog-to-digital conversion unit, wherein the data collection module collects real-time physiological data generated during anesthesia of a patient by using a sensor and sends the data to the analog-to-digital conversion unit, and the analog-to-digital conversion unit is used for converting received analog signal data into digital signal data and sending the data to the data cleaning unit in the data processing module. As a further improvement of the technical scheme, the data collection module comprises a historical data collection unit which collects historical physiological data by using a hospital database and sends the data to a data cleaning unit in the data processing module. The data processing module comprises a data cleaning unit, a characteristic and tag processing unit and a training and predicting classifying unit, wherein the data cleaning unit is used for receiving real-time physiological data and historical physiological data which are respectively transmitted by the analog-to-digital conversion unit and the historical data collecting unit, processing missing values and repeated values in the data and transmitting the data to the characteristic and tag processing unit, the characteristic and tag processing unit is used for extracting the characteristic of the real-time physiological data, extracting the characteristic and tag of the historical physiological data and transmitting the processed data to the training and predicting classifying unit, and the training and predicting classifying unit is used for distinguishing the real-time physiological data and the historical physiological data, transmitting the real-time physiological data to the predicted data receiving unit in the anesthesia predicting and responding module according to whether the tag column is empty or not, and transmitting the historical physiological data to the training data receiving unit in the cloud module. The cloud module comprises a training data receiving unit, a model updating unit and a par