Intelligent prognosis method and system for chronic disease patient based on recurrent neural network

Abstract

The invention discloses a chronic disease patient intelligent prognosis method and system based on a recurrent neural network. The method comprises the following steps: S100, acquiring and processing medical data of a patient; S200, based on the processed medical data and the recurrent neural network model or the gating recurrent unit, establishing a prediction model, and predicting the death risk of the patient at the corresponding moment through the prediction model. The prediction model is established based on the recurrent neural network model, the variant of the recurrent neural network model and medical data fusing static information of the patient and dynamic information changing along with time, the death risk of the patient at the corresponding moment is predicted, and clinical doctors are assisted to make decisions.

Description

technical field [0001] The invention relates to the field of intelligent medical treatment, in particular to an intelligent prognosis method and system for patients with chronic diseases based on a cyclic neural network. Background technique [0002] Peritoneal dialysis is an important alternative treatment for end-stage renal disease. Patients undergoing peritoneal dialysis have complex conditions. In order to better treat and manage such patients, it is necessary to comprehensively evaluate the condition according to various factors during the follow-up process. and develop an individualized treatment plan accordingly. [0003] At present, there have been attempts to apply data mining and machine learning technology, especially deep learning, to medical data analysis, among which recurrent neural network (RNN) and its variant gated recurrent unit (GRU) can flexibly process variable-length time-series data, It is very suitable for analyzing the data of peritoneal dialysis ...

AI Technical Summary

Problems solved by technology

[0003] At present, there have been attempts to apply data mining and machine learning technology, especially deep learning, to medical data analysis. Among them, recurrent neural network (RNN) and its variant gated recurrent unit (GRU) can flexibly process variable-length time series data. It is very suitable for analyzing the data of peritoneal dialysis patients, and can dig out the nonlinear relationship and hidden patterns, and there is no invention of peritoneal dialysis intelligent prognosis method and system based on cyclic neural network fusion of static and dynamic information.

Images