Panoramic health dynamic monitoring and prediction system based on time sequence knowledge graph

Abstract

The invention provides a panoramic health dynamic monitoring and prediction system based on a time sequence knowledge graph, and relates to the technical field of knowledge graphs. According to the invention, the requirement of health data collection can be met, the health data can be displayed in a panoramic real-time manner, and the requirement of health data analysis in dynamic health monitoring can be greatly met. Meanwhile, a health time sequence knowledge graph link prediction model based on incremental LSTM is adopted. The model is used for creating a health timing sequence knowledge graph with timing sequence characteristics on the basis of health dynamic monitoring data records, adding up-and-down associated information in the graph by adopting an LSTM recurrent neural network, carrying out serialized learning, then carrying out incremental calculation on timing sequence information, and extracting more accurate feature vectors from the timing sequence information. And finally, deep learning is continuously carried out through incremental computation and an LSTM recurrent neural network, so that the prediction accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of knowledge graphs, in particular to a panoramic health dynamic monitoring and prediction system based on time-series knowledge graphs. Background technique [0002] With the growth of age, the physiological functions of the heart, brain, kidney and other organs decline, the metabolic function is disordered, the immunity is low, and people are prone to various chronic diseases such as hypertension, diabetes, coronary heart disease and tumors. In major health institutions such as community hospitals and nursing homes, health monitoring and early warning management services will be provided for people over 65 years old. [0003] The existing health monitoring and early warning management is to use various health data collection terminals (used to collect heart rate, blood pressure, body temperature, weight and other health index data) in major health institutions such as community hospitals and nursing homes ...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the existing technology, the present invention provides a panoramic health dynamic monitoring and prediction system based on time-series knowledge graphs, which solves the technical problem of low accuracy of the prediction results of the existing health monitoring and early warning management

Images