Cerebral apoplexy risk prediction method and device based on hybrid deep transfer learning

Abstract

The invention belongs to the technical field of computers, and provides a cerebral apoplexy risk prediction method and device based on hybrid deep transfer learning. According to the invention, healthmonitoring data of a user is obtained and subjected to normalization processing to obtain prediction data for cerebral apoplexy risk prediction; the prediction data is input into a final cerebral apoplexy risk prediction model to obtain a cerebral apoplexy risk prediction result; the final cerebral apoplexy risk prediction model is obtained by training a primary prediction model with training samples obtained through instance migration and active learning, wherein the primary prediction model achieves network weight migration through migration learning; whether a user has a cerebral apoplexyrisk or not is determined according to the cerebral apoplexy risk prediction result; and therefore, the problems of unbalanced data distribution in a cerebral apoplexy risk prediction model training set, difficult acquisition of positive sample data with cerebral apoplexy and easy leakage of user privacy are solved, the problem of a low cerebral apoplexy data sample amount is also solved, and theaccuracy of the cerebral apoplexy risk prediction result is further improved.

Description

technical field [0001] The invention belongs to the field of computer technology, and in particular relates to a stroke risk prediction method and device based on hybrid deep transfer learning. Background technique [0002] Stroke, also known as cerebral apoplexy or cerebrovascular disease, is caused by the sudden rupture and bleeding of blood vessels in the human brain or cerebral ischemia and hypoxia caused by blood vessel blockage. According to the survey, stroke has become the primary cause of disability in Chinese adults The reason is that stroke has the characteristics of high morbidity, high mortality and high disability rate, and the one hour when the onset of stroke patients becomes the golden hour for the treatment of stroke patients, if the treatment is carried out within one hour of the onset , can effectively reduce the sequelae of patients after recovery. However, there is currently a problem that it is impossible to predict that a normal user may have a strok...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a stroke risk prediction method and device based on hybrid deep transfer learning, which aims to solve the problems of the existing stroke risk prediction model due to the unbalanced distribution of stroke sample data in the training set and the small number of stroke data samples. , The problem of low accuracy of the prediction results of the risk prediction model due to the difficulty in obtaining stroke sample data

Images