Research on Key Technologies of Multi-modal Heterogeneous Medical Big Data Based on Deep Learning

Abstract:
With the continuous deepening of social and economic development and medical reform, as well as the rapid development of some new technologies such as artificial intelligence, big data, and cloud computing, intelligent information technology combining medical and industrial has become a necessary trend. Among them, multi-source heterogeneous data fusion based on machine learning theory and supported by perceptual data has become a current hot research field, which is widely used in urban systems such as medical care, homes, and transportation. Massive perception data are generated in different fields or industries, and each type of data contains a large amount of unlabeled data, data sparse areas and domain knowledge, and there are big differences in these data types, data relationships and data quality. How to fully mine The information of single source data, and the fusion of these single sources of information, the establishment of a big data model based on these fusion information, and finally the establishment of an intelligent auxiliary diagnosis system, has become one of the important research contents. This article observes the patient's diagnosis and treatment process in the hospital, combines deep learning network and multi-modal data fusion technology, and uses deep semantic matching network to study the establishment of a hospital intelligent diagnosis and treatment auxiliary system, which provides a reference for medical big data detection and diagnosis and treatment.