Shared medical data intelligent auxiliary diagnosis system integrating block chain and federated learning

Abstract

The invention discloses a shared medical data intelligent auxiliary diagnosis system integrating block chain and federated learning. The system comprises an electronic medical record library establishment module used for acquiring medical data of a hospital and establishing an electronic medical record library; a shared sample extraction module used for screening samples with the same pathologicalfeatures from an electronic case library of hospitals participating in medical data sharing to serve as shared samples; a block chain module used for storing the information sent to a central serverby a local model training module in the block chain and compensating the hospitals with large contribution by utilizing a consensus mechanism of the block chain; a local model establishing module usedfor establishing the local model; the local model training module used for performing encryption training of the local model; and an auxiliary diagnosis module used for storing the trained local model so as to carry out auxiliary diagnosis on diseases. The system can solve the island problem of medical data on the premise that privacy and safety are guaranteed, and therefore, good assistance is provided for disease diagnosis.

Description

technical field [0001] This application relates to the fields of big data and medical technology, and specifically relates to an intelligent auxiliary diagnosis system for sharing medical data that integrates blockchain and federated learning. Background technique [0002] Massive data is an essential element for the development of artificial intelligence, and the acquisition and use of data requires two conditions. One is to have a large number of data sources, and the other is how to achieve efficient, safe, reliable, reasonable and legal sharing. [0003] Although our country has a great advantage in the number of hospital cases, due to the lack of sharing of medical data, there is a phenomenon of "data islands", which makes these data cannot be used reasonably, thus hindering the development of medical artificial intelligence. However, it is usually not possible to simply combine the data of all parties. Due to the privacy of medical data, the protection of relevant law...

AI Technical Summary

Problems solved by technology

[0003] Although our country has a great advantage in the number of hospital cases, due to the lack of sharing of medical data, there is a phenomenon of "data islands", which makes these data cannot be used reasonably, thus hindering the development of medical artificial intelligence

However, it is usually not possible to simply combine the data of all parties. Due to the privacy of medical data, the protection of relevant laws and regulations, and the competition among various institutions, it is unrealistic to achieve a simple combination of medical data.

Images