Disease data processing method, device and equipment based on federal learning and medium

Abstract

The invention discloses a disease data processing method, device and equipment based on federal learning and a medium. The method comprises the steps that an electronic health records and disease dataof diagnosed patients in a local database are acquired; feature extraction is performed on the electronic health records to obtain disease feature vectors of all the patients; a local training sampleset is constructed according to the disease feature vectors and disease data of all the patients; federal learning of all hospital data terminals is performed based on the local training sample set to obtain a disease prediction model. According to the disease data processing method based on federal learning, data of all the hospital terminals is combined for federal training, the high-quality disease prediction model can be trained on the basis of not disclosing the privacy of the patients, and a positive auxiliary role can be played accordingly in a diagnosis process by a doctor.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a federated learning-based disease data processing method, device, equipment and medium. Background technique [0002] With the development of machine learning and artificial intelligence, the application of machine learning in the field of medical and health is currently a direction of exploration. However, machine learning requires a large amount of data as the basis for learning. The patient data of each hospital is limited. It is difficult to train high-quality machine models only by relying on the patient data of one hospital. If the data of each hospital is combined for training, the privacy of patients in each hospital will be leaked. Therefore, how to realize the data union of various hospitals in a safe and reliable way to obtain high-quality machine models is an urgent problem to be solved. Contents of the invention [0003] The main purpose ...

AI Technical Summary

Problems solved by technology

However, machine learning requires a large amount of data as the basis for learning. The patient data of each hospital is limited. It is difficult to train high-quality machine models only by relying on the patient data of one hospital.

If the data of each hospital is combined for training, the privacy of patients in each hospital will be leaked

Images