Medical image annotation system based on homomorphic encryption and data processing method thereof

Abstract

The invention discloses a medical image annotation system based on homomorphic encryption and a data processing method thereof, and the method comprises the steps: receiving encrypted original data sent by a participant through a CPU, and carrying out the primary processing of the encrypted original data; the FPGA receiving the encrypted original data subjected to primary processing by the CPU, and performing operation on the encrypted original data subjected to primary processing so as to obtain an operation result of the encrypted data; the CPU receiving the operation result of the encrypted data and sending the operation result of the encrypted data to a participant, so that the participant decrypts the operation result of the encrypted data to obtain an actual operation result. According to the technical scheme of the invention, the homomorphic encryption can be realized to process the data, and meanwhile, the processing efficiency can be ensured.

Description

technical field [0001] The invention relates to the field of medical technology, in particular to a medical image labeling system based on homomorphic encryption and a data processing method thereof. Background technique [0002] The application of artificial intelligence in the medical field urgently needs to be driven by big data and supported by a large amount of data required by deep learning algorithms to summarize doctors' clinical experience into standardized and refined process rules to achieve accurate judgments on patients' physical information. At present, the research of medical artificial intelligence is mainly divided into two aspects: clinical application and medical research. Among them, medical research uses the advantages of artificial intelligence data extraction and simulation experiments to analyze medical data to help medical personnel conduct medical research and development, specific disease etiology analysis and treatment and other related research; ...

AI Technical Summary

Problems solved by technology

[0004] 1), the current crowdsourcing data management platform lacks professional processing in the medical field, and there is no medical data platform that can clarify the flow of data sets

[0005] 2) The existing medical data processing tools are rather messy. There are separate labeling tools and data desensitization tools (for DICOM format). There is a lack of a platform to centrally view and manage all kinds of medical data to ensure follow-up Reproducibility of relevant results, promoting efficient labeling of medical data

[0006] 3) With the development of medical information technology, some medical institutions collect a large amount of medical data information every year, and build their own medical image labeli

Images