Medical data training method and device and storage medium

Abstract

The invention relates to a medical data training method and device, and a storage medium. The training method comprises the steps: combining original medical data with standard information; inputting the classification algorithm model to obtain a deduction result; selecting to-be-selected training data through screening based on a deduction result; and generating medical data for retraining in combination with a labeling result of the to-be-selected training data. Or the training method comprises: inputting labeled medical data into the classification algorithm model to obtain a deduction result, and the labeled medical data comprising original information and an original labeling result contrasted with the standard information table; selecting to-be-selected training data through screening based on a deduction result; and generating medical data for retraining in combination with a labeling result of the to-be-selected training data. According to the embodiments of the invention, annotation resources can be saved, the performance can be improved, and the utilization rate of training data can be improved.

Description

technical field [0001] The present disclosure relates to the technical field of medical data intelligent processing, and in particular to a medical data training method, a processing device for medical data training, and a computer-readable storage medium. Background technique [0002] For the training of medical data in the prior art, in the deep learning algorithm, one of the main means to improve the performance is to increase the amount of training data, but there are two problems: 1) increasing the random data, but the amount is small, still not enough to improve the performance of the algorithm ; 2) The amount of data is greatly increased, but the training time and labeling resources are also greatly increased. Contents of the invention [0003] The present disclosure intends to provide a medical data training method, a processing device for medical data training, and a computer-readable storage medium, which can save labeling resources and improve performance, and i...

AI Technical Summary

Problems solved by technology

[0002] For the training of medical data in the prior art, in the deep learning algorithm, one of the main means to improve the performance is to increase the amount of training data, but there are two problems: 1) increasing the random data, but the amount is small, still not enough to improve the performance of the algorithm ; 2) The amount of data is greatly increased, but the training time and labeling resources are also greatly increased

Images