Lesion localization core data extraction method, system, electronic equipment and storage medium

Abstract

The invention discloses a lesion location core data extraction method, system, electronic equipment and storage medium. For any image in a medical image data set, the information entropy, contrast value, and inception score value of the image are calculated and fused to calculate the image. The core degree; all the images in the medical image dataset are sorted in descending order according to the core degree, and the top k images with the core degree are extracted as the core data. Using the previous batch of core data and non-pathological medical data, optimize the information entropy; repeat the above process to extract an appropriate number of core data. While continuously extracting the core data, the present invention continuously optimizes its extraction mechanism, so that the extraction performance of the present invention is continuously improved. Experiments have proved that the present invention has high practicability, can reduce the burden of a large amount of data labeling, and obtain an excellent lesion localization model through training, which can effectively assist doctors in diagnosis and reduce the rate of misdiagnosis.

Description

technical field [0001] The invention relates to the field of smart medical care, in particular to a method, system, electronic equipment and storage medium for extracting core data of lesion location based on active learning. Background technique [0002] In recent years, artificial intelligence has become increasingly mature in theory and technology, bringing a lot of convenience to human daily life, among which the development of intelligent medical care is very rapid, such as the algorithm based on deep learning proposed by Google, which can explain the signs of diabetic retinopathy; Ni uses deep learning to achieve high accuracy in abdominal organ segmentation and so on. These deep learning technologies can assist doctors in identifying diseases, greatly reduce the burden on doctors, and help doctors make more accurate diagnoses. These studies reflect the effectiveness of deep learning technology in medical image analysis, but most of the current research on intelligent...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to provide a lesion localization core data extraction method, system, electronic equipment and storage medium for the deficiencies of the existing technologies, and to extract core data from medical image data without any lesion labeling information to solve the problem of intelligent In medical treatment, it is difficult to locate lesions due to a large amount of unlabeled and variable quality data

Images