Focus classification system based on deep learning and probability imaging omics

Abstract

The invention relates to a focus classification system based on deep learning and probability imaging omics, and belongs to the technical field of medical image classification. The objective of the invention is to solve problems of ambiguity caused by classification ambiguity and low classification precision of an existing lesion classification system. According to the method, a deep convolutionalneural network is used as a main stem, a non-local shape analysis module is proposed to extract feature cloud of a focus on a medical image, interference of pixels around the focus on classificationjudgment is removed, and essential representation of the focus is obtained; meanwhile, the fuzziness of the label is captured; a fuzzy prior network is provided to simulate fuzzy distribution of different expert labels; ambiguity of expert annotation is displayed and modeled; the classification result of model training has better robustness, the fuzzy prior sample is combined with focus representation, a new focus classification system is constructed and can achieve controllability and probability; compared with a traditional convolutional neural network, a classification fuzziness problem isbetter solved, and higher classification precision can be acquired.

Description

technical field [0001] The invention relates to the classification technology of CT images of lesions, in particular to a lesion classification system based on deep learning and probabilistic radiomics, and belongs to the technical field of medical image classification. Background technique [0002] At present, the morbidity and mortality of various cancers in China rank first among all kinds of diseases, among which the morbidity and mortality of lung cancer rank first among various malignant tumors, posing a huge threat to people's health. Pulmonary nodule screening is an important means to achieve early diagnosis and treatment of lung cancer. The radiomics analysis method uses doctors to manually outline regions of interest, image processing, feature extraction, and feature screening, and then combines machine learning algorithms to predict target variables and assist doctors to analyze lesions. It has powerful functions and has been widely used. Compared with traditiona...

AI Technical Summary

Problems solved by technology

However, first of all, the method adopted by this patent is based on a 2D convolutional neural network, which makes this patent not capable of capturing three-dimensional spatial features

Secondly, the patent uses traditional black-box deep learning, which makes the patented technology not controllable and transparent

There is also the problem of ambiguity caused by classification fuzziness.

[0010] Therefore, it is urgent to propose a new classification system to solve the problem of ambiguity and insufficient classification accuracy caused by the ambiguity of classification of lesions in the existing classification technology.

Images