Medical image feature recognition prediction model

Abstract

The invention discloses a medical image feature recognition prediction model which comprises the following steps: constructing a multi-tissue segmentation model based on a deep convolutional neural network, obtaining data from a sample library, extracting image features, obtaining an image segmentation result based on pyramid up-sampling, and segmenting different tissue regions; constructing a cell detection model based on the deep convolutional neural network, firstly magnifying times to take blocks, and then standardizing images of the taken small blocks; sending the cells into a regression detection module to detect the image in each small block, and then cascading the detected cells with a deep classification network to obtain an interested target; and constructing a visual sub-vision module, and selecting the features with the most predictive capability from the features in all different tissue areas and interested objects by using a method of combining a feature selection method and cross validation. According to the invention, multi-tissue segmentation is carried out on the pathological image, cells are accurately identified, and a doctor is assisted in reading through sub-visual features.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a medical image feature recognition prediction model. Background technique [0002] In the past ten years, with the people's increasing material living standards and the accelerating pace of life, the incidence of malignant tumors in the population has been increasing year by year. Traditional methods of cancer diagnosis and analysis usually rely on pathologists to manually measure and characterize several indicators in cancer pathological image samples, including tumor staging, glandular differentiation grade of adenocarcinoma, and some immunohistochemical-based Antigenated molecular markers, such as estrogen receptor and HER2 for breast cancer patients, and specific antigen CaP for prostate cancer, etc. According to doctors with many years of medical experience, there are several key problems in the traditional medical diagnosis and analysis process. Firs...

AI Technical Summary

Problems solved by technology

[0005] In addition, the simple identification of tissues and cells is not enough to support and assist clinicians in qualitative and quantitative analysis of patient survival time and survival risk, and thus cannot formulate accurate and effective treatment strategies for patients.

Images