Deep-network-based tissue segmentation method of panoramic digital colorectum pathology image

Abstract

The invention discloses a deep-network-based tissue segmentation method of a panoramic digital colorectum pathology image. The method comprises the following steps: step one, acquiring a panoramic digital colorectum pathology image; step two, segmenting the panoramic digital colorectum pathology image; step three, establishing a training sample image; step four, extracting deep features of tissuesof different types; step five, determining types of tissues in the segmented images by using a classifier based on the extracted deep tissue features; step six, splicing image classification resultsand determining a tissue class of an overall picture; and step seven, splicing the images according to block coordinates. According to the invention, the panoramic digital colorectum pathology image is segmented and tissue type marking is carried out on all segmented images by using a sliding window and a trained model; and tissue type determination is carried out by using the classifier based onthe extracted deep tissue features to obtain an image classification result. The classification becomes accurate and the classification speed is increased.

Description

technical field [0001] The invention discloses a tissue segmentation method for colorectal panorama digital pathological images based on a deep network, belonging to the field of medical image processing. Background technique [0002] At present, the analysis of pathological images is mainly evaluated by pathologists. However, the manual analysis method is extremely time-consuming and involves the subjective judgment of doctors. There are large differences between doctors with different experiences, which will lead to inappropriate treatment or overtreatment. In poorer and backward areas, many people died due to missing treatment time due to lack of good doctors and medical equipment. [0003] For pathological tissue images, since they contain a lot of valuable information, different histopathological images can be classified by using some characteristics of the pathological image itself. Histopathological images are highly complex and have many targets. Existing research ...

AI Technical Summary

Problems solved by technology

Histopathological images are highly complex and have many targets. Existing research work on colorectal digital pathology images is very little. Generally, only some types of tissue areas in the images are detected, and the images are processed one-sidedly, and the processing results are not accurate. precise

[0004] Existing research on colorectal panoramic digital pathological images has not yet done research on colorectal panoramic digital pathological images, generally only for local area segmentation, for example, Multi-class texture analysis in colorectal cancer histology published in Science report in 2016 is Segmentation of various tissues on a small range of colorectal pathological pictures, the segmentation is rough, the accuracy is poor, and the error rate is high. Only cells or some types of tissue areas in the picture are detected, and the scope of application is small.

Images