Digital pathological section target area identification method and system, equipment and medium

Abstract

The embodiment of the invention discloses a digital pathological section target area identification method. The method comprises the following steps: acquiring a pathological section scanning image; inputting the pathological section scanning image into a preset recognition model based on deep learning; extracting contour features of the pathological section scanning image through the image contour feature extraction sub-model to obtain a contour image; segmenting the contour image through the image segmentation sub-model to obtain a plurality of sub-contour images; performing classification identification on the plurality of sub-contour images through an image classification sub-model to obtain a region category corresponding to each sub-contour image; the target area image is determinedaccording to the area category of each sub-contour image, so that the scanning efficiency and the scanning quality of the pathological section are improved, the accuracy of the target area image is improved, and the memory space of the digital pathological section is reduced. In addition, the invention also provides a digital pathological section target area identification system, equipment and amedium.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a method, system, device and medium for identifying a target region of a digital pathological slice. Background technique [0002] Pathological diagnosis refers to the observation of pathological sections through a microscope to diagnose diseases. Digital pathology refers to high-resolution digital images obtained by scanning and collecting digital pathological slice scanners, and high-precision multi-view seamless stitching and processing to obtain high-resolution digital pathological slices. [0003] However, the scanning methods of existing scanners to scan digital pathological slices have the following disadvantages: one is to select the area to be scanned by artificial naked eyes, which consumes manpower and material resources; Many blank areas and irrelevant areas will be scanned, thereby reducing the work efficiency and scanning quality of the entire scanning pr...

AI Technical Summary

Problems solved by technology

[0003] However, the scanning methods of existing scanners to scan digital pathological slices have the following disadvantages: one is to select the area to be scanned by artificial naked eyes, which consumes manpower and material resources; Many blank areas and irrelevant areas will be scanned, thereby reducing the work efficiency and scanning quality of the entire scanning process, and digital pathology slides take up a lot of space

Images