Psoriasis tissue pathological section segmentation method and system based on deep learning

Abstract

The embodiment of the invention discloses a psoriasis tissue pathological section segmentation method based on deep learning, and the method comprises the steps: obtaining a plurality of psoriasis tissue pathological images under a microscope with a preset size, and carrying out the preprocessing of the psoriasis tissue pathological images; initializing a pre-established deep neural network; inputting the image into a deep neural network for calculation to obtain a segmented image; comparing the segmented image output by the deep neural network with a pre-labeled label to obtain an error value; judging the number of times of comparing the output result of the deep neural network with a pre-labeled label, and if the preset number of times is not reached, updating the parameter according tothe error value; and if the number of times reaches the preset number of times, splicing the segmented images to generate a target image, and completing image segmentation. A deep learning algorithm is adopted to segment a psoriasis tissue pathology image, psoriasis tissue pathology characteristics are deeply analyzed, and therefore, psoriasis tissue pathology can be analyzed more accurately underthe condition that segmentation is ensured more accurately.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of image segmentation, and in particular to a method and system for segmenting psoriasis histopathological slices based on deep learning. Background technique [0002] In the field of computer vision, the purpose of image segmentation is to simplify or change the representation of the image, making the image easier to understand and analyze. Since the 1970s, it has been studied by people in related fields, and thousands of algorithms have been developed to realize image segmentation. The commonly used algorithms are roughly divided into the following algorithms, namely: threshold-based segmentation methods, region-based segmentation methods, edge-based segmentation methods, and segmentation methods based on specific theories. Coupled with the introduction of computer hardware performance and the popularization of the concept of deep learning in recent years, image segmentation technolo...

AI Technical Summary

Problems solved by technology

[0004] To this end, the embodiment of the present invention provides a psoriasis histopathological slice segmentation method based on deep learning to solve the problem of accurate segmentation and identification due to the wide variety of psoriasis-like skin diseases and complex tissue structures in the prior art. low rate problem

Images