Nodule calcification medical image processing method

Abstract

The invention provides a nodule calcification medical image processing method, and relates to the technical field of medical image processing, and the method comprises the steps: marking a nodule calcification region in each collected nodule calcification medical image to obtain corresponding marked images; training each marked image to obtain an image processing model, wherein the image processing model comprises a residual network module, an anti-stacking module, a group dilated convolutional network module, a stacking module and a decision network module which are connected in sequence; and processing a to-be-processed medical image in the image processing model to obtain a semantic probability heat map of the nodule calcification region in the to-be-processed medical image, and taking the semantic probability heat map as a processing result of the to-be-processed medical image. The method has the advantages that the problems that in the prior art, the nodule image calcification region semantic distinguishing capacity is poor, and calcification region semantic feature extraction is likely to be interfered with by similar backgrounds are solved, and the problems that the nodule calcification medical image semantic probability heat map extraction effect is poor due to the fact that the nodule calcification region is small are solved.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a method for processing medical images of nodule calcification. Background technique [0002] Nodular calcification refers to the necrosis of certain tissues in the human body under the action of some factors, and then calcium salts in the body are deposited in the necrotic focus, making the lesion localized and stable. Accurate identification of nodular calcification can help Physicians are better at making clinical diagnoses. [0003] Most of the traditional calcification point discrimination algorithms are based on the brightness characteristics of medical images, and judge whether it is a calcification point by comparing single or multiple thresholds. Since the calcification point is not necessarily the brightest area in the organ and nearby tissues, there are many other bright tissues near the nodule; in addition, the imaging quality of ultrasound machines ...

AI Technical Summary

Problems solved by technology

Therefore, the existing calcification feature extraction methods limit its applicable scope. At the same time, classic networks such as AlexNet and FCN, as general-purpose networks for feature extraction, have not been designed and optimized for the unique features of nodule calcification medical images.

Images