Auxiliary diagnosis system for automatic identification of benign and malignant thyroid nodules based on deep convolutional neural network

Abstract

The invention relates to auxiliary medical diagnoses, and aims to provide a method for automatically identifying whether a thyroid nodule is benign or malignant based on a deep convolutional neural network. The method for automatically identifying whether the thyroid nodule is benign or malignant based on the deep convolutional neural network comprises the following steps: reading B ultrasonic data of thyroid nodules; performing preprocessing for thyroid nodule images; selecting images, and obtaining nodule portions and non-nodule portions through segmentations; averagely dividing the extracted ROIs (regions of interest) into p groups, extracting characteristics of the ROIs by utilizing a CNN (convolutional neural network), and performing uniformization; taking p-1 groups of data as a training set, taking the remaining one group to make a test, and obtaining an identification model through training to make the test; and repeating cross validation for p times, and then obtaining an optimum parameter of the identification model. The method can obtain the thyroid nodules through the automatic segmentations by means of the deep convolutional neural network, and makes up for the deficiency that a weak boundary problem cannot be solved based on a movable contour and the like; and the method can automatically lean and extract valuable feature combinations, and prevent the complexity of an artificial feature selection.

Description

technical field [0001] The invention relates to the field of auxiliary medical diagnosis, in particular to a system for automatically identifying benign and malignant thyroid nodules based on a deep convolutional neural network. Background technique [0002] In recent years, with the rapid development of computer technology and digital image processing technology, digital image processing technology has been more and more used in the field of auxiliary medical diagnosis. Accurate, recognition and other image processing technologies to obtain valuable medical diagnosis information, the main purpose is to make the doctor observe the lesion more directly and clearly, and provide auxiliary reference for the doctor's clinical diagnosis, which has very important practical significance. [0003] Thyroid nodules are a common epidemic nowadays. According to surveys, the incidence of thyroid nodules in the population is nearly 50%, but only 4%-8% of thyroid nodules can be palpated in ...

AI Technical Summary

Problems solved by technology

However, the inherent imaging mechanism makes the quality of ultrasonic thyroid tumor images collected clinically poor, which affects the accuracy and automation of auxiliary diagnosis. Therefore, the most current segmentation of thyroid nodules is semi-automatic segmentation based on active contours. The classification is mainly It is to manually select features, and then use SVM, KNN, decision tree, etc. to classify and identify. These classifiers can only have good results for small sample data, but medical data is massive, and the classification and identification of large samples is very important for medical diagnosis. In order to have a better assisting effect

Images