Ultrasonic thyroid nodule benign and malignant prediction method based on deep learning

Abstract

The invention particularly relates to an ultrasonic thyroid nodule benign and malignant prediction method based on deep learning. Comprising the following steps: S1, collecting an ultrasonic nodule image sample, and preprocessing the ultrasonic nodule image sample, wherein the preprocessing comprises image sample quality improvement and image sample data labeling; s2, training a deep neural network based on the image sample annotation data; s3, applying the trained deep neural network to a real-time ultrasonic image to obtain a segmented nodule region image; s4, based on the segmented nodule region image, marking nodule types through a nodule judgment database, and taking the nodule types as training samples of a nodule prediction model; s5, adopting a feedforward neural network, using labeled sample data as a training set network, calculating accuracy and recall rate through feature selection, carrying out learning training through repeated iteration, and selecting a network model with the best evaluation result for output; and S6, predicting the type of the thyroid nodule based on the trained network model.

Description

technical field [0001] The present invention relates to the field of smart medical care, in particular to a method for predicting benign and malignant ultrasonic thyroid nodules based on deep learning. Background technique [0002] During the clinical treatment of thyroid nodules, the distinction between benign and malignant thyroid nodules is the basis for the diagnosis and treatment of thyroid nodules. At present, puncture examination and pathological examination are the main means to differentiate benign and malignant thyroid nodules. However, puncture examination and pathological examination have damage. Sex, causing damage to the normal thyroid tissue of the patient. Therefore, how to effectively use the results of non-invasive ultrasonography to predict benign and malignant thyroid nodules is of great significance for the diagnosis and treatment of thyroid nodules. [0003] In order to help doctors improve the accuracy of clinical diagnosis, avoid unnecessary examinat...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above problems, the purpose of the present invention is to provide a method for predicting benign and malignant ultrasound thyroid nodules based on deep learning for the problems of unsatisfactory effect of traditional image segmentation algorithms and low accuracy of computer-aided diagnosis in the prior art.

Images