Ultrasonic image identification method and device for papillary thyroid carcinoma

Abstract

The invention discloses an ultrasonic image identification method and an ultrasonic image identification device for papillary thyroid carcinoma. The ultrasonic image identification method comprises the step of identifying features of a papillary thyroid carcinoma ultrasonic image to be identified by adopting a pre-established papillary thyroid carcinoma ultrasonic image identification model. The establishment process of the papillary thyroid carcinoma ultrasonic image identification model comprises the steps of: preprocessing papillary thyroid carcinoma ultrasonic images used for deep learning; normalizing the pre-processed papillary thyroid carcinoma ultrasonic images used for deep learning; establishing a convolutional neural network; and training the papillary thyroid carcinoma ultrasonic images used for deep learning after normalization by utilizing the convolutional neural network, so as to obtain the papillary thyroid carcinoma ultrasonic image identification model. The ultrasonic image identification method and the ultrasonic image identification device can perform accurate identification on the features of the papillary thyroid carcinoma ultrasonic image to be identified in the using process, so as to assist the diagnosis of a doctor, reduce the uncertainty of diagnosis and improve the accuracy of the diagnosis.

Description

technical field [0001] The invention relates to the technical field of ultrasonic image recognition, in particular to an ultrasonic image recognition method and device for thyroid papillary carcinoma. Background technique [0002] In recent years, the prevalence of cancer has been increasing, and it is very important to make an accurate and rapid diagnosis of the patient's lesions. When diagnosing cancer, the more conventional diagnostic method is to conduct a qualitative analysis of the pathological morphology of the pathological sections under a microscope, and this diagnostic method often has a lot of subjectivity, especially when the characteristics of cancer are not very good. When it is obvious or the sampling is small, it will increase the uncertainty of subjective diagnosis. [0003] With the development of digital image processing technology, pattern recognition and artificial intelligence technology, computer-aided diagnosis can minimize the misjudgment of cancer ...

AI Technical Summary

Problems solved by technology

Usually, the corresponding diagnosis is made through the identification of ultrasound images of lesions. However, at present, the identification of ultrasound imag

Images