Ultrasonic image thyroid nodule intelligent detection system based on Fast-RCNN

Abstract

The invention discloses an ultrasonic image thyroid nodule intelligent detection system based on a Faster-RCNN. When the adopted Faster-RCNN model is used for carrying out target detection, speed increases and storage space is optimized compared with a regional convolutional neural network RCNN so that rapid and automatic detection of benign and malignant nodules is realized. The features are optimized in the model construction. Key features in an ultrasonic diagnosis standard ACR TI-RADS are calculated before deep learning training, key image feature factors influencing clinical diagnosis aresearched and added to a convolutional neural network feature layer for improvement, a new identification model is constructed, and therefore the accuracy of a prediction result is improved.

Description

technical field [0001] The invention belongs to the field of computer image processing, in particular to an identification system for identifying benign and malignant thyroid ultrasound nodules by using intelligent detection technology. Background technique [0002] Thyroid tumors are common and multiple tumors in the head and neck. Thyroid malignant tumors account for more than 10% of malignant tumors treated, much higher than other head and neck tumors. In recent years, the incidence of thyroid cancer has been increasing year by year, which has attracted extensive attention from clinicians and researchers. Early diagnosis and treatment of thyroid nodules can effectively prevent thyroid cancer. Some thyroid nodules have a high cancer rate, and early diagnosis and early surgery can avoid more serious consequences. Domestic medical resources are tight, and medical staff need to face a large number of patients every day. Real-time diagnosis of thyroid nodules requires heavy t...

AI Technical Summary

Problems solved by technology

However, at present, most researches on the intelligent diagnosis of thyroid nodules by artificial intelligence have the problem that the modeling data is not comprehensive enough. Most of them choose to build models based on thyroid ultrasound image data, or relatively complete text feature data after selection. In the actual diagnosis and treatment process, The patient's historical clinical data, such as the patient's past medical history, medication records, and test results, will affect the judgment of whether the patient's nodules are benign or malignant. The current research article directly uses machine learning methods to directly train the classification model, lacking the intelligence after classification. There is a certain room for improvement in detection and positioning. Accurate intelligent detection has a better auxiliary effect on doctors' clinical diagnosis and treatment.

Images