Paper
28 July 2023 Classification of thyroid ultrasound standard section based on ResNet-cbam
Yifeng Zhu, Lanzhen Yao, Xingyu Zhou, Xiaofeng An
Author Affiliations +
Proceedings Volume 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023); 127160V (2023) https://doi.org/10.1117/12.2685643
Event: Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 2023, Xi'an, China
Abstract
For organ ultrasound examination, it is very important to accurately obtain the standard section of classified organ ultrasound images. In this paper, a method based on ResNet-cbam is proposed to identify the classified ultrasonic standard plane. By collecting thyroid ultrasound images, which are divided into TPTI transverse section isthm us, TPRT transverse section right thyroid lobe, TPLT transverse section left thyroid lobe and lateral thyroid lobe longitudinal section. After image denoising and enhancement preprocessing, several models are first used for experiments, which show that ResNet-cbam has the best classification and recognition effect. By constantly adjusting the ResNet-cbam model structure, the number of iterations comparative experiments on changing the learning rate and activation function show that the best experimental effect is when Resnet18-cbam and the learning rate lr is 0.001 and the activation function is relu() function. Finally, the accuracy of classification and recognition is 89%, which proves that ResNet-cbam can recognize the standard section of thyroid well.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yifeng Zhu, Lanzhen Yao, Xingyu Zhou, and Xiaofeng An "Classification of thyroid ultrasound standard section based on ResNet-cbam", Proc. SPIE 12716, Third International Conference on Digital Signal and Computer Communications (DSCC 2023), 127160V (28 July 2023); https://doi.org/10.1117/12.2685643
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Thyroid

Ultrasonography

Image classification

Image processing

Standards development

Ultrasonics

Data modeling

Back to Top