Chu Chen, Zheng Jihui, Zhou Yong
Department of General Surgery, Fourth Affiliated Hospital of China Medical University, No. 4, Chongshan East Road, Huanggu District, Shenyang City, Liaoning Province, 110032, China.
Department of Ultrasound, Affiliated Hospital of China Medical University, No. 4, Chongshan East Road, Huanggu District, Shenyang City, Liaoning Province, 110032, China.
Comput Methods Programs Biomed. 2021 Feb;199:105906. doi: 10.1016/j.cmpb.2020.105906. Epub 2020 Dec 17.
Aiming at the time consuming processing of existing thyroid nodule detection and difficulty in feature extraction, U-Net-based thyroid nodule detection is proposed to perform computed aided diagnosis.
This paper proposes a mark-guided ultrasound deep network segmentation model of thyroid nodules. By comparing with VGG19, Inception V3, DenseNet 161, segmentation accuracy, segmentation edge and network operation time, it is found that the algorithm in this paper has relative advantages.
U-Net network-based ultrasound thyroid nodules segmented the nodule area overlapped with the manually depicted nodule area close to 100%, the segmentation accuracy rate was as high as 0.9785, and the U-Net segmentation result was closer to the manually depicted nodule. The accuracy of U-Net segmentation of the thyroid is about 3% higher than the other three networks.
The segmentation of nodules based on U-Net proposed in this paper significantly improves the segmentation accuracy of thyroid nodules with a small training data set, and provides a comprehensive reference for clinical diagnosis and treatment.
针对现有甲状腺结节检测处理耗时且特征提取困难的问题，提出基于U-Net的甲状腺结节检测方法以实现计算机辅助诊断。
本文提出一种标记引导的甲状腺结节超声深度网络分割模型。通过与VGG19、Inception V3、DenseNet 161比较分割精度、分割边缘及网络运行时间，发现本文算法具有相对优势。
基于U-Net网络的超声甲状腺结节分割出的结节区域与手动描绘的结节区域重叠接近100%，分割准确率高达0.9785，且U-Net分割结果更接近手动描绘的结节。U-Net对甲状腺的分割准确率比其他三个网络约高3%。
本文提出的基于U-Net的结节分割方法在训练数据集较小的情况下显著提高了甲状腺结节的分割精度，为临床诊疗提供了全面参考。
