Jiangsu Provincial Engineering Research Center of Intelligent Technology for Healthcare, Ministry of Education, Jiangnan University, 1800 Lihu Avenue, Wuxi, 214122, Jiangsu, China.
Jiangsu Provincial Engineering Research Center of Intelligent Technology for Healthcare, Ministry of Education, Jiangnan University, 1800 Lihu Avenue, Wuxi, 214122, Jiangsu, China.
Comput Med Imaging Graph. 2024 Oct;117:102439. doi: 10.1016/j.compmedimag.2024.102439. Epub 2024 Sep 28.
Ultrasound examination plays a crucial role in the clinical diagnosis of thyroid nodules. Although deep learning technology has been applied to thyroid nodule examinations, the existing methods all overlook the prior knowledge of nodules moving along a straight line in the video. We propose a new detection model, DiffusionVID-Line, and design a novel tracking algorithm, ByteTrack-Line, both of which fully leverage the prior knowledge of linear motion of nodules in thyroid ultrasound videos. Among them, ByteTrack-Line groups detected nodules, further reducing the workload of doctors and significantly improving their diagnostic speed and accuracy. In DiffusionVID-Line, we propose two new modules: Freq-FPN and Attn-Line. Freq-FPN module is used to extract frequency features, taking advantage of these features to reduce the impact of image blur in ultrasound videos. Based on the standard practice of segmented scanning by doctors, Attn-Line module enhances the attention on targets moving along a straight line, thus improving the accuracy of detection. In ByteTrack-Line, considering the characteristic of linear motion of nodules, we propose the Match-Line association module, which reduces the number of nodule ID switches. In the testing of the detection and tracking datasets, DiffusionVID-Line achieved a mean Average Precision (mAP50) of 74.2 for multiple tissues and 85.6 for nodules, while ByteTrack-Line achieved a Multiple Object Tracking Accuracy (MOTA) of 83.4. Both nodule detection and tracking have achieved state-of-the-art performance.
超声检查在甲状腺结节的临床诊断中起着至关重要的作用。虽然深度学习技术已经应用于甲状腺结节检查，但现有的方法都忽略了结节在视频中沿直线移动的先验知识。我们提出了一种新的检测模型 DiffusionVID-Line，并设计了一种新颖的跟踪算法 ByteTrack-Line，这两种方法都充分利用了甲状腺超声视频中结节线性运动的先验知识。其中，ByteTrack-Line 对检测到的结节进行分组，进一步减轻了医生的工作量，显著提高了他们的诊断速度和准确性。在 DiffusionVID-Line 中，我们提出了两个新模块：Freq-FPN 和 Attn-Line。Freq-FPN 模块用于提取频率特征，利用这些特征来减少超声视频中图像模糊的影响。基于医生分段扫描的标准做法，Attn-Line 模块增强了对沿直线移动的目标的注意力，从而提高了检测的准确性。在 ByteTrack-Line 中，考虑到结节的线性运动特征，我们提出了 Match-Line 关联模块，该模块减少了结节 ID 切换的数量。在检测和跟踪数据集的测试中，DiffusionVID-Line 在多个组织上的平均精度（mAP50）达到了 74.2，在结节上的平均精度（mAP50）达到了 85.6，而 ByteTrack-Line 在多个目标跟踪精度（MOTA）上达到了 83.4。结节检测和跟踪都达到了最先进的性能。
