Medical UltraSound Image Computing (MUSIC) Lab, School of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, China.
Department of Ultrasound, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China.
Med Image Anal. 2022 Aug;80:102490. doi: 10.1016/j.media.2022.102490. Epub 2022 Jun 5.
Ultrasound (US) plays a vital role in breast cancer screening, especially for women with dense breasts. Common practice requires a sonographer to recognize key diagnostic features of a lesion and record a single or several representative frames during the dynamic scanning before performing the diagnosis. However, existing computer-aided diagnosis tools often focus on the final diagnosis process while neglecting the influence of the keyframe selection. Moreover, the lesions could have highly-irregular shapes, varying sizes, and locations during the scanning. The recognition of diagnostic characteristics associated with the lesions is challenging and also faces severe class imbalance. To address these, we proposed a reinforcement learning-based framework that can automatically extract keyframes from breast US videos of unfixed length. It is equipped with a detection-based nodule filtering module and a novel reward mechanism that can integrate anatomical and diagnostic features of the lesions into keyframe searching. A simple yet effective loss function was also designed to alleviate the class imbalance issue. Extensive experiments illustrate that the proposed framework can benefit from both innovations and is able to generate representative keyframe sequences in various screening conditions.
超声（US）在乳腺癌筛查中起着至关重要的作用，特别是对于乳腺致密的女性。通常的做法是要求超声医师识别病变的关键诊断特征，并在进行诊断之前在动态扫描过程中记录单个或多个有代表性的帧。然而，现有的计算机辅助诊断工具通常侧重于最终的诊断过程，而忽略了关键帧选择的影响。此外，在扫描过程中，病变可能具有高度不规则的形状、不同的大小和位置。与病变相关的诊断特征的识别具有挑战性，并且还面临严重的类不平衡问题。针对这些问题，我们提出了一种基于强化学习的框架，可以自动从不定长的乳腺超声视频中提取关键帧。它配备了基于检测的结节过滤模块和一种新颖的奖励机制，可以将病变的解剖学和诊断特征整合到关键帧搜索中。还设计了一个简单而有效的损失函数来减轻类不平衡问题。广泛的实验表明，所提出的框架可以从这两项创新中受益，并能够在各种筛查条件下生成有代表性的关键帧序列。
