ImViA Laboratory, University of Burgundy, Batiment I3M, 64b rue sully, 21000, Dijon, France.
Radiation Oncology Department, CGFL, Dijon, France.
Int J Comput Assist Radiol Surg. 2020 Sep;15(9):1467-1476. doi: 10.1007/s11548-020-02231-x. Epub 2020 Jul 20.
This paper addresses the detection of the clinical target volume (CTV) in transrectal ultrasound (TRUS) image-guided intraoperative for permanent prostate brachytherapy. Developing a robust and automatic method to detect the CTV on intraoperative TRUS images is clinically important to have faster and reproducible interventions that can benefit both the clinical workflow and patient health.
We present a multi-task deep learning method for an automatic prostate CTV boundary detection in intraoperative TRUS images by leveraging both the low-level and high-level (prior shape) information. Our method includes a channel-wise feature calibration strategy for low-level feature extraction and learning-based prior knowledge modeling for prostate CTV shape reconstruction. It employs CTV shape reconstruction from automatically sampled boundary surface coordinates (pseudo-landmarks) to detect the low-contrast and noisy regions across the prostate boundary, while being less biased from shadowing, inherent speckles, and artifact signals from the needle and implanted radioactive seeds.
The proposed method was evaluated on a clinical database of 145 patients who underwent permanent prostate brachytherapy under TRUS guidance. Our method achieved a mean accuracy of [Formula: see text] and a mean surface distance error of [Formula: see text]. Extensive ablation and comparison studies show that our method outperformed previous deep learning-based methods by more than 7% for the Dice similarity coefficient and 6.9 mm reduced 3D Hausdorff distance error.
Our study demonstrates the potential of shape model-based deep learning methods for an efficient and accurate CTV segmentation in an ultrasound-guided intervention. Moreover, learning both low-level features and prior shape knowledge with channel-wise feature calibration can significantly improve the performance of deep learning methods in medical image segmentation.
本文针对经直肠超声（TRUS）图像引导下永久性前列腺近距离放疗中的临床靶区（CTV）检测问题展开研究。开发一种用于术中 TRUS 图像CTV 自动检测的稳健、自动方法对于实现更快、更具可重复性的干预措施至关重要，这有利于临床工作流程和患者健康。
我们提出了一种基于多任务深度学习的方法，用于术中 TRUS 图像中前列腺 CTV 边界的自动检测，充分利用了低水平和高水平（先验形状）信息。我们的方法包括用于低水平特征提取的通道特征校准策略和基于学习的前列腺 CTV 形状重建的先验知识建模。该方法通过从自动采样的边界表面坐标（伪标记）中进行 CTV 形状重建，来检测前列腺边界的低对比度和噪声区域，同时减少了来自阴影、固有斑点和针及植入放射性种子的伪影信号的偏差。
该方法在接受 TRUS 引导下永久性前列腺近距离放疗的 145 名患者的临床数据库上进行了评估。我们的方法的平均准确率为[公式：见正文]，平均表面距离误差为[公式：见正文]。大量消融和对比研究表明，与以前基于深度学习的方法相比，我们的方法在 Dice 相似系数上提高了 7%以上，3D Hausdorff 距离误差降低了 6.9mm。
我们的研究表明，基于形状模型的深度学习方法在超声引导干预中具有高效、准确的 CTV 分割的潜力。此外，通过通道特征校准学习低水平特征和先验形状知识可以显著提高医学图像分割中深度学习方法的性能。
