变形还是不变形：通过无监督关键点检测实现新辅助化疗期间乳腺 DCE-MRI 的治疗感知纵向配准
To deform or not: treatment-aware longitudinal registration for breast DCE-MRI during neoadjuvant chemotherapy via unsupervised keypoints detection
AI总结：
本文提出基于无监督关键点检测与选择性体积保持的条件金字塔纵向配准网络，在 NAC 乳腺 DCE-MRI 上保护肿瘤体积并提升配准及 pCR 预测性能。
摘要：
临床医生将新辅助化疗（NAC）后的乳腺 DCE-MRI 与治疗前扫描进行比较，以评估对 NAC 的反应。临床证据表明，在不使受治疗肿瘤区域变形的情况下进行准确的纵向可变形配准，是量化肿瘤变化的关键。我们提出了一种基于无监督关键点检测和选择性体积保持的条件金字塔配准网络，用于量化随时间变化的情况。在该方法中，我们从 DCE-MRI 中提取结构关键点和异常关键点，将结构关键点用于配准算法以限制大变形，并采用基于异常关键点的体积保持损失，使肿瘤体积在配准后保持不变。我们使用了一个临床数据集，其中包含来自 314 名接受 NAC 治疗患者的 1630 次 MRI 扫描。结果表明，我们的方法具有更好的配准性能和更好的肿瘤体积保持效果。此外，基于所提方法的局部—全局联合生物标志物在病理完全缓解（pCR）预测中取得了较高准确率，表明预测信息存在于肿瘤区域之外。这些生物标志物有可能用于避免某些患者接受不必要的手术。对于临床医生和/或计算机系统在通过我们的方法配准的图像上进行后续肿瘤分割和反应预测，它可能具有价值。我们的代码可在 https://github.com/fiy2W/Treatment-aware-Longitudinal-Registration 获取。
英文摘要：
Clinicians compare breast DCE-MRI after neoadjuvant chemotherapy (NAC) with pre-treatment scans to evaluate the response to NAC. Clinical evidence supports that accurate longitudinal deformable registration without deforming treated tumor regions is key to quantifying tumor changes. We propose a conditional pyramid registration network based on unsupervised keypoint detection and selective volume-preserving to quantify changes over time. In this approach, we extract the structural and the abnormal keypoints from DCE-MRI, apply the structural keypoints for the registration algorithm to restrict large deformation, and employ volume-preserving loss based on abnormal keypoints to keep the volume of the tumor unchanged after registration. We use a clinical dataset with 1630 MRI scans from 314 patients treated with NAC. The results demonstrate that our method registers with better performance and better volume preservation of the tumors. Furthermore, a local-global-combining biomarker based on the proposed method achieves high accuracy in pathological complete response (pCR) prediction, indicating that predictive information exists outside tumor regions. The biomarkers could potentially be used to avoid unnecessary surgeries for certain patients. It may be valuable for clinicians and/or computer systems to conduct follow-up tumor segmentation and response prediction on images registered by our method. Our code is available on \url{https://github.com/fiy2W/Treatment-aware-Longitudinal-Registration}.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
