Department of Radiation Oncology Physics and Technology, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, No.440, Jiyan Road, Jinan 250117, Shandong Province, P.R.China.
Manteia Technologies Co.,Ltd, 1903, B Tower, Zijin Plaza, No.1811 Huandao East Road, Xiamen, 361001, China.
Radiother Oncol. 2022 Dec;177:222-230. doi: 10.1016/j.radonc.2022.11.004. Epub 2022 Nov 12.
Deep Learning (DL) technique has shown great potential but still has limited success in online contouring for MR-guided adaptive radiotherapy (MRgART). This study proposed a patient-specific DL auto-segmentation (DLAS) strategy using the patient's previous images and contours to update the model and improve segmentation accuracy and efficiency for MRgART.
A prototype model was trained for each patient using the first set of MRI and corresponding contours as inputs. The patient-specific model was updated after each fraction with all the available fractional MRIs/contours, and then used to predict the segmentation for the next fraction. During model training, a variant was fitted under consistency constraints, limiting the differences in the volume, length and centroid between the predictions for the latest MRI within a reasonable range. The model performance was evaluated for both organ-at-risks and tumors auto-segmentation for a total of 6 abdominal/pelvic cases (each with at least 8 sets of MRIs/contours) underwent MRgART through Dice Similarity Coefficient (DSC) and 95% Hausdorff Distance (HD95), and was compared with deformable image registration (DIR) and frozen DL model (no updating after pre-training). The contouring time was also recorded and analyzed.
The proposed model achieved superior performance with higher mean DSC (0.90, 95 % CI: 0.88-0.95), as compared to DIR (0.63, 95 %CI: 0.59-0.68) and frozen DL models (0.74, 95 % CI: 0.71-0.79). As for tumors, the proposed method yielded a median DSC of 0.95, 95 % CI: 0.94-0.97, and a median HD95 of 1.63 mm, 95 % CI: 1.22 mm-2.06 mm. The contouring time was reduced significantly (p < 0.05) using the proposed method (73.4 ± 6.5 secs) compared to the manual process (12 ∼ 22 mins). The online ART time was reduced to 1650 ± 274 seconds with the proposed method, as compared to 3251.8 ± 447 seconds using the original workflow.
The proposed patient-specific DLAS method can significantly improve the segmentation accuracy and efficiency for longitudinal MRIs, thereby facilitating the routine practice of MRgART.
深度学习（DL）技术在磁共振引导自适应放疗（MRgART）的在线勾画中显示出巨大潜力，但仍存在一定的局限性。本研究提出了一种基于患者既往图像和勾画的患者特异性深度神经网络自动勾画（DLAS）策略，以更新模型并提高勾画的准确性和效率。
使用第一组 MRI 和相应的勾画作为输入，为每位患者训练一个原型模型。在每个分次治疗后，利用所有可用的分次 MRI/勾画对患者特异性模型进行更新，然后用于预测下一个分次的勾画。在模型训练过程中，根据一致性约束对变体进行拟合，将最新 MRI 预测结果之间的体积、长度和质心差异限制在合理范围内。使用 Dice 相似系数（DSC）和 95%Hausdorff 距离（HD95）评估器官危及器官和肿瘤的自动勾画性能，共对 6 例接受 MRgART 的腹部/盆腔病例（每个病例至少有 8 组 MRI/勾画）进行了评估，并与可变形图像配准（DIR）和冻结的深度神经网络模型（预训练后不更新）进行了比较。还记录并分析了勾画时间。
与 DIR（0.63，95%CI：0.59-0.68）和冻结的深度神经网络模型（0.74，95%CI：0.71-0.79）相比，提出的模型在器官危及器官和肿瘤的勾画中均取得了更高的平均 DSC（0.90，95%CI：0.88-0.95）。对于肿瘤，该方法的中位 DSC 为 0.95，95%CI：0.94-0.97，中位 HD95 为 1.63mm，95%CI：1.22mm-2.06mm。与手动勾画（12∼22 分钟）相比，使用提出的方法（73.4±6.5 秒）显著减少了勾画时间（p<0.05）。使用提出的方法，在线 ART 时间从 3251.8±447 秒减少到 1650±274 秒。
提出的患者特异性深度神经网络自动勾画方法可以显著提高纵向 MRI 的勾画准确性和效率，从而促进 MRgART 的常规应用。
