King Abdulaziz University, Department of Diagnostic Radiology, Faculty of Applied Medical Sciences, King Abdulaziz University, Jeddah, Saudi Arabia.
University of Leeds, School of Medicine, Leeds, United Kingdom.
Phys Med Biol. 2023 Aug 29;68(17). doi: 10.1088/1361-6560/acf023.
Deep-learning auto-contouring (DL-AC) promises standardisation of organ-at-risk (OAR) contouring, enhancing quality and improving efficiency in radiotherapy. No commercial models exist for OAR contouring based on brain magnetic resonance imaging (MRI). We trained and evaluated computed tomography (CT) and MRI OAR autosegmentation models in RayStation. To ascertain clinical usability, we investigated the geometric impact of contour editing before training on model quality.Retrospective glioma cases were randomly selected for training (= 32, 47) and validation (= 9, 10) for MRI and CT, respectively. Clinical contours were edited using international consensus (gold standard) based on MRI and CT. MRI models were trained (i) using the original clinical contours based on planning CT and rigidly registered T1-weighted gadolinium-enhanced MRI (MRIu), (ii) as (i), further edited based on CT anatomy, to meet international consensus guidelines (MRIeCT), and (iii) as (i), further edited based on MRI anatomy (MRIeMRI). CT models were trained using: (iv) original clinical contours (CTu) and (v) clinical contours edited based on CT anatomy (CTeCT). Auto-contours were geometrically compared to gold standard validation contours (CTeCT or MRIeMRI) using Dice Similarity Coefficient, sensitivity, and mean distance to agreement. Models' performances were compared using paired Student's t-testing.The edited autosegmentation models successfully generated more segmentations than the unedited models. Paired t-testing showed editing pituitary, orbits, optic nerves, lenses, and optic chiasm on MRI before training significantly improved at least one geometry metric. MRI-based DL-AC performed worse than CT-based in delineating the lacrimal gland, whereas the CT-based performed worse in delineating the optic chiasm. No significant differences were found between the CTeCT and CTu except for optic chiasm.T1w-MRI DL-AC could segment all brain OARs except the lacrimal glands, which cannot be easily visualized on T1w-MRI. Editing contours on MRI before model training improved geometric performance. MRI DL-AC in RT may improve consistency, quality and efficiency but requires careful editing of training contours.
深度学习自动勾画（DL-AC）有望实现危及器官（OAR）勾画的标准化，提高放疗质量和效率。目前还没有基于脑磁共振成像（MRI）的 OAR 勾画商业模型。我们在 RayStation 中训练和评估了 CT 和 MRI OAR 自动分割模型。为了确定临床可用性，我们研究了在训练前对模型质量有影响的轮廓编辑的几何形状。回顾性脑胶质瘤病例被随机选择用于 MRI（n = 32，47）和 CT（n = 9，10）的训练和验证。临床轮廓使用基于 MRI 和 CT 的国际共识（金标准）进行编辑。MRI 模型（i）使用基于计划 CT 和刚性注册 T1 加权钆增强 MRI（MRIu）的原始临床轮廓进行训练，（ii）如（i），进一步基于 CT 解剖结构编辑，以满足国际共识指南（MRIeCT），和（iii）如（i），进一步基于 MRI 解剖结构编辑（MRIeMRI）。CT 模型使用：（iv）原始临床轮廓（CTu）和（v）基于 CT 解剖结构编辑的临床轮廓（CTeCT）进行训练。使用 Dice 相似系数、灵敏度和平均吻合距离，将自动轮廓与金标准验证轮廓（CTeCT 或 MRIeMRI）进行几何比较。使用配对学生 t 检验比较模型性能。
经过编辑的自动分割模型比未经编辑的模型成功生成了更多的分割。配对 t 检验显示，在训练前对垂体、眼眶、视神经、晶状体和视交叉进行 MRI 编辑显著提高了至少一个几何度量。MRI 基于的 DL-AC 在勾画泪腺方面的表现不如 CT 基于的好，而 CT 基于的在勾画视交叉方面的表现不如 MRI 基于的好。除了视交叉，CTeCT 和 CTu 之间没有发现显著差异。T1w-MRI DL-AC 可以分割除泪腺以外的所有脑 OAR，而泪腺在 T1w-MRI 上不易可视化。在模型训练前对 MRI 轮廓进行编辑可以提高几何性能。RT 中的 MRI DL-AC 可能会提高一致性、质量和效率，但需要仔细编辑训练轮廓。
