A recently developed hybrid method for automatic segmentation of 18 cardiac structures, combining deep learning, multi-atlas mapping and geometric segmentation of small challenging substructures, was independently validated on 30 lung cancer cases. These included anatomical and imaging variations, such as tumour abutting heart, lung collapse and metal artefacts. Automatic segmentations were compared with manual contours of the 18 structures using quantitative metrics, including Dice similarity coefficient (DSC), mean distance to agreement (MDA) and dose comparisons.
A comparison of manual and automatic contours across all cases showed a median DSC of 0.75-0.93 and a median MDA of 2.09-3.34 mm for whole heart and chambers. The median MDA for great vessels, coronary arteries, cardiac valves, sinoatrial and atrioventricular conduction nodes was 3.01-8.54 mm. For the 27 cases treated with curative intent (planned target volume dose ≥50 Gy), the median dose difference was -1.12 to 0.57 Gy (absolute difference of 1.13-3.25%) for the mean dose to heart and chambers; and -2.25 to 4.45 Gy (absolute difference of 0.94-6.79%) for the mean dose to substructures.
The novel hybrid automatic segmentation tool reported high accuracy and consistency over a validation set with challenging anatomical and imaging variations. This has promising applications in substructure dose calculations of large-scale datasets and for future studies on long-term cardiac toxicity.
方法:最近开发的用于自动分割18个心脏结构的混合方法,结合深度学习,小型挑战性子结构的多图集映射和几何分割,在30例肺癌病例中进行了独立验证。这些包括解剖学和影像学变化,如肿瘤邻接心脏,肺塌陷和金属伪影.使用定量指标将自动分割与18个结构的手动轮廓进行比较,包括骰子相似系数(DSC),平均协议距离(MDA)和剂量比较。
结果:所有病例的手动和自动轮廓比较显示,整个心脏和心室的中位DSC为0.75-0.93,中位MDA为2.09-3.34mm。大血管的MDA中位数,冠状动脉,心脏瓣膜,窦房和房室传导淋巴结为3.01-8.54mm。对于27例治愈意向治疗(计划目标体积剂量≥50Gy),心脏和心室平均剂量的中位剂量差异为-1.12~0.57Gy(绝对差异为1.13~3.25%);亚结构平均剂量差异为-2.25~4.45Gy(绝对差异为0.94~6.79%).
结论:新的混合自动分割工具报告了具有挑战性的解剖和成像变化的验证集的高准确性和一致性。这在大规模数据集的子结构剂量计算以及长期心脏毒性的未来研究中具有广阔的应用前景。