Department of Nuclear Medicine, Nanjing Drum Tower Hospital, the Affiliated Hospital of Nanjing University Medical School, No. 321, Zhongshan Road, Nanjing City, 210008, Jiangsu Province, China.
National Institute of Healthcare Data Science at Nanjing University, Nanjing, China.
Eur Radiol. 2022 Jul;32(7):4801-4812. doi: 10.1007/s00330-022-08573-1. Epub 2022 Feb 15.
To demonstrate the effectiveness of automatic segmentation of diffuse large B-cell lymphoma (DLBCL) in 3D FDG-PET scans using a deep learning approach and validate its value in prognosis in an external validation cohort.
Two PET datasets were retrospectively analysed: 297 patients from a local centre for training and 117 patients from an external centre for validation. A 3D U-Net architecture was trained on patches randomly sampled within the PET images. Segmentation performance was evaluated by six metrics, including the Dice similarity coefficient (DSC), Jaccard similarity coefficient (JSC), sensitivity (Se), positive predictive value (PPV), Hausdorff distance 95 (HD 95), and average symmetric surface distance (ASSD). Finally, the prognostic value of predictive total metabolic tumour volume (pTMTV) was validated in real clinical applications.
The mean DSC, JSC, Se, PPV, HD 95, and ASSD (with standard deviation) for the validation cohort were 0.78 ± 0.25, 0.69 ± 0.26, 0.81 ± 0.27, 0.82 ± 0.25, 24.58 ± 35.18, and 4.46 ± 8.92, respectively. The mean ground truth TMTV (gtTMTV) and pTMTV were 276.6 ± 393.5 cm and 301.9 ± 510.5 cm in the validation cohort, respectively. Perfect homogeneity in the Bland-Altman analysis and a strong positive correlation in the linear regression analysis (R linear = 0.874, p < 0.001) were demonstrated between gtTMTV and pTMTV. pTMTV (≥ 201.2 cm) (PFS: HR = 3.097, p = 0.001; OS: HR = 6.601, p < 0.001) was shown to be an independent factor of PFS and OS.
The FCN model with a U-Net architecture can accurately segment lymphoma lesions and allow fully automatic assessment of TMTV on PET scans for DLBCL patients. Furthermore, pTMTV is an independent prognostic factor of survival in DLBCL patients.
•The segmentation model based on a U-Net architecture shows high performance in the segmentation of DLBCL patients on FDG-PET images. •The proposed method can provide quantitative information as a predictive TMTV for predicting the prognosis of DLBCL patients.
使用深度学习方法展示在 3D FDG-PET 扫描中自动分割弥漫性大 B 细胞淋巴瘤（DLBCL）的有效性，并在外部验证队列中验证其在预后中的价值。
回顾性分析了两个 PET 数据集：来自本地中心的 297 名患者用于训练，以及来自外部中心的 117 名患者用于验证。在 PET 图像内随机采样的斑块上训练 3D U-Net 架构。使用六个指标评估分割性能，包括 Dice 相似系数（DSC）、Jaccard 相似系数（JSC）、敏感性（Se）、阳性预测值（PPV）、Hausdorff 距离 95（HD 95）和平均对称表面距离（ASSD）。最后，在实际临床应用中验证了预测总代谢肿瘤体积（pTMTV）的预后价值。
验证队列的平均 DSC、JSC、Se、PPV、HD 95 和 ASSD（标准差）分别为 0.78 ± 0.25、0.69 ± 0.26、0.81 ± 0.27、0.82 ± 0.25、24.58 ± 35.18 和 4.46 ± 8.92。验证队列中的平均地面真实 TMTV（gtTMTV）和 pTMTV 分别为 276.6 ± 393.5 cm 和 301.9 ± 510.5 cm。在 Bland-Altman 分析中表现出完美的同质性，在线性回归分析中表现出强烈的正相关（R 线性= 0.874，p < 0.001），gtTMTV 与 pTMTV 之间存在。pTMTV（≥201.2 cm）（PFS：HR = 3.097，p = 0.001；OS：HR = 6.601，p < 0.001）被证明是 PFS 和 OS 的独立因素。
基于 U-Net 架构的 FCN 模型可以准确分割淋巴瘤病变，并允许在 DLBCL 患者的 PET 扫描上进行完全自动的 TMTV 评估。此外，pTMTV 是 DLBCL 患者生存的独立预后因素。
•基于 U-Net 架构的分割模型在 FDG-PET 图像上显示出对 DLBCL 患者的高分割性能。•该方法可以提供定量信息，作为预测 TMTV，用于预测 DLBCL 患者的预后。
