TMTV-Net:淋巴瘤PET/CT图像中全自动总代谢肿瘤体积分割——多中心泛化性分析
研究纳入四个中心共1418例2-[18F]FDG PET/CT扫描,开发级联深度学习模型实现淋巴瘤总代谢肿瘤体积(TMTV)全自动分割。内部测试平均DSC为0.68±0.12,多中心外部测试(n=518)平均DSC为0.66±0.18,显著优于nnU-Net和SWIN UNETR等先进方法(p<0.001),TMTV预测与真实值相关系数为0.89(p<0.001)。
要点
- 级联软投票引导方法在外部多中心数据(n=518)上平均DSC为0.66±0.18,显著优于nnU-Net和SWIN UNETR(p<0.001)。
- TMTV预测与真实值相关系数为0.89(p<0.001),外部测试整体性能较内部仅下降2%。
- 测试时间增强对本级联方法影响可忽略(DSC:0.66±0.16)。
核心结论
- TMTV-Net在多中心外部数据集上对多种淋巴瘤亚型的TMTV分割表现出较强性能和泛化性,外部测试整体性能仅下降2%。
- 模型已公开可用,便于在不同机构数据集上进行多中心评估和泛化性分析。
研究设计与样本
- 研究纳入四个中心共1418例2-[18F]FDG PET/CT扫描。
- 900例扫描用于开发/验证/测试阶段,其中450例为淋巴瘤、肺癌和黑色素瘤扫描,450例为阴性扫描。
- 518例扫描用于多中心外部测试,均为来自不同中心的淋巴瘤患者,包括弥漫大B细胞淋巴瘤、原发纵隔大B细胞淋巴瘤和经典霍奇金淋巴瘤。
- 方法为级联策略:先将PET/CT图像重采样为不同体素大小,在每个重采样数据集上采用五折交叉验证训练多分辨率3D U-Net并集成模型;对预测掩膜进行软投票后,将概率平均预测与输入影像数据输入另一个3D U-Net,采用半监督损失训练。
- 额外评估了测试时间增强(TTA)对分割性能的影响。
主要发现
- 内部测试数据平均DSC为0.68±0.12。
- 多中心外部数据(n=518)平均DSC为0.66±0.18,显著优于nnU-Net和SWIN UNETR等先进方法(p<0.001)。
- TTA对部分对比方法有性能提升,但对本级联方法影响可忽略(DSC:0.66±0.16)。
- TMTV预测与真实值相关系数为0.89(p<0.001)。
- 外部多中心数据集(n=518)TMTV预测平均相对误差(ARE)为0.43±0.54,绝对误差(AE)为157.32±378.12 mL;排除10%离群值(n=53)后,ARE为0.30±0.22,AE为82.05±99.78 mL。
关键数值与推荐
- 内部测试数据平均DSC为0.68±0.12。
- 外部多中心测试数据(n=518)平均DSC为0.66±0.18,显著优于nnU-Net和SWIN UNETR(p<0.001)。
- TMTV预测与真实值相关系数为0.89(p<0.001)。
- 外部多中心数据集(n=518)TMTV预测ARE=0.43±0.54,AE=157.32±378.12 mL。
- 排除10%离群值(n=53)后,ARE=0.30±0.22,AE=82.05±99.78 mL。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
PURPOSE: Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. METHODS: Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. RESULTS: Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. CONCLUSION: TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.
引用
- Yousefirizi F, Klyuzhin IS, O JH, et al. TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis. Eur J Nucl Med Mol Imaging. 2024. PMID: 38326655. doi:10.1007/s00259-024-06616-x
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。