基于深度学习在表观扩散系数图上分割和量化肺部肿块的多中心可重复性研究
该研究开发并验证了ADCSegNet深度学习模型，用于在ADC图上自动分割肺部肿块并评估自动ADC量化的可重复性。模型在中心1（303例）训练测试，在中心2（70例）和中心3（12例）外部评估，分割精度DSC内部0.843、外部0.701，优于nnU-Net等模型。模型与高年资放射科医生的ADC测量一致性最高，平均ADC平均差异0.00（95%一致性界限-0.13至0.13）×10⁻³ mm²/s，重复扫描稳定性优于放射科医生。
要点
- ADCSegNet在ADC图上自动分割肺部肿块，内部DSC 0.843，外部DSC 0.701。
- 模型与高年资放射科医生ADC测量一致性最高，平均ADC差异0.00×10⁻³ mm²/s。
- 模型重复扫描稳定性优于放射科医生，平均ADC的CCC为0.931 vs. 0.908。
关键影像表现/征象
- ADCSegNet在ADC图上自动分割肺部肿块，内部DSC为0.843，外部DSC为0.701，优于nnU-Net和其他当代骨干网络。
- 模型与高年资放射科医生的一致性最高，平均ADC的平均差异为0.00（95%一致性界限-0.13至0.13）×10⁻³ mm²/s，优于模型与低年资医生及放射科医生之间的一致性。
诊断标准与关键数值
- 分割精度以Dice相似系数（DSC）评估，内部0.843，外部0.701。
- 一致性以Bland-Altman分析、ICC(2,1)和QIBA重复性系数（RDC）评估，多数ADC指标的ICC(2,1)超过0.75。
- 关键指标重复性系数：平均ADC的RDC%模型-高年资医生为12.98%，高年资-低年资医生为22.87%。
- 重复扫描稳定性以Lin一致性相关系数（CCC）评估，平均ADC的CCC模型为0.931，放射科医生为0.908。
临床应用与阅片要点
- ADCSegNet可用于ADC图上肺部肿块的自动分割和ADC量化，与高年资放射科医生测量高度一致，有望辅助日常ADC测量。
- 模型外部泛化性能下降（DSC 0.701），跨中心应用时需注意分割精度可能低于内部测试。
局限性
- 外部评估数据集样本量较小（中心2为70次检查，中心3为12次检查），可能影响泛化性结论的稳健性。
- 重复扫描稳定性评估的前瞻性数据集样本量有限（22×2次检查）。
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）。
英文原文（摘要）· 点击展开
OBJECTIVES: To develop and validate a deep learning model for automatic segmentation of pulmonary masses on apparent diffusion coefficient (ADC) maps and to assess repeatability of automated ADC quantification.
METHODS: We proposed ADCSegNet, a deep learning model tailored to pulmonary ADC maps, trained and tested on ADC maps from centre 1 (303 MRI examinations) and externally evaluated on datasets from centre 2 (70 examinations) and centre 3 (12 examinations) for generalisability. Segmentation accuracy was evaluated using the Dice similarity coefficient (DSC). Agreement between model-derived and manual ADC measurements (senior and junior radiologists) was examined using Bland-Altman analysis, the intraclass correlation coefficient [ICC(2,1)], and the reproducibility coefficient (RDC) following QIBA profiles. Repeat-scan stability was assessed on an independent prospective test-retest dataset (22 × 2 examinations) using Lin's concordance correlation coefficient (CCC).
RESULTS: ADCSegNet achieved DSCs of 0.843 internally and 0.701 externally, outperforming nnU-Net and other contemporary backbones. Agreement was highest between the model and the senior radiologist; for mean ADC, the mean difference was 0.00 (95% limits of agreement, -0.13 to 0.13) × 10⁻3 mm2/s, exceeding model-junior and inter-radiologist agreement. ICC(2,1) exceeded 0.75 for most ADC metrics, and RDC confirmed high reproducibility for key metrics (mean ADC RDC%: 12.98% for model-senior vs. 22.87% for senior-junior radiologist). Test-retest analyses showed higher repeatability of ADCSegNet than radiologists for core metrics such as mean ADC (CCC 0.931 vs. 0.908).
CONCLUSIONS: ADCSegNet enables accurate segmentation of pulmonary masses on ADC maps and provides reproducible automated ADC quantification, outperforming state-of-the-art backbone models while showing strong agreement with expert readers and stable performance on repeat scans.
引用
- Ding L, Chen T, Gao Q, et al. Deep learning-based segmentation and quantification of pulmonary masses on apparent diffusion coefficient maps: a multicentre reproducibility study. Eur J Radiol. 2026. PMID: 42035562. doi:10.1016/j.ejrad.2026.112881
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。