Scholarlink

<p>Automated segmentation of PDAC is a prerequisite to realize the promise of precision oncology, yet existing approaches remain insufficiently validated at the cohort scale and acquisition heterogeneity required for clinical deployment. We developed and validated a pancreas-localized three-dimensional convolutional neural network model (Model-BB) using 1859 multi-institutional treatment-naïve, biopsy-confirmed PDAC CT examinations. Model-BB achieved a DSC of 0.76 ± 0.13 on internal testing and 0.76 ± 0.09 on external validation, with stable performance across acquisition site, scanner vendor, slice thickness, and temporal epochs. Upstream localization analysis showed complete tumor enclosure in 224 of 241 test cases (92.9%), with preserved performance (DSC 0.78) among cases with partial peripheral exclusion, indicating that localization error was not the principal source of reduced performance. In a controlled architectural benchmark, Model-BB outperformed Swin UNETR, a 3D vision transformer (DSC, 0.76 ± 0.13 versus 0.68 ± 0.18; <i>p</i> &lt; 0.001). On a difficulty-enriched subset (<i>n</i> = 50), Model-BB achieved DSC 0.71 against a STAPLE-derived consensus, exceeding individual reader-pair agreement (DSC range, 0.57–0.65), with concordance correlation coefficient 0.93 for tumor volume. These findings support anatomically constrained, task-specific segmentation as a reproducible geometric substrate for volumetric tumor-burden quantification, treatment-response assessment, and multimodal outcomes modeling, pending prospective validation in clinical-trial workflows.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Anatomically constrained deep learning for clinical-grade volumetric pancreatic cancer segmentation: development, validation, and architectural benchmarking

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月3日

查看全文
  • Sovanlal Mukherjee,
  • Khurram Khaliq Bhinder,
  • Armin Zarrintan,
  • Ajith Antony,
  • Angela Ammirabile,
  • Ahmed Jadoon,
  • Subhosree Dey,
  • Baloy J. Talukdar,
  • Takeru Yamaguchi,
  • Rondell P. Graham,
  • Suresh T. Chari,
  • Ajit Harishkumar Goenka

摘要

Automated segmentation of PDAC is a prerequisite to realize the promise of precision oncology, yet existing approaches remain insufficiently validated at the cohort scale and acquisition heterogeneity required for clinical deployment. We developed and validated a pancreas-localized three-dimensional convolutional neural network model (Model-BB) using 1859 multi-institutional treatment-naïve, biopsy-confirmed PDAC CT examinations. Model-BB achieved a DSC of 0.76 ± 0.13 on internal testing and 0.76 ± 0.09 on external validation, with stable performance across acquisition site, scanner vendor, slice thickness, and temporal epochs. Upstream localization analysis showed complete tumor enclosure in 224 of 241 test cases (92.9%), with preserved performance (DSC 0.78) among cases with partial peripheral exclusion, indicating that localization error was not the principal source of reduced performance. In a controlled architectural benchmark, Model-BB outperformed Swin UNETR, a 3D vision transformer (DSC, 0.76 ± 0.13 versus 0.68 ± 0.18; p < 0.001). On a difficulty-enriched subset (n = 50), Model-BB achieved DSC 0.71 against a STAPLE-derived consensus, exceeding individual reader-pair agreement (DSC range, 0.57–0.65), with concordance correlation coefficient 0.93 for tumor volume. These findings support anatomically constrained, task-specific segmentation as a reproducible geometric substrate for volumetric tumor-burden quantification, treatment-response assessment, and multimodal outcomes modeling, pending prospective validation in clinical-trial workflows.