Scholarlink

Background <p>Sarcopenia, characterized by progressive skeletal muscle loss, is associated with poor outcomes in various diseases. Traditional methods for assessing muscle cross-sectional area using computed tomography (CT) scans are manual, time-consuming and prone to variability.</p> Aim <p>This study comprehensively validates a deep-learning (DL) pipeline for accurate and reproducible sarcopenia detection on computed tomography across diverse disease abdominal conditions and imaging protocols.</p> Methods <p>We utilized the publicly available Sparsely Annotated Region and Organ Segmentation (SAROS) CT dataset (<i>n</i> = 550 CT scans, 6516 slices) for model training. Testing was conducted on 601 CT scans from public (SAROS, Cancer Imaging Archive [TCIA]&#xa0;, WAW-TACE) and in-house multi-center datasets representing varied clinical conditions (acute pancreatitis, inflammatory bowel disease, gallbladder cancer and distal bile duct obstruction). The implemented pipeline integrated TotalSegmentator for L3 vertebral segmentation, automated L3 slice extraction and skeletal muscle segmentation using nnU-Net. Performance evaluation included expert qualitative scoring, Dice scores, intersection over union (IoU) and diagnostic accuracy metrics for sarcopenia detection.</p> Results <p>The DL pipeline demonstrated consistent segmentation accuracy across diverse datasets, with mean Dice scores ranging from 0.9287 to 0.9701 and mean IoU values up to 0.9423. Expert evaluation confirmed reliable L3 vertebral segmentation (78%–85% rated as complete) and skeletal muscle segmentation (90%–92.6% rated as excellent). Sarcopenia detection was consistent across varied patient populations, with sensitivity (0.94–0.97), specificity (0.84–0.97) and AUC values up to 0.92. Importantly, sub-group analysis confirmed comparable performance across varying disease conditions, CT protocols, contrast usage and radiation doses.</p> Conclusion <p>This study demonstrates that a deep-learning pipeline can achieve consistent and reliable performance for skeletal muscle segmentation and sarcopenia detection across heterogeneous abdominal CT protocols and diverse clinical conditions.</p> Graphical Abstract <p></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

Deep-learning pipeline for automated skeletal muscle segmentation and sarcopenia detection

  • 期刊论文

  • 发表日期：

    2026年1月19日

查看全文
  • Pankaj Gupta,
  • Niharika Dutta,
  • Saroj K. Sinha,
  • Harjeet Singh,
  • Santosh Irrinki,
  • Ajay Gulati,
  • Madhurima Sharma,
  • Mahesh Prakash,
  • Anindita Sinha,
  • Gaurav Prakash,
  • Thakur Deen Yadav,
  • Lileshwar Kaman,
  • Rajnikant Yadav,
  • Archana Gupta,
  • Ishan Kumar,
  • Kajal Kumari,
  • Rajesh Gupta,
  • Usha Dutta

摘要

Background

Sarcopenia, characterized by progressive skeletal muscle loss, is associated with poor outcomes in various diseases. Traditional methods for assessing muscle cross-sectional area using computed tomography (CT) scans are manual, time-consuming and prone to variability.

Aim

This study comprehensively validates a deep-learning (DL) pipeline for accurate and reproducible sarcopenia detection on computed tomography across diverse disease abdominal conditions and imaging protocols.

Methods

We utilized the publicly available Sparsely Annotated Region and Organ Segmentation (SAROS) CT dataset (n = 550 CT scans, 6516 slices) for model training. Testing was conducted on 601 CT scans from public (SAROS, Cancer Imaging Archive [TCIA] , WAW-TACE) and in-house multi-center datasets representing varied clinical conditions (acute pancreatitis, inflammatory bowel disease, gallbladder cancer and distal bile duct obstruction). The implemented pipeline integrated TotalSegmentator for L3 vertebral segmentation, automated L3 slice extraction and skeletal muscle segmentation using nnU-Net. Performance evaluation included expert qualitative scoring, Dice scores, intersection over union (IoU) and diagnostic accuracy metrics for sarcopenia detection.

Results

The DL pipeline demonstrated consistent segmentation accuracy across diverse datasets, with mean Dice scores ranging from 0.9287 to 0.9701 and mean IoU values up to 0.9423. Expert evaluation confirmed reliable L3 vertebral segmentation (78%–85% rated as complete) and skeletal muscle segmentation (90%–92.6% rated as excellent). Sarcopenia detection was consistent across varied patient populations, with sensitivity (0.94–0.97), specificity (0.84–0.97) and AUC values up to 0.92. Importantly, sub-group analysis confirmed comparable performance across varying disease conditions, CT protocols, contrast usage and radiation doses.

Conclusion

This study demonstrates that a deep-learning pipeline can achieve consistent and reliable performance for skeletal muscle segmentation and sarcopenia detection across heterogeneous abdominal CT protocols and diverse clinical conditions.

Graphical Abstract