Large-scale multi-center CT and MRI segmentation of pancreas with deep learning

Automated volumetric segmentation of the pancreas on cross-sectional imaging is needed for diagnosis and follow-up of pancreatic diseases. While CT-based pancreatic segmentation is more established, MRI-based segmentation methods are understudied, largely due to a lack of publicly available datasets, benchmarking research efforts, and domain-specific deep learning methods. In this retrospective study, we collected a large dataset (767 scans from 499 participants) of T1-weighted (T1 W) and T2-weighted (T2 W) abdominal MRI series from five centers between March 2004 and November 2022. We also collected CT scans of 1,350 patients from publicly available sources for benchmarking purposes. We introduced a new pancreas segmentation method, called PanSegNet, combining the strengths of nnUNet and a Transformer network with a new linear attention module enabling volumetric computation. We tested PanSegNet's accuracy in cross-modality (a total of 2,117 scans) and cross-center settings with Dice and Hausdorff distance (HD95) evaluation metrics. We used Cohen's kappa statistics for intra and inter-rater agreement evaluation and paired t-tests for volume and Dice comparisons, respectively. For segmentation accuracy, we achieved Dice coefficients of 88.3% (±7.2%, at case level) with CT, 85.0% (±7.9%) with T1 W MRI, and 86.3% (±6.4%) with T2 W MRI. There was a high correlation for pancreas volume prediction with R<sup>2</sup> of 0.91, 0.84, and 0.85 for CT, T1 W, and T2 W, respectively. We found moderate inter-observer (0.624 and 0.638 for T1 W and T2 W MRI, respectively) and high intra-observer agreement scores. All MRI data is made available at https://osf.io/kysnj/. Our source code is available at https://github.com/NUBagciLab/PaNSegNet.

Authors

Institutions

Publication Details

Journal
Medical Image Analysis
Published
2024-11-08
DOI
https://doi.org/10.1016/j.media.2024.103382
Citations
45
Primary Topic
Medical Image Segmentation Techniques
Type
article
Field-Weighted Citation Impact
11.21

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Large-scale multi-center CT and MRI segmentation of pancreas with deep learning

Hongyi Pan, Onkar Susladkar, Rajesh N. Keswani, Megan Engels et al.
45 citations
Medical Image Analysis
Medical Image Segmentation Techniques
11.21
article

Large-scale multi-center CT and MRI segmentation of pancreas with deep learning

Hongyi Pan, Onkar Susladkar, Rajesh N. Keswani, Megan Engels, Michael B. Wallace, Chenchan Huang, Derk C.F. Klatte, Boqing Gong, Barış Türkbey, Emil Agarunov, Ulaş Bağcı, Görkem Durak, Linkai Peng, Pallavi Tiwari, Marco J. Bruno, Debesh Jha, Ilkin Isler, Alpay Medetalibeyoğlu, Ayis Pyrros, Candice W. Bolan, Vandan Gorade, Nassier Harfouch, Tamas A. Gonda, Yury Velichko, Sanne Hoogenboom, Elif Keleş, Ivo G. Schoots, Temel Tirkes, Concetto Spampinato, Frank H. Miller, Cemal Yazıcı, Amir Bourhani, Yavuz Taktak, Zheyuan Zhang, Bin Wang, Camila Lopes Vendrami, Asli C. Ormeci, Lanhong Yao
article en
45 citations

Abstract

Automated volumetric segmentation of the pancreas on cross-sectional imaging is needed for diagnosis and follow-up of pancreatic diseases. While CT-based pancreatic segmentation is more established, MRI-based segmentation methods are understudied, largely due to a lack of publicly available datasets, benchmarking research efforts, and domain-specific deep learning methods. In this retrospective study, we collected a large dataset (767 scans from 499 participants) of T1-weighted (T1 W) and T2-weighted (T2 W) abdominal MRI series from five centers between March 2004 and November 2022. We also collected CT scans of 1,350 patients from publicly available sources for benchmarking purposes. We introduced a new pancreas segmentation method, called PanSegNet, combining the strengths of nnUNet and a Transformer network with a new linear attention module enabling volumetric computation. We tested PanSegNet's accuracy in cross-modality (a total of 2,117 scans) and cross-center settings with Dice and Hausdorff distance (HD95) evaluation metrics. We used Cohen's kappa statistics for intra and inter-rater agreement evaluation and paired t-tests for volume and Dice comparisons, respectively. For segmentation accuracy, we achieved Dice coefficients of 88.3% (±7.2%, at case level) with CT, 85.0% (±7.9%) with T1 W MRI, and 86.3% (±6.4%) with T2 W MRI. There was a high correlation for pancreas volume prediction with R<sup>2</sup> of 0.91, 0.84, and 0.85 for CT, T1 W, and T2 W, respectively. We found moderate inter-observer (0.624 and 0.638 for T1 W and T2 W MRI, respectively) and high intra-observer agreement scores. All MRI data is made available at https://osf.io/kysnj/. Our source code is available at https://github.com/NUBagciLab/PaNSegNet.

Medical Image AnalysisVol. 99
Northwestern University (US), Google (United States) (US), National Institutes of Health (US), Intel (United States) (US), University of Wisconsin–Madison (US), Chicago Department of Public Health (US), Jacksonville College (US), Erasmus MC (NL), WinnMed (US), University of Florida (US), University of Catania (IT), University of Illinois Chicago (US), Northwestern University (PH), University of Illinois Hospital & Health Sciences System (US), National Cancer Institute (US), Mayo Clinic in Florida (US), Indiana University School of Medicine, Indiana University – Purdue University Indianapolis (US), New York University (US), Istanbul University (TR), University of Amsterdam (NL), Erasmus University Rotterdam (NL)
National Institutes of Health
Zero hunger
Openalex Percentile: Top 1%
Medical Image Segmentation Techniques
11.21
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.