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

Journal: Medical image analysis
PMID: 39541706

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 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

  • Zheyuan Zhang
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University.
  • Elif Keles
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University.
  • Gorkem Durak
  • Yavuz Taktak
    Department of Internal Medicine, Istanbul University Faculty of Medicine, Istanbul, Turkey.
  • Onkar Susladkar
    Vishwakarma Institute of Information Technology, Pune, India.
  • Vandan Gorade
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Debesh Jha
    Department of Information and Communication Engineering, Chosun University, 309 Pilmun-Daero, Dong-Gu, Gwangju 61452, Republic of Korea.
  • Asli C Ormeci
    Department of Internal Medicine, Istanbul University Faculty of Medicine, Istanbul, Turkey.
  • Alpay Medetalibeyoglu
  • Lanhong Yao
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University.
  • Bin Wang
    State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau, Northwest A&F University, Yangling 712100, China; New South Wales Department of Primary Industries, Wagga Wagga Agricultural Institute, Wagga Wagga 2650, Australia. Electronic address: bin.a.wang@dpi.nsw.gov.au.
  • Ilkin Sevgi Isler
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA; Department of Computer Science, University of Central Florida, Florida, FL, USA.
  • Linkai Peng
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Hongyi Pan
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Camila Lopes Vendrami
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Amir Bourhani
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Yury Velichko
    Department of Radiology, Feinberg School of Medicine, Northwestern University, 737 N. Michigan Ave, Suite 1600, Chicago, IL, 60611, USA.
  • Boqing Gong
    Google Research, Seattle, WA, USA.
  • Concetto Spampinato
    Department of Computer and Telecommunications Engineering, University of Catania, Catania, Italy.
  • Ayis Pyrros
    DuPage Medical Group, Radiology. Electronic address: ayis@ayis.org.
  • Pallavi Tiwari
    Department of Radiology, University of Wisconsin, Madison, WI, USA.
  • Derk C F Klatte
    Department of Gastroenterology and Hepatology, Amsterdam Gastroenterology and Metabolism, Amsterdam UMC, University of Amsterdam, Netherlands; Department of Radiology, Mayo Clinic, Jacksonville, FL, USA.
  • Megan Engels
    Department of Gastroenterology and Hepatology, Amsterdam Gastroenterology and Metabolism, Amsterdam UMC, University of Amsterdam, Netherlands; Department of Radiology, Mayo Clinic, Jacksonville, FL, USA.
  • Sanne Hoogenboom
    Department of Gastroenterology and Hepatology, Amsterdam Gastroenterology and Metabolism, Amsterdam UMC, University of Amsterdam, Netherlands; Department of Radiology, Mayo Clinic, Jacksonville, FL, USA.
  • Candice W Bolan
  • Emil Agarunov
    Division of Gastroenterology and Hepatology, New York University, NY, USA.
  • Nassier Harfouch
    Department of Radiology, NYU Grossman School of Medicine, New York, NY, USA.
  • Chenchan Huang
    Department of Radiology, NYU Grossman School of Medicine, New York, NY, USA.
  • Marco J Bruno
    Departments of Gastroenterology and Hepatology, Erasmus Medical Center, Rotterdam, Netherlands.
  • Ivo Schoots
    Department of Radiology, The Netherlands Cancer Institute-Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands; Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, The Netherlands.
  • Rajesh N Keswani
    Northwestern University, Chicago, Illinois, USA.
  • Frank H Miller
    Machine & Hybrid Intelligence Lab, Department of Radiology, Northwestern University, Chicago, USA.
  • Tamas Gonda
    Division of Gastroenterology and Hepatology, New York University, NY, USA.
  • Cemal Yazici
    Division of Gastroentrrology and Hepatology, University of Illinois Chicago, Chicago, Illinois.
  • Temel Tirkes
    Department of Radiology and Imaging Sciences, Indiana University School of Medicine, 550 N. University Blvd. Suite 0663, Indianapolis, IN, 46202, USA. atirkes@iu.edu.
  • Baris Turkbey
    Molecular Imaging Program, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA.
  • Michael B Wallace
  • Ulas Bagci
    Department of Radiology, Feinberg School of Medicine, Northwestern University, 737 N Michigan Ave, Ste 1600, Chicago, IL 60611.

Keywords

External Resources

Popular Topics
Recent Journals