Image-Based Machine Learning Algorithms for Disease Characterization in the Human Type 1 Diabetes Pancreas.

Journal: The American journal of pathology
Published Date: Mar 1, 2021

Abstract

Emerging data suggest that type 1 diabetes affects not only the β-cell-containing islets of Langerhans, but also the surrounding exocrine compartment. Using digital pathology, machine learning algorithms were applied to high-resolution, whole-slide images of human pancreata to determine whether the tissue composition in individuals with or at risk for type 1 diabetes differs from those without diabetes. Transplant-grade pancreata from organ donors were evaluated from 16 nondiabetic autoantibody-negative controls, 8 nondiabetic autoantibody-positive subjects with increased type 1 diabetes risk, and 19 persons with type 1 diabetes (0 to 12 years' duration). HALO image analysis algorithms were implemented to compare architecture of the main pancreatic duct as well as cell size, density, and area of acinar, endocrine, ductal, and other nonendocrine, nonexocrine tissues. Type 1 diabetes was found to affect exocrine area, acinar cell density, and size, whereas the type of difference correlated with the presence or absence of insulin-positive cells remaining in the pancreas. These changes were not observed before disease onset, as indicated by modeling cross-sectional data from pancreata of autoantibody-positive subjects and those diagnosed with type 1 diabetes. These data provide novel insights into anatomic differences in type 1 diabetes pancreata and demonstrate that machine learning can be adapted for the evaluation of disease processes from cross-sectional data sets.

Authors

  • Xiaohan Tang
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida; Department of Metabolism and Endocrinology, The Second Xiangya Hospital, Central South University, Changsha, China.
  • Irina Kusmartseva
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Shweta Kulkarni
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Amanda Posgai
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Stephan Speier
    Paul Langerhans Institute Dresden of the Helmholtz Zentrum München at the University Clinic Carl Gustav Carus of Technische Universität Dresden, Helmholtz Zentrum München, Neuherberg, Germany; Institute of Physiology, Faculty of Medicine, Technische Universität Dresden, Dresden, Germany; German Center for Diabetes Research, München-Neuherberg, Germany.
  • Desmond A Schatz
    Department of Pediatrics, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Michael J Haller
    Department of Pediatrics, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Martha Campbell-Thompson
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida; J. Crayton Pruitt Family Department of Biomedical Engineering, University of Florida, Gainesville, Florida.
  • Clive H Wasserfall
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida.
  • Bart O Roep
    Department of Diabetes Immunology, Diabetes and Metabolism Research Institute, City of Hope/Beckman Research Institute, Duarte, California.
  • John S Kaddis
    Department of Diabetes Immunology, Diabetes and Metabolism Research Institute, City of Hope/Beckman Research Institute, Duarte, California; Department of Diabetes and Cancer Discovery Science, Diabetes and Metabolism Research Institute, City of Hope/Beckman Research Institute, Duarte, California.
  • Mark A Atkinson
    Department of Pathology, Immunology and Laboratory Medicine, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida; Department of Pediatrics, College of Medicine, University of Florida Diabetes Institute, Gainesville, Florida. Electronic address: atkinson@ufl.edu.

Keywords

External Resources

Popular Topics
Recent Journals