Digital Pathology and Artificial Intelligence in Pediatric Pathology.

Journal: Surgical pathology clinics
Published Date:

Abstract

Applications of artificial intelligence (AI) and machine learning (ML) are rapidly developing to support the diagnosis and classification of pathology specimens. These tools rely on digitization of pathology glass slides as whole slide images, allowing computers to interpret image information. Tools to support the evaluation of pediatric pathology specimens have been slower to develop, in part because specimens that can be used to train these tools are less common. Here, we selectively highlight diagnostic and prognostic applications of AI and ML tools uniquely designed to support the evaluation of neoplastic and non-neoplastic pediatric pathology specimens.

Authors

  • Nakul Shankar
    Department of Pathology, University of Colorado, Boulder, USA.
  • Portia A Kreiger
    Division Chief, Anatomic Pathology, Children's Hospital of Philadelphia; Clinical Pathology & Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania, 3401 Civic Center Boulevard, 5NW10, Philadelphia, PA 19104, USA.
  • Derek A Oldridge
    Clinical Pathology & Laboratory Medicine, Perelman School of Medicine, University of Pennsylvania, 3401 Civic Center Boulevard, 5NW10, Philadelphia, PA 19104, USA; Department of Pathology and Laboratory Medicine, Children's Hospital of Philadelphia, 3501 Civic Center Boulevard, CTRB 9028, Philadelphia, PA 19104, USA.
  • Jennifer Picarsic
    Children's Hospital of Pittsburgh of University of Pittsburgh Medical Center, Pittsburgh, PA, USA.
  • Michael A Arnold
    Department of Pathology University of Colorado, 12631 East 17th Avenue, Mail Stop B216, Aurora, CO 80045, USA; Department of Pathology and Laboratory Medicine, Children's Hospital Colorado, Box 120, 13123 East 16th Avenue, Aurora, CO 80045, USA. Electronic address: Michael.Arnold@childrenscolorado.org.