Indica
Backed by cutting edge technology, an uncompromising focus on ease-of-use and dedicated customer service, Indica Labs’ software and services are being used to make vital discoveries in pathology labs and research organizations around the world.

A Deep Learning Convolutional Neural Network Can Differentiate Between Helicobacter Pylori Gastritis and Autoimmune Gastritis with Results Comparable to Gastrointestinal Pathologists

A Deep Learning Convolutional Neural Network Can Differentiate Between Helicobacter Pylori Gastritis and Autoimmune Gastritis with Results Comparable to Gastrointestinal Pathologists

13 January 2022
8:00 – 9:00 PST | 11:00 – 12:00 EST | 16:00 – 17:00 GMT

LEARNING OBJECTIVES

• Learn how to design a clinical study evaluating the performance of a convolutional neural network
• Learn how to train a HALO AI algorithm
• Learn how an AI algorithm has the potential to assist pathologists with generalized training to recognize H. pylori gastritis and autoimmune gastritis
• Learn how a deep learning tool could serve as a learning tool for pathology residents and other possible future applications

This webinar is ideal for infectious disease researchers, pathologists, attending physicians, image analysts, and others working in digital pathology.

ABSTRACT

In this 60-minute webinar, researchers and pathologists from the University of New Mexico will discuss their recently published study using the HALO AI™ deep learning convolutional neural network (CNN) to determine whether a CNN can differentiate autoimmune gastritis from Helicobacter pylori gastritis.

AI has been widely studied in neoplastic pathology, but this technology also holds promise in the realm of inflammatory pathology. In a prior publication, pathologists at the University of New Mexico demonstrated that a CNN can accurately discriminate between the three most common histologic patterns in medical gastric biopsies: H. pylori gastritis, reactive gastropathy, and normal gastric mucosa. In their latest publication, they demonstrate that HALO AI can distinguish between two morphologically similar inflammatory patterns: H. pylori gastritis, and autoimmune gastritis, with equal accuracy of GI pathologists. As many pathologists who lack specialty training in inflammatory GI pathology find it challenging to distinguish these diagnoses, this has the potential to improve clinical practice in the future.

The webinar will include:

1) A brief introduction to the histologic features of H. pylori gastritis, autoimmune gastritis, and normal gastric mucosa
2) An overview of how the HALO AI algorithm was trained and how the test set images were evaluated
3) A detailed description of the results and potential future applications

PRESENTERS

Joshua Hanson, MD
Interim Division Chief of Anatomic Pathology
University of New Mexico

Dr. Joshua Hanson is the Interim Division Chief of Anatomic Pathology at the University of New Mexico in Albuquerque. He specializes in gastrointestinal and liver pathology. His primary research interests are in optimizing surgical pathology diagnostics with digital pathology and AI.

Fred Schultz, MA
Informatics Manager of the Human Tissue Repository
University of New Mexico

Fred Schultz, MA, is the informatics manager of the Human Tissue Repository at the University of New Mexico. He is the designated ‘super-user’ of the HALO platform for digital pathology at UNM.

David Martin, MD
Senior Director of Surgical Pathology
University of New Mexico

Dr. David Martin is the Senior Director of Surgical Pathology at the University of New Mexico in Albuquerque, specializing in gastrointestinal and liver pathology. His main research focus is AI analysis of histologic images for diagnosis and patient prognostics.