Cautious Artificial Intelligence Improves Outcomes and Trust by Flagging Outlier Cases.

Journal: JCO clinical cancer informatics
Published Date:

Abstract

PURPOSE: Artificial intelligence (AI) models for medical image diagnosis are often trained and validated on curated data. However, in a clinical setting, images that are outliers with respect to the training data, such as those representing rare disease conditions or acquired using a slightly different setup, can lead to wrong decisions. It is not practical to expect clinicians to be trained to discount results for such outlier images. Toward clinical deployment, we have designed a method to train cautious AI that can automatically flag outlier cases.

Authors

  • Abhiraj S Kanse
    Department of Electrical Engineering Indian Institute of Technology Bombay, Mumbai, India.
  • Nikhil C Kurian
    Department of Electrical Engineering Indian Institute of Technology Bombay, Mumbai, India.
  • Himanshu P Aswani
    Department of Electrical Engineering Indian Institute of Technology Bombay, Mumbai, India.
  • Zakia Khan
    Independent Researcher, Palatine, IL.
  • Peter H Gann
    Department of Pathology, University of Illinois at Chicago, Chicago, IL, USA.
  • Swapnil Rane
    Department of Pathology, Tata Memorial Centre-ACTREC, HBNI, Mumbai, India.
  • Amit Sethi