AI in the Analytical Phase

Abstract

As one prime example, the variability in the visual inspection of medical images by pathologists is a well-known problem. Both inter- and intra-observer variability may affect image assessment and subsequently the ensuing diagnosis. A large body of work have reported high rates of diagnostic inaccuracy as a result of major discordance among participating physicians with respect to case diagnoses and propose a combination of “routine second opinions” and “directed retrospective peer review”. As most proposed AI-driven solutions for digital pathology focus on the concept of object classification, it appears that algorithmic decision-making may not necessarily contribute to supporting concordance by providing a framework for consensus building. Most capable classification schemes trained with immense effort are supposed to be used for triaging cases in the pathology laboratory (e.g. detecting region-of-interest = ROI), and not for direct and immediate assistance in the pathologist’s office. In contrast, instantly retrieving multiple recently diagnosed cases from a virtual repository with histopathologic and diagnostic similarity to the patient’s biopsy about to be diagnosed offers a new generation of decision support that may even enable “virtual” peer review. This warrants the establishing “centers of excellence” (CoE) that will lead such initiatives and provide the technical and conceptual groundwork before a broader roll-out to the medical community.