Short Course 51: Adopting Digital and Computational Pathology in Clinical Practice: A Practical, Technical, Regulatory and Legal Primer

Abstract

Despite there being two FDA-cleared commercial whole slide imaging systems for primary diagnosis, very few pathology practices have rushed to adopt a fully digital workflow. Frequently cited barriers to the adoption of digital pathology include increased cost, potential disruptions to workflow, decreased productivity, regulatory ambiguity, and a general skepticism about the technology. These barriers have contributed to a lukewarm reception for what was commonly considered a significant milestone in the practice of anatomic pathology. Similar concerns have been raised about the application of artificial intelligence (AI) to anatomic pathology (a.k.a. computational pathology) which also elicits a fear of computer algorithms replacing human pathologists.

Implementing a digital pathology system for clinical use is not plug and play. There are many infrastructures, hardware, software, interface, digital storage, physical space, human factors, computational power, and organizational needs that one must take into consideration prior to the selection, purchase, and implementation of these systems. Unfortunately, pathology training programs rarely address these topics in any depth, if at all. This has left the vast majority of our profession, including the generation currently in training, without a basic fund of knowledge to draw upon when considering digital pathology and AI.

The overall goal of this course is to demystify the adoption of digital pathology and AI. This course will provide a well-integrated and comprehensive curriculum that starts with the fundamentals of digital pathology and AI and builds up knowledge to clinical implementation. The course content is targeted to practicing pathologists and will prepare them to participate in the evaluation and implementation of a digital pathology system for clinical practice.

Learning Objectives:

1. Understand the components required to go fully digital in anatomic pathology, including gross imaging, whole slide imaging, LIS integration, telepathology, and artificial intelligence platforms, including use cases for different types of pathology practices
2. Recognize technical, workflow, financial, regulatory, legal, and human barriers affecting adoption of digital workflows in anatomic pathology and how to overcome or mitigate them
3. Understand the basics of artificial intelligence as applied to the practice of pathology, the role of the pathologist in developing artificial intelligence algorithms, and what will be required to implement artificial intelligence in clinical practice