"...although AI’s use will fundamentally alter pathology practice, it is perhaps pathologists’ greatest opportunity to invest themselves more fully in their patients’ care."

Arch Pathol Lab Med. 2019 Jul 3. doi: 10.5858/arpa.2019-0229-ED. [Epub ahead of print]

Regulating Artificial Intelligence for a Successful Pathology Future.

Allen TC1.

Author information

1

From the Department of Pathology, University of Mississippi Medical Center, Jackson.

"Artificial intelligence (AI), a set of techniques aimed at approximating some aspect of human cognition using machines,1 promises to provide pathologists a tool to improve inefficiencies and inaccuracies in diagnosis and laboratory testing, and to facilitate an anticipated smaller number of pathologists’ ability to serve an increasing number of patients with timely diagnoses, even as those diagnoses exhibit an increasingly complicated molecular complexity.2–4 Pathology has long examined the use of AI in

pathologic diagnosis5 and is already using AI. Papanicolaou test imaging, approved by the US Food and Drug Administration (FDA) for screening purposes, is now fully, if not widely, used.6,7 Other areas of cytopathology are investigating the use of AI for diagnosis,8 and AI will undoubtedly play a strong and central role in its continuing progress toward becoming fully automated.9–11 Artificial intelligence in pathology promises to outperform human beings in their assessment of the specific characteristics

necessary for pathologic and radiologic diagnosis.9,12 Despite the initial hysteria,13–16 it is now clear that AI will be, at least for the foreseeable future, a tool for pathologists, not a replacement for  pathologists, and it is evident that although AI’s use will fundamentally alter pathology practice, it is

perhaps pathologists’ greatest opportunity to invest themselves more fully in their patients’ care.3,7 Pathologists’ success with AI, however, will depend to a large degree on the successful implementation of efficient AI governance and regulation."