Over the past few years, deep learning, artificial intelligence (AI) and healthcare AI has become a hot topic owing to development of advanced computer-assisted analytical tools. Pathologists too have embraced this revolution with the recent FDA approval of whole slide scanners. The promise of digital pathology is not merely transfer of analog data from glass slide to a monitor but augment the human eye with information/data to perform intelligent interpretations. AI and deep learning technologies mimic human cognitive capabilities using sophisticated imaging processing techniques to train deep convolutional neural networks. These trained multilayered convolutional networks then perform all complex computational tasks and reveal more information answering queries pertaining to a given disease.
AI/deep learning systems offer further advantages by integrating the workflow into the operational environment that is hardware agnostic, thereby making it a good fit. A larger set of variables can thus be assessed and analyzed by these computer algorithms on the digitized tissue sections. The inevitability of these digital transition AI tools will be particularly beneficial as the industry moves towards value-based payments. Experts in this field suggest that healthcare providers who pay for performance will appreciate the speed and predictive power of AI. Pathology specialty, as prompted by some, will soon metamorphose into a role of “information specialists” to make room for new technologies.
Image analysis algorithms are already being used for automated or semi-automated immunohistochemistry (IHC) quantification, thereby facilitating standardization, speed of analysis and objective reporting to complement the role of the pathologist. Thus, an early intervention towards making AI as an “assisting tool” is combining the human cognitive capabilities and the predictive power of computational algorithms for providing insights into patient specific treatment options.
Intelligent digital pathology will thus bring about a paradigm shift from conventional microscopy to pathology AI.