Artificial intelligence applications in histopathology

Abstract

Histopathology is a vital diagnostic discipline in medicine, fundamental to our understanding, detection, assessment and treatment of conditions such as cancer, dementia and heart disease. Traditionally, the standard workflow in histopathology has primarily relied on the visual interpretation of tissue samples carried out by human experts under a light microscope. Since the 2000s, thanks to advances in scanning technologies such as whole-slide imaging, histopathology is undergoing a digital transformation. The rapid increase in digital data is fuelling the development and application of artificial intelligence (AI) methods. In this Review, we delve into the latest progress in AI methods for histopathology, which promise to yield accurate, scalable, useful and affordable support tools for clinical decision. We examine the challenges and opportunities in this domain, exploring historically important approaches and problems that have shaped the field, while also highlighting recent technological breakthroughs that are poised to redefine its future. Furthermore, we offer an overview of publicly available datasets that have been instrumental in propelling the development of AI methods in histopathology. Increase in clinical digital data is propelling the development and application of artificial intelligence methods in histopathology. In this Review, machine learning algorithms and models and their clinical use cases are discussed, highlighting the computational and operational challenges in the field.