Developing image analysis methods for digital pathology

Abstract

The potential to use quantitative image analysis and artificial intelligence is one of the driving forces behind digital pathology. However, despite novel image analysis methods for pathology being described across many publications, few become widely adopted and many are not applied in more than a single study. The explanation is often straightforward: software implementing the method is simply not available, or is too complex, incomplete, or dataset‐dependent for others to use. The result is a disconnect between what seems already possible in digital pathology based upon the literature, and what actually is possible for anyone wishing to apply it using currently available software. This review begins by introducing the main approaches and techniques involved in analysing pathology images. I then examine the practical challenges inherent in taking algorithms beyond proof‐of‐concept, from both a user and developer perspective. I describe the need for a collaborative and multidisciplinary approach to developing and validating meaningful new algorithms, and argue that openness, implementation, and usability deserve more attention among digital pathology researchers. The review ends with a discussion about how digital pathology could benefit from interacting with and learning from the wider bioimage analysis community, particularly with regard to sharing data, software, and ideas. © 2022 The Author. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Hourglass, a tool to mine bioimaging data, uncovers sex-disparities in the IL-6-associated T cell response in pancreatic tumors

“…Digital image analysis, as opposed to semi-quantitative scoring, is becoming a common standard and has significant benefits in terms of larger scale experiments such as a wide staining panel applied to a TMA, normalization and implementation on analysis standards across labs and institutions, and cost-effectiveness. (33)(34)(35)(36) Hourglass was designed as a user-friendly and robust analysis tool to systematically and reproducibly mine large bioimaging datasets in the rapidly expanding digital pathology space. Naturally, alongside the boom of big data in biomedical and clinical spaces, tools are constantly created to query public and private datasets.…”

Reciprocal effects of mTOR inhibitors on pro-survival proteins dictate therapeutic responses in tuberous sclerosis complex

“…Ki67 and Cleaved Caspase 3 staining was quantified in a blinded fashion using QuPath software on 3-4 non-overlapping fields per tumor. 65 …”

Recent Advances in Pathology: the 2022 Annual Review Issue of The Journal of Pathology

“…The second review, from Peter Bankhead in Edinburgh, UK, focuses on image analysis methods, highlighting the current disconnect between the existing possibilities for digital pathology and the reality of using sophisticated image analysis systems in a coherent, consistent manner. In addition to reviewing the basic approaches, the article discusses the challenges of developing a novel algorithm past proof‐of‐concept and the need for collaborative and multidisciplinary approaches, with openness and sharing as a key principle for the future [2].…”