AlphaGo, Deep Learning, and the Future of the Human Microscopist

• Published in: Archives of pathology & laboratory medicine (1976) Vol. 141; no. 5; pp. 619 - 621
• Main Authors: Granter, Scott R, Beck, Andrew H, Papke, Jr, David J
• Format: Journal Article
• Language: English
• Published: United States College of American Pathologists 01.05.2017
• Subjects: Academic libraries; Accuracy; Algorithms; Analysis; Artificial intelligence; Birds; Breast cancer; Computational Biology - methods; Computational Biology - trends; Computers; Deep Learning; Games; Games, Recreational; Humans; Humor; Microscopy - methods; Microscopy - trends; Neural networks; Pathology; Pathology, Clinical - methods; Pathology, Clinical - trends; Pattern recognition; Software; Studies; Websites; Womens health
• ISSN: 0003-9985; 1543-2165; 1543-2165
• DOI: 10.5858/arpa.2016-0471-ED

In March of last year, Google's (Menlo Park, California) artificial intelligence (AI) computer program AlphaGo beat the best Go player in the world, 18-time champion Lee Se-dol, in a tournament, winning 4 of 5 games.1 At first glance this news would seem of little interest to a pathologist, or to anyone else for that matter. ''2 To beat Se-dol, Google's AlphaGo program used artificial neural networks that simulate mammalian neural architecture to study millions of game positions from expert human-played Go games. In a simple regression fit, we might determine a line that predicts an outcome y given an input x. With increased computational power, machine learning algorithms are able to fit a huge number of input variables (for example, moves in a game of Go) to determine a desired output (maximizing space gained on the Go board). On a practical level, there are financial barriers to incorporating slide scanners and computers into pathology workflow,6 although presumably hospitals would undertake these steps if computers could improve diagnostic accuracy or increase the efficiency of pathologists. Levenson et al10 showed that pigeons can be trained to distinguish malignant from benign breast tissue with 85% accuracy for individual birds and with an impressive 99% accuracy for a flock consensus. Currently, development of state-of-the-art computer vision algorithms requires millions of training images.11 For AlphaGo, it was feasible to generate millions of example games, but establishing a whole-slide image database of millions of images is currently not practical. Systematic analysis of breast cancer morphology uncovers...