AI and physicians work best in unison

In one of the first studies to investigate how AI can improve physicians’ diagnostic accuracy, the research highlighted the benefit of incorporating AI into practice, while also allaying fears of a complete technological takeover.Published in the April edition of Ophthalmology, the study expanded on previous work from Google AI that donstrated its algorithm performed as well as practitioners in screening patients for diabetic retinopathy.The researchers also examined whether the algorithm could do more than simply diagnose disease, and developed a new computer-assisted syst that could ‘explain’ the algorithm’s diagnosis.According to the study’s authors, previous attpts to utilise computer-assisted diagnosis donstrated that some people were overly reliant on the equipment, leading to th repeating the machine’s errors. Others were under-reliant or ignored the equipment’s recommendations and missed accurate predictions.The Google AI research team believed that some of these pitfalls might be avoided if the computer provided an explanation for its predictions.The team found that this new explanation syst not only improved the ophthalmologists’ diagnostic accuracy, but it also improved algorithm’s accuracy.In testing the theory, the researchers developed two types of assistance to help physicians read the algorithm’s predictions; Grades: a set of five scores that represent the strength of evidence for the algorithm’s prediction; and Grades + heatmap: an improved grading syst which includes a heatmap that measures the contribution of each pixel in the image to the algorithm’s prediction.Ten ophthalmologists were asked to read each image once under one of three conditions: unassisted, grades only, and grades + heatmap. Both types of assistance improved physicians’ diagnostic accuracy. It also improved their confidence in the diagnosis, however, the degree of improvent depended on the physician’s level of expertise.“What we found is that AI can do more than simply automate eye screening, it can assist physicians in more accurately diagnosing diabetic retinopathy,” lead researcher Dr Rory Sayres said.Akin to medical technologies that preceded it, Sayres said that AI was another tool that would make the knowledge, skill, and judgment of physicians even more central.