Rise of the machines.

We talk a lot about prognostic models and how to keep up with the tidal wave of genomic markers coming our way. It seems big data computational analysis is the only way forward when our ability to produce data far exceeds our ability to analyze it. This week’s JAMA viewpoint discusses some of the unforeseen consequences of the inevitable influx of artificial intelligence (AI) into our daily lives, with specific shout-outs to fields with heavy focuses on algorithms and pattern recognition (looking at you oncology, radiology, and pathology). Major concerns include: deskilling more discriminate physicians, misinterpretations resulting from deficient contextualization (i.e., gestalt), and a growing chasm between our push for accuracy and our capacity to interpret the meaning of our results. But with AIs now enabling electronic medical records to provide a window to the future as well as the past, machine learning is undoubtedly here to stay. Luckily, at least for the time being, so is the value of a doctor’s humanity.


Popular Posts