A fantastic review came out in the British Medical Journal. Nagendran et al. analyzed the design, reporting standards, risk of bias and claims of studies comparing the performance of diagnostic deep learning algorithms for medical imaging with that of expert physicians. There is a line in their paper that should be on the wall of every company involved with medical A.I.: "The danger is that public and commercial appetite for healthcare AI outpaces the development of a rigorous evidence base to support this comparatively young field." They could only find about a dozen studies and pointed out how many studies are not prospective and have not been validated on real world data. Their results stand as a proof that while the field is new and just started to skyrocket, the same rules must apply to them as to any other medical technologies, no matter how exciting A.I. results sound. There is a long path ahead and we plan to support this mission with two new papers at The Medical Futurist Institute, I hope I can share those with you in no time. Paper: https://lnkd.in/g8Qx-4D #themedicalfuturist #digitalhealth #future #healthcare #medicine #technology #AI #artificialintelligence #deeplearning #machinelearning #study #research …see more