But how do medical students learn how to diagnose ear infections, as well as the hundreds of other middle ear and ear canal disorders that can be seen with an otoscope?
OtoSim claim that, “95% of graduating medical students do not feel comfortable with their level of otology skills at the end of their clerkship.” To address this problem they developed an artificial ear, in which images of ear disorders are presented under the control of instructors:
But can expert, intelligent systems be trained on similar images, so that they can make diagnoses as well as a human physician? Although the training set of images would likely have to be much larger, I imagine that the answer is that they can.
If so, then uploaded images could be automatically diagnosed at any time, making a mobile, consumer product such as Cellscope’s Otoscope that much more effective for parents, even if they will still have to visit a physician to get a drug prescribed.