Artificial intelligence is only as strong as its ability to properly process aberrations. The Greeks said it best, a chan is only as strong as its weakest link! In healthcare, using AI for diagnosis is a useful tool but FAR from being a replacement for hands-on diagnoses.
Well planned aberrations could also be used to trick medical A.I. vision systems that are designed to identify diseases designed to identify diseases. In that study, the researchers tested deep learning systems with adversarial examples on three popular medical imaging tasks – classifying diabetic retinopathy from retinal images, pneumothorax from chest X-rays, and melanoma from skin photos. In such attacks, pixels within images are modified in a way that might seem like a minimal amount of noise to humans but could trick these systems into classifying these images incorrectly. The scientists noted that their attacks could make deep learning systems misclassify images up to 100 percent of the time and that modified images were imperceptible from real ones to the human eye. They add that such attacks could work on any image and could even be incorporated directly into the image-capture process.
Source: (14) How To Play Tricks On Artificial Intelligence? | LinkedIn