An article by Michael Pencina, PhD and Lawrence Carin, PhD in JAMA “On Deep Learning for Medical Image Analysis“ explains how a neural network can be trained to recognize complex visual patterns and demystifies the use of convolutional neural networks to analyze medical images. Neural network technology is used to identify pathology in medical images by “learning” from information in prior images. Machine learning can greatly improve a clinician’s ability to deliver medical care.
The online article also includes a short video with Drs. Carin and Pencina and other researchers explaining the concepts in a clear concise manner.
Watch it here. Read the complete article in JAMA .
Michael Pencina is also the Vice Dean for Data Science and Information Technology.