Putting the "big" in Big Data: Learning to Be Just as (Un)certain as a Clinician at EEG

Abstract

The deployment of machine learning to augment clinical judgment is an exciting avenue in the effort to improve patient care. In basic terms, machine learning can be thought of as an algorithm that learns how to distinguish patterns within data, and it can learn better with larger amounts of data. Distinguishing patterns is at the heart of many aspects of clinical care. EEG interpretation is a prime example: a trained, laborious task that might benefit from algorithms. The challenge, however, is to build an algorithm that is reliable enough to be useful and trustworthy in standard clinical care.