University Park, Pa.-based Penn State University has received a $599,883 grant from the National Science Foundation to develop machine learning algorithms that can better analyze messy healthcare data.
The researchers had been studying EHR data to predict disease risk or progression, but noticed it was filled with gaps because the info isn't collected at regular intervals and there are so many variables at play.
The scientists plan to use artificial intelligence techniques that can make more accurate medical predictions through methods such as flexible modeling of longitudinal data and learning of complex correlations.
They intend to explore the use of a deep learning model called "deep kernel gaussian process regression" that can "discover the patterns between unobserved or hidden states," according to the Oct. 11 university news release.