Skip to main content

Prof. Joydeep Ghosh Receives Two NSF Grants for Work with Complex Data

UT ECE professor Joydeep Ghosh has received two research awards from the National Science Foundation (NSF) totalling more than $1 Million focusing on topics in complex data modeling in the healthcare field..

Prof. Ghosh was awarded over $660,000 to develop a computational framework for semi-automatic, high-throughput phenotyping of EHR at UT Austin. His approach employs machine learning primarily based on multi-tensor factorization. Prof. Ghosh’s collaborators include Abel Kho from Northwestern University at Chicago, Bradley Malin and Joshua Denny from Vanderbilt University Medical Center and Jimeng Sun from the Georgia Institute of Technology. The total award across the four institutions exceeds $2 million.

The goal of Prof. Ghosh’s research is to model data as multiple, interconnected relationships, such as the relationship between a patient and their medication and diagnosis, or a patient and their symptoms. His research team will develop scalable algorithms to analyze these relationships and derive hidden concepts from the available data. Clinical experts will refine these concepts into specific phenotypes.

Prof. Ghosh also received a grant of nearly $500,000 to develop scalable ordering and ranking algorithms for complex data using the concept of Monotonic Retargeting (MR). Using tools from convex optimization, function approximation and stochastic learning theory, Prof. Ghosh will create a framework to establish efficient solutions for different classes of associated learning problems. This framework will affect ordering, ranking and determine top choices for recommendation, multi-label classification and multi-dimensional isotonic regression.

This research has widespread applications in the study of diseases, where only a small number of genes associated with certain illnesses are known. Yet several sources of information on gene-to-gene relations exist, such as co-expression and protein interactions. The question becomes how to prioritize, simultaneously and for each disease, a small number of additional genes associated with the illness.