CHiP’s Doctoral Research Assistant, Yu Deng, Receives Prize at Northwestern’s Biomedical Data Science Day
We’re proud to announce that CHiP’s Doctoral Research Assistant, Yu Deng, received first prize for the poster competition at Northwestern University’s Second Annual Biomedical Data Science Day earlier this month. Yu’s project, “Use of Clinical Phenotypes and Non-negative Tensor Factorization for Heart Failure Prediction,” utilizes machine learning to identify patients who are at high risk of heart failure using Electronic Health Record (EHR) data. According to Yu, EHR data is high dimensional, sparse and noisy, and there are few algorithms that can be used to efficiently yield results using that data. Through non-negative tensor factorization, a type of statistical analysis, Yu and her research team were able to generate more understandable and concise information, or phenotypes, from EHR data. Physicians then evaluated those phenotypes, and determined that they were clinically meaningful.
Yu’s project is part of the larger High-Throughput Phenotyping on EHRs Using Multi-Tensor Factorization project at CHiP, which is funded by the National Science Foundation. CHiP’s Director, Abel Kho, is the co-principal investigator.
For more information on Yu’s project, read the abstract below.