Mei Liu, PhD, a computer scientist who uses advanced informatics approaches to improve health care, recently joined NJIT’s College of Computing Sciences as an assistant professor. Her long-term research goal is to develop data-mining methodologies to uncover clinical knowledge from Electronic Medical Records (EMRs) that improves the quality, safety, efficiency and effectiveness of health care.  EMRs have created an unprecedented resource for observational studies as it contains not only detailed patient information but also large amounts of longitudinal clinical data.  Despite the promise of EMR as a research tool, challenges exist for large-scale observational studies.  First, much of the relevant clinical information is embedded in narrative text.  Second, multiple factors conspire to make drawing specific conclusions from EMR data more challenging than data collected specifically to answer research hypotheses.  Thus, it is highly desirable to develop effective and efficient computational methods to mine EMR data for conducting large-scale observational research.

Adverse drug reaction (ADR) for instance, is one of the major causes for failure in drug development.  And severe ADRs that go undetected until the post-marketing phase of a drug often lead to patient morbidity, as exemplified by numerous drug withdrawals.  Currently, she is leading three projects to simultaneously examine ADRs from different angles.  First, she aims to predict ADRs from the chemical, biological, and phenotypic properties of drugs.  Second, she uses laboratory and retrospective medication order data from EMR to ascertain ADRs.  Third, she is exploring the use of natural language processing (NLP) techniques to extract adverse events from the narrative notes in EMR and correlate those events with medications through association mining.  She is also interested in other data mining tasks for clinical informatics, such as drug repurposing (i.e. application of known drugs to new diseases).  She is also interested in using patient medical records to build predictive models for diseases such as diabetes and cancer.

In addition, she plans to combine her experience in bioinformatics and clinical informatics to better understand how molecules function together in achieving a particular clinical outcome.  For instance, she plans to utilize her knowledge in protein/gene interaction networks to study ADRs because a drug acts on a human body by inducing perturbations to biological systems, which involve various molecular interactions such as protein-protein interactions, signaling pathways, and pathways of drug action and metabolism.  She’s especially interested in understanding gene-disease relationships; this understanding will assist in the designs of new drug and therapeutic treatments.

Liu’s preliminary work on ADR detection by mining EMRs is recently published in the 2013 May issue of JAMIA (Journal of American Medical Informatics Association), entitled “Comparative Analysis of Pharmacovigilance Methods in Detection of Adverse Drug Reactions from Electronic Medical Records”. Furthermore, Liu is involved in international collaborations with universities in China to develop machine learning methodologies for intelligent software project risk planning to minimize the impacts of project risks and achieve better project outcome. The collaboration has resulted in two recent publications in DSS (Decision Support Systems) entitled “An Integrative Framework for Intelligent Software Project Risk Planning” in press on December 30, 2012, and “Software Project Risk Analysis using Bayesian Networks with Causality Constraints” in press on Nov. 8, 2012.

She received her doctorate in computer science with a research focus in bioinformatics from the University of Kansas in 2009.  Her dissertation involved the development and optimization of data mining algorithms for understanding protein interactions and protein functions.  She developed a domain-based random decision forest framework (RDFF) to predict protein interactions and a cross-species interacting domain patterns (CSIDOP) approach to discover new functions of proteins.  She also designed a method called K-GIDDI that expanded existing knowledge in protein domain-domain interaction (DDI). Upon receiving her doctorate, she was awarded the National Library of Medicine Postdoctoral Training Fellowship in biomedical informatics, and completed the training in the department of biomedical informatics at Vanderbilt University in 2012.

Last updated:  May 23, 2013