UTHealth gets $7.3M for biomed IT study

Machine learning, enhanced medical records among topics of research
By Mike Miliard
11:11 AM
Share

"We’re at the forefront of a transformational era – the health information technology revolution," said Jiajie Zhang, dean of the School of Biomedical Informatics at the The University of Texas Health Science Center at Houston.

[See also: EMR usability seen lacking]

"There’s a huge and growing demand for what we do," added Jiang, in a press statement announcing that UTHealth has been awarded grants totaling $7.3 million for advanced research into biomedical discovery through the use of health IT.

Study at the School of Biomedical Informatics is primarily aimed at new ways to mine medical data for information that could lead to better treatments – and also to making medical records more available and easy to use for caregivers, officials say.

[See also: Object of beauty, or ungainly nuisance?]

With 177 students, UTHealth's school is the only academic program of biomedical informatics in Texas and the only free-standing school of biomedical informatics in the nation. A total of seven grants have been awarded to five principal researchers there since June.

One of the school’s big data researchers, Hua Xu, was awarded money to explore enhance use of elecronic medical records in research – approximately $4.3 million, from a combination of the National Institute of Allergy and Infectious Diseases, the U.S. National Library of Medicine and the National Institute of General Medical Sciences.

"One of the major challenges facing data researchers is how to integrate machine learning algorithms and human wisdom into one system," said Xu, associate professor and holder of the Robert H. Graham Professorship in Entrepreneurial Biomedical Informatics and Bioengineering, in a press statement.

Xu and his team are developing new interactive machine learning algorithms and software for biomedical text processing problems; Xu also applies such advanced text mining approaches to biomedical data management. The research is part of the Data Discovery Index Coordination Consortium, a project that falls under the umbrella of the NIH's BD2K -- it stands for Big Data to Knowledge -- program.

"The consortium will engage the research community to pilot an National Institutes of Health Data Discovery Index, which will catalyze the discoverability, accessibility and citation standards for biomedical big data," said Xu.

Meanwhile, assistant Professor Cui Tao is using a $1.4 million grant from the U.S. National Library of Medicine to develop a software program to make electronic health records more user-friendly for clinicians.

"We want to make it easier for physicians to access the medical histories of their patients," Tao said in a statement. "Some systems may store these patient notes in different places. We’re developing a system that will automatically put this information together.”

This software package is called Temporal Information Modeling, Extracting and Reasoning, or TIMER.

Associate Professor Yang Gong, MD, was awarded a $1.2 million grant for patient safety research by the Agency for Healthcare Research and Quality.

When it comes to investigating patient safety events, health care organizations often follow different procedures, and timely reporting and effective learning from medical incidents are considered useful ways to develop strategies for reducing medical errors.

Utilizing an innovative user-centered, learning-supportive and ontological approach combined with case-based reasoning and natural language processing techniques, Gong’s team is developing a knowledge base and learning toolkit that can systematically collect and analyze incident reports. They link historical reports with previous analytical results.

"The long-term goal is to understand the causes of medical incidents and to develop interventions to reduce the risk of recurrence," he said.

Professor James Langabeer II, meanwhile, was awarded a $240,000 grant from the American Heart Association for a Phase 2 study working to enhance the care of heart disease through an analytic model and database.

Langabeer, who also has an appointment as professor and research director in the Department of Emergency Medicine at UTHealth Medical School, said he is building a large database of all heart attacks to help model and assess outcomes for a public service campaign on the importance of seeking immediate care for the symptoms of a heart attack.

The data will be used to improve emergency cardiovascular care, reduce mortality and improve quality outcomes. Phase 1 of this study resulted in some novel findings that helped to provide a foundation for emergency medical service transport of patients, which were reported in several important clinical journals. The study is being done in conjunction with data from the American College of Cardiology.

Associate Professor Trevor Cohen, MD, received a $160,000 grant from the U.S. National Library of Medicine to contribute to the development of the DDICC.

"It will facilitate dataset search, retrieval and re-use," said Cohen. "These datasets are all publicly available and this search engine will be open to the public."