Published August 2, 2018
A five-year, $2.5 million grant awarded last summer to the Department of Biomedical Informatics in the Jacobs School of Medicine and Biomedical Sciences is helping to train a new generation of research leaders in the analysis and interpretation of complex health care datasets.
The T15 award from the National Library of Medicine of the National Institutes of Health supports postdoctoral fellows and junior faculty doing biomedical informatics research. The fellowships are designed to meet a growing need for investigators trained in biomedical computing, data science and related fields with applications in health care clinical informatics, translational bioinformatics and clinical research informatics.
“The purpose of the grant is to expose junior faculty to the field of biomedical informatics and to increase collaboration between biomedical informatics and other disciplines,” said Peter Elkin, MD, professor and chair of the Department of Biomedical Informatics and director of UB’s Clinical and Translational Science Institute informatics core. “These are NIH-funded eight- to 12-week immersed training experiences which can be performed in the summer, and we’re in the process of accepting applications now.”
Michael Dwyer, PhD, assistant professor of neurology and biomedical informatics and technical imaging director of the Buffalo Neuroimaging Analysis Center, was one of the first recipients. His three-month research project was an extension of his ongoing work in measuring brain atrophy in neurological diseases like multiple sclerosis. Research tools to measure progressive loss of brain tissue are not really feasible in clinical routine for individual patients and, according to Dwyer, “methods require highly optimized imaging that isn't often done, and that there don't exist large-scale adaptable datasets capable of providing individualized comparatives and predictives.”
In his fellowship, he addressed these limitations with new computational approaches and the establishment of a 10,000-scan normative database combining clinical and imaging data.
“I think one of the most exciting things about the biomedical informatics field and about the program at UB is its real-world, multi-discliplinary focus,” said Dwyer. “Many times, when we are working on projects, there's a tendency to get a bit ‘silo-ed’ — to lose the forest for the trees — and this is especially likely when everyone working together has the same background.”
In the Biomedical Informatics (BMI) department, he said, “there are clinicians, bio-statisticians, computer scientists, geneticists, ontological philosophers and public health experts all working hand-in-hand and interacting on a daily basis. Everyone brings something unique to the table, and there is really a sense when working together that you are approaching the problem from many angles at the same time in a way that you wouldn't be able to do alone.”
Dwyer said the experience also allowed him to get up to date with the most recent advances in deep learning techniques, and to talk with other faculty and students using those methods on real-world problems.
“For me, the fellowship itself has led to a longer-term collaboration that I think will really help foster more and better translational neuroimaging research.”
Dwyer advises potential applicants to identify a clear project goal, “to provide a central reference point” for discussions about the breadth of knowledge and synergies in the field. “BMI is all about continual improvement, so try to learn as much as you can about other people's work and build professional relationships that can lead to future research and collaboration.”
The Department of Biomedical Informatics is currently accepting applications for three short-term (three-month) fellowships per grant year. For more information, contact Cheryl Kennedy at (716) 888-4847 or firstname.lastname@example.org.