News Release

Biomedical informatics research group develops teaching/diagnostic software

Peer-Reviewed Publication

Virginia Tech

A multidisciplinary informatics research group working in the Virginia-Maryland Regional College of Veterinary Medicine (VMRCVM) has created a Java-based "Problem List Generator" learning tool that helps veterinary students improve deductive reasoning skills as they learn the art and science of diagnosing disease.

Currently being beta-tested by second year students in the VMRCVM, the software is the "brainchild" of veterinary clinical pathologist Holly Bender and a 16-member interdisciplinary team of faculty members and graduate students from across the university collaborating as the Biomedical Informatics Research Group.

The technology appears so successful the United States Department of Agriculture has funded the group with a Higher Education Challenge Grant to develop a diagnostic training module for food animal diseases using the technology.

Bender began working on the project to improve the way veterinary students learn how to apply knowledge in a problem-solving environment. While first-year students enrolled in the DVM curriculum spend much of their time learning facts, second year students are challenged to begin developing "higher order thinking skills" by applying their knowledge in "case-based" problems.

"This can be a very painful process," explains Bender, an associate professor in the Department of Biomedical Sciences and Pathobiology. "This software facilitates their thinking."

One of the biggest problems she saw in her students was a tendency to jump to diagnostic conclusions on the basis of pre-conceived notions without fully considering all available evidence. To counter that, the Web-based auto-tutorial forces the students to work through a process where they "arrange data abnormalities in a causal hierarchy."

For example, clinical pathologists routinely evaluate blood and urinary chemistry assays that provide information on blood hemoglobin, white blood cells and a range of other parameters.

By systematically identifying all available data as it relates to normative values, hypothesizing about mechanisms responsible for the aberrations, and factoring those into a problem-solving hierarchy, the students build a deductive "argument for the complete pathogenesis of the disease," Bender says.

"Theyare thinking better," says Bender. "They are accounting for their decisions better."

A second project underway is developing a set of 16 Web-based interactive tutorials that describe the physiological processes behind disease states affecting different organ systems.

The Biomedical Informatics Research Group has been meeting once a week for two-and-a-half years and includes faculty, graduate students, and professional students from the Departments of Teaching and Learning, Computer Science, and Accounting and Information Systems, the VMRCVM, and the University of Virginia at Wise.

"I have been struggling with this problem for close to 20 years," says Bender. "It takes the talents of professionals in these key fields to approach the problem with novel solutions." The tool is currently undergoing a formal assessment as the subject of a graduate level instructional technologies course led by Greg Sherman in the Department of Teaching and Learning.

The group has been funded with approximately $200,000 by the University Center for Innovation in Learning, ASPIRES and Reach-Out Grants over the past several years, Bender says, and it appears that university seed funding is beginning to reap dividends.

The $90,000 USDA project is just one example of what Bender believes are many other applications for the new software technology, including continuing education for practicing veterinarians, extending the tool to assist experts in clinical pathology, and improving case analysis by other disciplines including human medicine and other health related fields.

"There is even a faculty member in the Department of Marketing with great interest in our tool," says Bender, adding that the tool will be extended to develop insights into the human learning process itself.

###

PR CONTACT: Jeffrey S. Douglas
540-231-7911
jdouglas@vt.edu



Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.