Brian C. Dean, Ph.D.
School of Computing
Contact: 864-656-5866 or firstname.lastname@example.org
Who is Professor Dean?
Brian Dean earned his undergraduate and graduate degrees in computer science from M.I.T., and is currently an associate professor of computer science in the Clemson School of Computing. Brian’s research is in applied algorithms, developing computational techniques and platforms that support the acquisition and analysis of data from a variety of areas of application, primarily biomedical domains. Due to his background in theoretical computer science and applied mathematics, his work uses a wide range of computational methods, such as optimization, machine learning, signal and image processing, and big data analytics. Brian also uses ideas from biology in his computing research, studying “bio-inspired” heuristics for solving large-scale optimization problems. His portfolio of current and past research ranges from fundamental work in biological systems (e.g., protein folding prediction, cellular imaging and mechanical modeling, studying chaotic dynamics in biological systems, and data mining in biological ontologies) through applications in human health (e.g., algorithms for genome-wide association studies, analysis of EEG, MRI, and other biosensor data related to neurological disorders, and modeling radiation in cancer treatment). He currently collaborates with academic and clinical partners at Clemson, MUSC, and the Greenwood Genomics Center.
How Professor Dean's research is transforming health care
Brian is currently involved in several projects that could have significant positive impact on health care; they are briefly summarized below:
- Novel algorithms for genome-wide association studies. He has recently developed sophisticated methods for localizing faults (e.g., “bugs”) in computer software. Exploiting the analogy between software and the human genome, he is now using these methods to perform genome-wide association studies, in which context a “bug” is a mutation that causes a malfunction in genomic execution.
- Developing automated tools for understanding epilepsy. In a long-term collaboration with neurologists at MUSC, he has developed a web-based platform, called EEGnet, for collection and analysis of expert opinion on diagnosing Epilepsy. To date, dozens of neurologists have used this platform to help us analyze data from hundreds of subjects. Goals of the project include: understanding human inter-rater reliability issues and construction of automated machine learning models for diagnosis of Epilepsy from EEG data.
- Network analysis of the human brain. In collaboration with researchers at MUSC, he has designed novel algorithmic methods for analyzing high-resolution network models of the human brain, derived from MRI data. Through analysis of both structural and functional brain networks, he hopes to understand connectivity patterns indicative of Autism, and how these relate to developmental changes in the brain.
Health Research Expertise Keywords
Computational Neuroscience, Bioinformatics, Genome-wide Association Studies, Data Mining, Machine Learning, Medical Imaging, Biological Modeling