EECS Special Seminar

Dennis A. Dean, II, Ph.D.
Graphs, Languages, Automation and other Approaches for Understanding Sleep and Public Health Outcomes
Harvard Medical School
Glennan 313
2:00 - 3:00 PM
April 16, 2014

The sleep-wake and circadian (internal time keeping) systems directly and indirectly influence nearly all bodily functions, including cognitive performance/behavior and cardiovascular outcomes. Traditionally, dynamical models have been used to predict sleep-wake and circadian states and their effects on outcome measures from impatient studies.  A growing need in sleep and circadian biology is to apply models and methods developed experimentally to field and occupational settings, and to the analysis of existing epidemiological scale data sets. This requires the development of analytical methods that are appropriate for finding solutions to a range of applied problems. To address this need, problem specification methods from formal language theory in computer science are integrated with traditional analysis approaches. The resulting algorithms allow robust and efficient incorporation of sleep and circadian principles in applied problems where these systems are strong determinants of important outcomes. Specific applications of these algorithms will include designing crew schedules for NASA and analyzing individual cortisol pulsatility. Recent work in automating epidemiological scale analysis of sleep studies will be presented to motivate opportunities to advance sleep research with informatics and engineering principles.  Specific examples will include automating spectral analysis of the electroencephalogram (EEG) signal and extracting novel indices from plethysmography.


Dr. Dean is a postdoctoral fellow in the Program in Sleep Epidemiology and the Program in Sleep and Cardiovascular Medicine at Brigham and Women’s Hospital and Harvard Medical School, under the direction of Dr. Susan Redline. His research is directed at elucidating the physiological relationship between sleep and hypertension by integrating objective statistical model building techniques with traditional sleep data analysis techniques. Dr. Dean is collaborating with Brown University’s Sleep for Science Laboratory to study the relationship between ADHD, obstructive sleep apnea, and plethysmography. Collectively, Dr. Dean’s postdoctoral research targets the development of ‘Big Data’ techniques that can be applied to large databases of sleep studies in order to identify associations of sleep with public health outcomes.

Dr. Dean has fifteen years of scientific software tool development experience and is skilled in developing automated analyses that include formal language methods and mathematical modeling techniques. He obtained a bachelor’s of science in Computer Science from SUNY-Empire State and a Master’s of Science in Computer Science with an emphasis in Machine Learning from the University of Massachusetts-Lowell. He is a graduate of the University of Massachusetts intercampus PhD program in Biomedical Engineering and Biotechnology and completed a certificate in applied statistics sponsored by the Harvard Catalyst. Dr. Dean’s research is funded by the Heart, Lung and Blood Institute at the National Institutes Health and the Periodic Breathing Foundation.