In this section, Professor Verghese describes his research interests, sharing that the class often fuels his work in the field with new questions.
Of particular interest to me is adapting and simplifying dynamic models from physiology so that model parameters can be estimated from available data, and so the models can facilitate inferences about important unmeasured signals from the available measured signals.
— George Verghese
For the past dozen years I have mostly worked in biomedical signal processing, for data collected in clinical and ambulatory settings. Of particular interest to me is adapting and simplifying dynamic models from physiology so that model parameters can be estimated from available data, and so the models can facilitate inferences about important unmeasured signals from the available measured signals. (Many of the dynamic models have electrical circuit analogs, which can make them especially accessible to Electrical Engineering and Electrical Engineering and Computer Science students.) All this fits very well with the themes of 6.011 Signals, Systems and Inference, and I routinely discuss with my class such examples originating in research. I also bring in application examples from other fields, as opportunity arises. These various examples are motivating for the students, as they illustrate the relevance of the course material. It is also almost invariably the case that each time I lecture the subject, I encounter new questions and ideas to carry back to my research!
Electrocardiogram data, an example of measured signals. (Image courtesy of kenfagerdotcom on flickr. License: BY-NC-SA.)