Clinical Predictive Modeling

Clinical care involves making many predictions under uncertainty, including risk assessment, diagnosis, prognosis and therapeutic management. The better those predictions can be made, the better clinical care is likely to be. The increasing availability and richness of electronic health records (EHRs) are increasing the opportunities for developing and deploying computer-based clinical prediction methods. Such methods can serve as key components of computer-based decision support systems. The data in EHRs can be used to construct prediction models using machine learning methods. Individual patient data from EHRs can also serve as input to the predictions models.

Gregory F. Cooper, M.D., Ph.D. and Shyam Visweswaran, M.D., Ph.D. are leading projects to apply artificial intelligence, machine learning, and Bayesian modeling to develop clinical prediction models from data. These projects are currently developing predictive models from both clinical and genome-wide data using Bayesian statistics and machine learning methods. Bayesian methods are especially well suited for combining prior knowledge (e.g., from the literature) with current data (e.g., from high throughput experiments) to derive predictive models.

These projects are also developing patient-specific prediction models. In contrast to population-based models that are constructed to perform well on average on all future patient cases, patient-specific models are optimized to predict well for a particular patient case under consideration.

Sample of Related Publications:

Visweswaran S, Cooper GF. Instance-specific Bayesian model averaging for classification. In: Advances in Neural Information Processing Systems (NIPS) (2004) 1449-1456.

Cooper GF, Abraham V, Aliferis CF, Aronis JM, Buchanan BG, Caruana R, Fine MJ, Janosky JE, Livingston G, Mitchell T, Monti S, Spirtes P. Predicting dire outcomes of patients with community acquired pneumonia. Journal of Biomedical Informatics (2005) 347-366.

Visweswaran S, Cooper GF. Patient-specific models for predicting the outcomes of patients with community acquired pneumonia. In: Proceedings of the Fall Symposium of the American Medical Informatics Association (2005) 759-763.

Visweswaran S, Angus DC, Hsieh M, Weissfeld L, Yealy D, Cooper, GF. Learning patient-specific predictive models from clinical data. Journal of Biomedical Informatics (2010) 669-685.

Cooper GF, Hennings-Yeomans P, Visweswaran S, Barmada M. An efficient Bayesian method for predicting clinical outcomes from genome-wide data. In: Proceedings of the Fall Symposium of the American Medical Informatics Association (November 2010).

Visweswaran S, Cooper GF. Learning instance-specific predictive models. Journal of Machine Learning Research (to appear).