MIT Department of Electrical Engineering & Computer Science

EECS Spring 1995 Catalogue Supplement

6.893 Medical Computing (H)

MW 2:30-4, 24-307
3-0-9
Prof. Peter Szolovits, NE43-416, x3-3476, psz@mit.edu

Medicine is (finally) in the middle of an information processing revolution. Dramatic improvements in computer technology are making possible comprehensive record-keeping systems, automated laboratories and therapeutic devices, and sophisticated visualization techniques. Societal demands for cost-effective care result in vast collections of clinical outcomes data that provide opportunities to learn from experience. New organizational structures for providing health care (such as HMOs) and an increased importance of the role of the patient in his or her health care provide additional new opportunities for innovative technical contributions.

This class will analyze the computational needs of clinical medicine, review the history of interesting systems and approaches that have begun to support those needs, and present the computer technologies that appear most applicable to contributing to revolutionary advances. We will also have participation of doctors from those institutions who are currently developing new systems and applications.

Some of the homework and a final paper wil be based on innovative applications of sophisticated computing techniques to real clinical data. We will explore topics in automated diagnosis, monitoring, detection of errors in treatment, learning new models from data, personalizing explanations to specific patients, etc. We will also examine the architectures of recently built comprehensive clinical information systems.

Students are expected to have a flexible competence in programming, because assignments will involve use of diverse databases and programming languages. A knowledge of artificial intelligence at the level of 6.034 will also be assumed, as many AI techniques will find application in the class material. The class is open to both undergraduate and graduate students, and will qualify for H-level graduate credit. We will use the text Medical Informatics: Computer Applications in Health Care, Shortliffe, et. al., Addison-Wesley, 1990 and additional handouts.