HST-951 Medical Decision Support
Grad H, Fall term, 3-0-9
Prerequisites: 6.034
, HST-947 or permission from the instructors; programming skills
The page has moved. In case it does not redirect immediately, please click
here
Course
Schedule and Administrivia
Handouts
Time/Location: Mondays/Wednesdays 2:30pm--4pm, in
24-407.
Course Description
The course objective is to present data-driven modeling in terms of
- Theory - how it works
- Practicalities - when to apply
- Implementation - how to apply
We consider two types of data-driven modeling: predictive modeling,
and creation of hypotheses. The former we will relate to medical informatics,
while the latter is an essential part of bioinformatics.
We will present the main concepts of decision analysis, machine learning,
and model construction and evaluation in the specific context of biomedical applications.
Included are some specific methodologies such as:
- Support vector machines
- Rough and fuzzy sets
- Neural networks
- Sequence alignment
- Haplotype tagging
- Bagging, boosting
Students
produce a final project using the methods learned in the subject, based
on actual clinical data. (Required for students in the Master's Program
in Medical Informatics, but open to other graduate students and advanced
undergraduates.)
L. Ohno-Machado,
P. Szolovits,
S. A. Vinterbo
Some examples of projects developed in this course, by domain:
-
Anonymization of databases
-
Diagnostic models
-
Prognostic models
Last updated 9/7/05