02-518/718 Computational Medicine
02-518/718 COURSE PROFILE
|Special Permission Required? (If yes, see “Notes:)||No|
|Course Relevance (who should take this course?)||
Computational biology Majors
Computational Biology Minors
Biological Sciences Majors
Computer Science Majors
Computer Science Minors
Artificial Intelligence Majors
Biomedical Engineering Majors
Biomedical Engineering Minors
Anyone interested in the application of Machine Learning and Artificial Intelligence to Medicine
Students who complete the course successfully will be able to:
1) Explain the core computational challenges in medicine
2) Explain common clinical data types, and what they are used for
3) Discuss and critique computational methods used to address these challenges
4) Use software to analyze clinical data and build predictive models
5) Learn to independently research a computational medicine challenge using online resources
6) Learn and apply concepts from supervised and unsupervised machine learning as applied to clinical problems.
Basic familiarity with cell and molecular biology (e.g., 03-121 Modern Biology or Biology Advanced Placement), including the central dogma, Mendelian genetics, cell structure, biochemical reactions, and the cell cycle
Familiarity with core concepts of Machine Learning (prerequisite: 10-401, or 10-601, or 10-701), including supervised learning and unsupervised learning.
Grades are determined 80% by homeworks, 20% by a final project report.
Four homeworks are assigned to cover the material in each of the four major modules in the course (Phenotyping; Biomarker Discovery; Statistical Modeling; Causal modeling).
Most assignments involve the analysis of provided data sets using open-source software. No programming is required.
All assignments are submitted electronically through Canvas. 3 late days total can be used without penalty during the semester. There are no partial late days.
Late homeworks will not otherwise be accepted unless you have made prior arrangements for an extension. Please note that extensions will only be granted under exceptional circumstances.
Students define and complete a final project that addresses a computational challenge in medicine.
The definition and scope of the project are determined in consultation with the instructor.
Students are free to use existing software/libraries to complete the course project.
A 10-12 page report describing the project and its outcomes is submitted at the end of the semester.
|Most Recent Syllabus||Link|
|Pre-reqs, Cross List, Related||Pre-requisite or co-requisite: 10-401 or 10-601 or 10-701|
The course is largely self-contained, but does assume some basic familiarity with core concepts from Machine Learning (classification; regression; clustering)
There is a graduate section of the course, 02-718. Lectures for 02-518 and 02-718 are held at the same time. Students in 02-718 do an extra assignment and are required to give an in-class presentation on their final project at the end of the semester.
|Updated July 2019|