Meeting times
- Tuesday & Thursday 12:30 - 1:45 pm, Smith 201
- NOTE: Occasionally, when there is a special guest lecture, there may
be changes to time and/or place.
Watch the web site and the course schedule for announcements of such changes.
Everything you need to know if you are taking CISC844:
Detailed and Hyperlinked Course Schedule (tentative)
HO1: Overview, Policies & Grading
HO2: Presentation Guidelines
Overview
Computational Biomedicine is a seminar-based course, in which we will survey mostly well-established - but also some newer - approaches in machine learning, algorithms and computational theory, along with their applications in biomedical informatics. The past fifteen years have seen tremendous growth in the amount of biomedical data, and much recent research in biomedical computing/informatics is focused on its interpretation. The ultimate goal is to understand and predict normal function of organisms and, more importantly, the mechanisms underlying disease. Given the wealth of data, the interpretation cannot be done manually. It requires advanced algorithms and computational tools, mimicking some aspects of the manual interpretation process, but expediting it several folds.
Machine Learning is concerned with the formulation and automatic acquisition of models from data, as well as with using such models for automatic inference and prediction. Both modeling and inference/prediction are inherent to biomedical data-analysis problems; as such machine learning methods are well-suited for this domain and are indeed applied to a wide variety of biological and medical problems.
Throughout the course we will survey methods and approaches in algorithms, statistics and machine learning, along with their current applications in biomedical informatics.
Planned topics in algorithms and machine learning include:
- Unsupervised learning and clustering
- Supervised learning and classification
- Hidden Markov models and time series
- Text mining and topic models
Planned discussions of biomedical applications include:
- Protein subcellular localization
- Computational analysis of gene expression in the context of disease;
- Biomedical text and image mining
- Mining electronic health records
- Computational analysis of medical signals: Electrocardiograms.
- Diagnosis and prediction of disease, in populations and in
individuals.
More details will be posted shortly
Course mechanics and grading
The course is primarily a seminar, consisting of reading, presenting
and participating in discussions.
A Detailed Course Schedule
Seminar presentations, 50%.
Each student will present and lead discussion, most likely two times during the course. Each presentation is typically about 2 papers discussing a common topic.
A list of topics/papers for presentation will be provided during the first week of classes.
Project, 30%.
Exact details will be provided early in the semester.
Typically, each student will choose a topic,
and carry out a small research project.
Examples include implementation of an existing method, a survey of a specific domain, or a brief write-up developing a new idea.
Project proposal presentations will take place mid-semester.
Presentations of completed projects, will take place in a class event at the end of the semester. Project reports are due at the end of the semester.
Class participation and submission of reading comments, 20%.
Showing up (on-time!), reading the papers prior to each seminar (+ short comments submission), asking questions and participating in discussions, and filling up evaluation forms at the end of the term.
Prerequisites
Understanding of basic concepts in probability and statistics, background in algorithms and theoretical computer science, as well as some familiarity with artificial intelligence.
Background in biology or medicine is an advantage - but is not required.
Course material
There is no required textbook. A collection of papers will be used throughout.
Supplementary recommended reading is listed below.
Supplementary books reserved at the Morris Library:
- Machine Learning. T.M. Mitchell. 1997. [Q325.5 M58]. (Chapters 3-6).
- Pattern Classification and Scene Analysis. R.O. Duda and P.E. Hart. 1973.[Q327 .D83]. (Chapter 6 in particular).
- Artificial Intelligence: A Modern Approach. S. Russell and P. Norvig.
- Bioinformatics: The Machine Learning Approach. P. Baldi and S. Brunak.
- Molecular Biology of the Cell. B. Alberts et al. .
Some helpful web sites:
Last update: February 10, 2020