6.7930_[6.871]/HST.956: Machine Learning for Healthcare

Instructors: Peter Szolovits, Manolis Kellis
Teaching Assistants: Hussein Mozannar, Eric Lehman
Graduate level; Units 4-0-8 (counts as an AUS2, AAGS, and II subject; also a EECS AI TQE)
Lectures: Tuesdays & Thursdays, 2:30-4:00pm Eastern Time, 35-225
Recitations (required): Friday, 3:00-4:00pm, 4-270
Prerequisite: 6.3900_[6.036] or 6.7900_[6.867] or 9.520/6.7910_[6.860] or 6.8611/6.8610_{[6.806/6.864]} or 6.4102/6.4100_{[6.438/6.034]} or equivalent machine learning class from other schools
_{(Subscripted bracketed numbers are the class numbers
before this year's mass renumbering of all EECS classes.)}
Office Hours: Monday 11:00am-12:00pm (32-370), Friday 4:00-5:00pm (36-156)
Contact staff: mlhc2023@googlegroups.com

Announcements

Lectures, readings and assignments will be posted to Canvas.
Please add yourself to the course Piazza (find the link in Canvas)
We do not plan to support remote participation in lectures and our expectation is that students will attend class in person unless an exception is granted (e.g., students needing to quarantine/isolate, or a clinician who is on service).
See last year's site for the 2022 schedule and problem sets; this year's will be somewhat similar, but not identical. See 2020's site or MIT OpenCourseWare for slides and videos of past lectures.
This class will have a final exam, but no midterm.
Class group projects will be organized with clinician mentors. Students will present their results in the last week of class.
Recitation are required; they will include deep dives into medical data, tutorials on tools, and paper discussions.

Course description

Introduces students to machine learning in healthcare, including the nature of clinical data and the use of machine learning for risk stratification, disease progression modeling, precision medicine, diagnosis, subtype discovery, and improving clinical workflows. Topics include causality, interpretability, algorithmic fairness, time-series analysis, graphical models, deep learning and transfer learning. Guest lectures by clinicians from the Boston area and course projects with real clinical data emphasize subtleties of working with clinical data and translating machine learning into clinical practice.

Schedule

The schedule is tentative, and will be updated as the class proceeds.

Please visit Canvas page to access all slide decks, recitation materials, and more.

	Class	Date	Lecture & Materials	Assignments
Overview of Clinical Care & Data	1	Tue, Feb 7	Introduction: What makes healthcare unique? [slides] Week 1 reading : AI in health and medicine Machine Learning in Medicine
	2	Thu, Feb 9	Overview of Clinical Care [slides]
	3	Tue, Feb 14	Overview of Clinical Data Science [slides] Week 2 reading [required] EHR Safari Data is Contextual
	4	Thu, Feb 16	Cautionary Tales; Discussion of Project Ideas; Guest speaker Dr. Leo Anthony Celi [slides]
ML with Clinical Text, Imaging, Physiological, and Administrative Data		Tue, Feb 21	No class -- Monday schedule of classes
	5	Thu, Feb 23	Risk Stratification from Structured Health Data [slides] Reading: Razavian N, Blecker S, Schmidt AM, Smith- McLallen A, Nigam S, Sontag D (2015) Population-level prediction of type 2 diabetes from claims data and analysis of risk factors. Big Data 3:4, 277-287, DOI: 10.1089/big.2015.0020. Population-Level Prediction of Type 2 Diabetes From Claims Data and Analysis of Risk Fa
	6	Tue, Feb 28	Risk Stratification (continued); Physiological Time-Series [slides]
	7	Thu, Mar 2	Intro to Clinical NLP [slides] Reading: Beam AL, Kompa B, Schmaltz A, Fried I, Weber G, Palmer N, et al. Clinical Concept Embeddings Learned from Massive Sources of Multimodal Medical Data. In: Biocomputing 2020. Kohala Coast, Hawaii, USA: WORLD SCIENTIFIC; 2019. p. 295-306.
	8	Tue, Mar 7	Contemporary Clinical NLP Methods [slides]
	9	Thu, Mar 9	Survival Analysis, Censoring, Proportional Hazard Models [slides]
	10	Tue, Mar 14	Cancelled because of storm shutdown
Causal Inference	11	Thu, Mar 16	Human-AI Collaboration in Clinical ML [slides]
	12	Tue, Mar 21	Causal Inference, Conditional Treatment Effects [slides]
	13	Thu, Mar 23	Causal Inference, continued, and Intro to Reinforcement Learning [slides]
		Mar 27-31	Spring Break -- No classes
	14	Tue, Apr 4	Dataset Shift [slides]
Real World Deployment Challenges	15	Thu, Apr 6	Regulation, Law and Deployment; Guest lecture from BU/MIT Technology Law Clinic (Chris Conley, Sophie Volpe, Lucas Batties) [slides for FDA and SaMD] [slides for FDA and CDSS]
	16	Tue, Apr 11	Learning with Noisy Labels, Unsupervised Learning Applications, Weak Supervision [slides]
	17	Thu, Apr 13	Privacy and Confidentiality [slides]
	18	Tue, Apr 18	Interpretability [slides]
	19	Thu, Apr 20	Genomics, Genetics, Cohort data, personalized predictions, Polygenic risk scores, rare variant prediction, pre-natal testing, ethics [slides]
	20	Tue, Apr 25	Intro to ML for Medical Imaging: algorithms & applications; Guest lecture by Alex Goehler, Novartis [slides]
	21	Thu, Apr 27	Fairness; Guest lecture by Marzyeh Ghassemi
	22	Tue, May 2	Genetics + EHR integration through eQTLs, patient subtyping, functional genomics, deep learning for multi-modal integration [slides]
	23	Thu, May 4	Visualization and User Interfaces; Guest lecture by Arvind Satyanarayan [slides]
	24	Tue, May 9	Deep learning for drug design, and integration with patient modeling, and target identification
	25	Thu, May 11	Rare disease vs. common disease matching and drug prioritization
	26	Tue, May 16	Student Project Presentations -- Grier Room (34-401), using posters
		May 19 9am-noon	Final Exam -- Johnson Athletic Center, Track level

Problem sets

The problem sets pdfs are available here (not that some data for the problem sets is not publicly available):

Some of the recitations are available here:

Projects

We will release the project guidelines and detailed information in Canvas.

Late Policy

(starting for pset1 onwards)

[4 "slack" days] We understand that sometimes things outside one's control prevent submitting by the deadline. As such, each student is given 4 "slack" days that they can use throughout the semester (e.g. you could submit four psets one day late each or you could submit one pset three days late and another one day late) without a late penalty. The days do not subdivide into sub-day units: 2 hours late would spend one of the slack days without 22 hours of "rollover". In your pdf writeup, specify how many slack days you are using (they cannot be used retroactively).
[10% off per unexcused late day.] If you submit a pset 3 days late and use 1 slack day, then this is 2 unexcused late days, which translates to 20% off your homework.
[Max 4 days late.] Homework will not be accepted after 4 days late, regardless of how many slack days are used, absent communication from S3 or OGE.
[write on homework] In order to use a slack day, students must include it in writing on their submission pdf. Otherwise, TAs will assume no slack days used and deduct 10% for each late day.

Scenarios:

Sam uses 2 slack days on HW3. This is the first time Sam has used any slack days. Sam now has 2 remaining slack days and receives her homework score with no penalty.
Jamie uses 1 slack day on HW3 but submits 52 hours after the deadline. Therefore Jamie is 3 days late (rounded up) and receives 20% off the graded homework. This is the first time Jamie has used any slack days, so Jamie now has 1 slack day remaining.
Cory has never used any slack days, but submits HW5 100 hours late with no note / communication from S3 or OGE. This is more than 4 days after the deadline, so the work is not accepted, and Cory receives a zero.

6.7930[6.871]/HST.956: Machine Learning for Healthcare