With massive amounts of data flowing from EMRs, wearables, and countless other new sources, the potential for machine learning and AI to transform healthcare is perhaps more drastic and profound than any other industry. However, there are unique obstacles that exist in healthcare that can make it difficult to apply machine learning. Oftentimes, data are missing, inaccurate or stored in silos. Connecting patient records across providers and insurers is a challenge due to the lack of interoperability and reliable patient identification methods. And in some cases, such as when dealing with patients with rare conditions, data is insufficient or incomplete.
- Understand current ML trends and opportunities that they bring in healthcare
- Outline practical problems that impact the application
- See how to break down data silos between patients, providers and payers
- Discover how to deploy ML to improve patient outcomes and/or impact the financial performance of your organization
- Grasp what predictive analytics often does not provide
Who Should Attend:
This course will be applicable to data scientists, software engineers, software engineering managers, and those working on health outcomes data from a range of industries including insurance, pharmaceuticals, electronic health records, health-related start-ups, Example companies include Philips Research, Kaiser, Blue Cross Blue Shield, Athena Health, Atrius Health, IBM Watson Health, Shire, Johnson & Johnson, Novartis, PatientsLikeMe.
Participants should be comfortable programming in Python, performing basic data analysis, and using the machine learning toolkit Scikit-learn. Additionally participants should be familiar with machine learning (we recommend the MIT Professional Education course Machine Learning for Big Data and Text Processing: Foundations for participants who feel they need preparation in this area).
Laptops with Python and Scikit-learn installed are required.
- What makes Healthcare unique? (9:00 am - 9:30 am)
- Supervised Learning For Risk Stratification (9:30 am -10:30 am)
- Break (10:30 am -10:45 am)
- Case Study for Risk Stratification (10:45 am -Noon)
- Lunch (Noon-1:00 pm)
- Lab Work (1:00 pm -3:00 pm)
- Break (3:00 pm -3:15 pm)
- Time-Series Modeling (3:15 pm -5 pm)
- Predicting the Outcome of Interventions: Causal Interference from Observational Data (9:00 am -10:00 am)
- Break (10:00 am -10:15 am)
- Lab Work (10:15 am -12:30 pm)
- Lunch (12:30 pm -1:30 pm)
- Interpretability of Machine Learning Models (1:30 pm -3:00 pm)
- Break (3:00 pm -3:15 pm)
- Deploying Machine Learning in Healthcare Settings (3:15 pm -4:30 pm)
- Conclusion (4:30 pm -5:00 pm)
This course runs 9:00 am - 5:00 pm each day.
David Sontag joined the MIT faculty in 2017 as Hermann L. F. von Helmholtz Career Development Professor in the Institute for Medical Engineering and Science (IMES) and as Assistant Professor in the Department of Electrical Engineering and Computer Science (EECS). He is also a principal investigator in the Computer Science and Artificial Intelligence Laboratory (CSAIL). Professor Sontag’s research interests are in machine learning and artificial intelligence. As part of IMES, he leads a research group that aims to transform healthcare through the use of machine learning. Prior to joining MIT, Sontag was an Assistant Professor in Computer Science and Data Science at New York University’s Courant Institute of Mathematical Sciences from 2011 to 2016, and postdoctoral researcher at Microsoft Research New England from 2010 to 2011. He received the Sprowls award for outstanding doctoral thesis in Computer Science at MIT in 2010, best paper awards at the conferences Empirical Methods in Natural Language Processing (EMNLP), Uncertainty in Artificial Intelligence (UAI), and Neural Information Processing Systems (NIPS), faculty awards from Google, Facebook, and Adobe, and a NSF CAREER Award.
This course takes place on the MIT campus in Cambridge, Massachusetts. We can also offer this course for groups of employees at your location. Please complete the Custom Programs request form for further details.
|Fundamentals: Core concepts, understandings, and tools (60%)||60|
|Latest Developments: Recent advances and future trends (20%)||20|
|Industry Applications: Linking theory and real-world (20%)||20|
|Lecture: Delivery of material in a lecture format (60%)||60|
|Discussion or Groupwork: Participatory learning (10%)||10|
|Labs: Demonstrations, experiments, simulations (30%)||30|
|Introductory: Appropriate for a general audience (25%)||25|
|Specialized: Assumes experience in practice area or field (65%)||65|
|Advanced: In-depth explorations at the graduate level (10%)||10|