Research has been traditionally viewed as a purely academic undertaking, especially in limited-resource healthcare systems. Clinical trials, the hallmark of medical research, are expensive to perform, and take place primarily in countries which can afford them. Around the world, the blood pressure thresholds for hypertension, or the blood sugar targets for patients with diabetes, are established based on research performed in a handful of countries. There is an implicit assumption that the findings and validity of studies carried out in the US and other Western countries generalize to patients around the world.
This course was created by members of MIT Critical Data, a global consortium that consists of healthcare practitioners, computer scientists, and engineers from academia, industry, and government, that seeks to place data and research at the front and center of healthcare operations.
Big data is proliferating in diverse forms within the healthcare field, not only because of the adoption of electronic health records, but also because of the growing use of wireless technologies for ambulatory monitoring. The world is abuzz with applications of data science in almost every field – commerce, transportation, banking, and more recently, healthcare. These breakthroughs are due to rediscovered algorithms, powerful computers to run them, and most importantly, the availability of bigger and better data to train the algorithms. This course provides an introductory survey of data science tools in healthcare through several hands-on workshops and exercises.
Who this course is aimed at
The most daunting global health issues right now are the result of interconnected crises. In this course, we highlight the importance of a multidisciplinary approach to health data science. It is intended for front-line clinicians and public health practitioners, as well as computer scientists, engineers and social scientists, whose goal is to understand health and disease better using digital data captured in the process of care.
We highly recommend that this course be taken as part of a team consisting of clinicians and computer scientists or engineers. Learners from the healthcare sector are likely to have difficulties with the programming aspect while the computer scientists and engineers will not be familiar with the clinical context of the exercises and workshops.
The MIT Critical Data team would like to acknowledge the contribution of the following members: Aldo Arevalo, Alistair Johnson, Alon Dagan, Amber Nigam, Amelie Mathusek, Andre Silva, Chaitanya Shivade, Christopher Cosgriff, Christina Chen, Daniel Ebner, Daniel Gruhl, Eric Yamga, Grigorich Schleifer, Haroun Chahed, Jesse Raffa, Jonathan Riesner, Joy Tzung-yu Wu, Kimiko Huang, Lawerence Baker, Marta Fernandes, Mathew Samuel, Philipp Klocke, Pragati Jaiswal, Ryan Kindle, Shrey Lakhotia, Tom Pollard, Yueh-Hsun Chuang, Ziyi Hou.
Section 1 provides a general perspective about digital health data, their potential and challenges for research and use for retrospective analyses and modeling. Section 2 focuses on the Medical Information Mart for Intensive Care (MIMIC) database, curated by the Laboratory for Computational Physiology at MIT. The learners will have an opportunity to develop their analytical skills while following a research project, from the definition of a clinical question to the assessment of the analysis’ robustness. The last section is a collection of the workshops around the applications of data science in healthcare.
Louis Agha-Mir-Salim, Leo Anthony Celi, Marie-Laure Charpignon, Kenneth Eugene Paik, Julia Situ and Wesley Yeung
Start your review of Collaborative Data Science for Healthcare