MIT Critical Data Marathon: Intro to the MIMIC-II Database and Meet the Teams!

Today was the first full day of the Massachusetts Institute of Technology (MIT) Critical Data Marathon–a weekend hackathon designed to highlight the incredible resource of the MIMIC-II database.

Stata Hall

Brilliant Minds Gathered in the Iconic MIT Stata Center Hall

The weekend is run in partnership with Sana Mobile and MIT H@cking Medicine. In true MIT H@cking Medicine form, the weekend was designed to bring together various disciplines that would not otherwise interact in order to brainstorm and crowd source innovative ideas. Brilliant minds from computer science, medicine, nursing, pharmacy, biostatisticts, epidemiology, informatics, business, health policy, and the social sciences were all present. More importantly, academia and industry mingled together.

Team BIDMC ICU Nurses

A true collaboration: Beth Israel Deaconess ICU Nurses Bringing Real World Insight to the Hackathon! Leo Anthony Celi, MD, MSc, MPH, Judy Bertone, Annie Anderson, Justin Dilibero and Sharon O’Donoghue.

MIMIC-II stands for Multi-parameter Intelligent Monitoring in Intensive Care II. It is a publicly available database that is part of the Multi-parameter Intelligent Monitoring in Intensive Care project funded by the National Institute of Biomedical Imaging and Bioengineering at the Laboratory of Computational Physiology at MIT.

MIMIC-II is the only ICU database that encompasses patient demographics, clinical laboratory data, categorical admission diagnoses as well as detailed therapeutic profiles such as intravenous medication drip rates and hourly fluid balance trends for the duration of the ICU stay.

The MIMIC-II is an intensive care unit research database applying automated techniques to aggregate high-resolution diagnostic and therapeutic data from a large, diverse population of adult intensive care unit patients. This freely available database is intended to support epidemiological research in critical care medicine and serve as a resource to evaluate new clinical decision support and monitoring algorithms.

The data is from all adult intensive care units (medical intensive care unit, surgical intensive care unit, cardiac care unit, cardiac surgery recovery unit) between 2001 and 2007 at a tertiary care hospital, the Beth Israel Deaconess Medical Center in Boston, MA. It consists of 25,328 intensive care unit stays. Learn more about the database by clicking here.

Leo Anthony Celi, MD, MSc, MPH, a member of the Organizing Committee, tells us more about the database.

The format of the hackathon mixed traditional and nontraditional.  In traditional fashion, there was a social meetup Friday night and pitching and team forming on Saturday morning. The teams will work throughout the weekend and then present their projects on Sunday evening. At that time, they will be judged, and winners will be announced.

To change up the pace, the weekend is also intermingled with hip hop, kickboxing, and mindful meditation.

Leo Andy Dan

Ned McCague, Andrew Zimolzak, MD, and Leo Anthony Celi, MD, MSc, MPH

There were about 13 teams that formed ideas.

Queen Bee of Boston Med Tech hackathons, Andrea Ippolito, Co-Leader of MIT H@cking Medicine, was at the center of the event, managing people and inspiring ideas! She tells us about the weekend and the teams.

 We interviewed a number of the teams.

One of my old co-fellow nephrologists from Massachusetts General Hospital, Chunmei Huang, MD, MS, was the physician on Team JUCCY. The team consisted of 4 members interested in solving the age old mystery of a patient’s true fluid status to help determine if a diuretic or IV saline needs to be given.

Nephrologists will just not give up seeking a non-invasive measure of a person’s true fluid status. There must be something better than skin tinting, the mysterious jugular venous pressure cardiologists wave about, and ultrasounding a patient’s inferior vena cava.

 Team Oxygenators looked at all the parameters that affect oxygenation of an ICU patient and how this affects outcomes.

Team Oxygenators

Team Oxygenators: Luciana Armijos, Nupur Garg, MD, Yousuf Ahmed, and Melanie Mueller, PhD

They have a fancy diagram. Apparently, a lot of factors affect oxygenation.

Oxygen Graph

It is true that pulmonologists obsess over successfully predicting if a patient “will fly” after extubation–if a patient will successfully breathe on his own after the intubation tube and ventilator support is removed.Team Intubate Extubate looked at factors affecting intubation and successful extubation from a ventilator.

Team Survival Index, which consisted of Matthew Doiron, Vipan Nikore, Stephen Flaherty, Aaron Stupple, and Susan Desanto-Madeya, looked at factors that predict mortality or survival.

Aaron Stupple and Team

This is a fantastic database open to any researcher. You can apply here. Other super successful studies and papers have already been completed, but the opportunities are endless!

Stay tuned to find out the winners!

Jennifer M. Joe, MD

Jennifer M. Joe, MD

My passion is healthcare optimization, whether that is with innovation, making scientific discoveries, or improving delivery. I love bringing people and ideas together and making projects work. With this, medicine exists to improve lives, and I will strive to always help patients and those around me.
Jennifer M. Joe, MD

3 Comments

    Follow us!

    Send this to a friend