Hartford HealthCare and MIT launch “Holistic AI in Medicine Collaboration”

HHC & MIT Collaboration logoHartford HealthCare (HHC) and the Massachusetts Institute of Technology (MIT) have launched an ambitious and promising collaboration.

Fourteen MIT doctoral students will work side-by-side with clinical experts from HHC to use artificial intelligence and machine learning to advance healthcare.

Doctoral Students Research Projects Holistic AI
In the News
In the News     

The student-physician teams, under the direction of Professor Dimitris Bertsimas, will study how AI and ML can be used to predict disease probability sooner, provide more exacting tests and use massive amounts of data to deliver care that is more personalized. (What is "Holistic AI in Medicine"?)

Dimitris Bertsimas

Dimitris Bertsimas

Professor Bertsimas is the Boeing Leaders for Global Operations Professor of Management, a
professor of Operations Research, and the associate dean for the Master of Business Analytics program at the Massachusetts Institute of Technology ...

Dimitris is a serial entrepreneur in the areas of artificial intelligence and machine learning, healthcare, education, financial services and transportation. He has co-authored more than 200 scientific papers, and is one of the pre-eminent academics for the next generation of prescriptive, predictive and optimization analytics.

Together, HHC and MIT are shifting the frontiers of artificial intelligence and machine learning in healthcare.

Meet the MIT doctoral students who will work with our experts to advance AI and ML in healthcare

Each of the participating students are part of MIT’s Operations Research Center, studying and working with Dr. Bertsimas.

Kimberly Villalobos Carballo

Kimberly Villalobos Carballo

A fifth-year doctoral student, Kimberly’s research integrates optimization and machine learning tools to enhance the performance of conventional algorithms. She is passionate about healthcare applications, and much of her research has been inspired by collaborations with hospitals (including work with Hartford Hospital on predictive machine learning models for length of stay reduction and identification of life-threatening patient events) that aim to improve the quality of their services and operations.

Yubing Cui

Yubing Cui

Yubing began her first year as a doctoral student after graduating in April from the University of Michigan with bachelor of science degree in math and computer science. As an undergraduate, she maintained 3.99 grade point average and earned a variety of math awards and merit scholarships. In addition to science, she is a music lover, with interests ranging from classical orchestra to Kpop.

Lisa Everest

Lisa Everest

Lisa is a second-year doctoral student and her research integrates optimization and machine learning with applications in healthcare and special interest in prescription. She previously worked for three years as a quantitative analyst in Goldman Sachs' Special Situations Group in private asset management. She received bachelor’s degrees in mathematics and management from MIT. In her free time, she enjoys snowboarding, figure skating and baking.

Carol Gao

Carol Gao

Carol is a research fellow at MIT. She graduated from Smith College with a bachelor’s degree in mathematics and quantitative economics.

Jiayi Gu

Jiayi Gu

Jiayi is a second-year doctoral student with research interests that center around the synergy of machine learning and optimization in healthcare applications. She holds a bachelor's degree in industrial engineering and global health studies from Northwestern University, and is particularly interested in using holistic, data-driven approaches to address public health challenges.

Emily Hahn

Emily Hahn

Emily is currently pursuing a master’s of business analytics degree at the MIT Sloan School of Management. She is interested in applying data analytics and machine learning to improve healthcare efficiency and effectiveness. Before arriving at MIT, she earned a bachelor of science in operations research at Cornell University.

Rohan Kumar

Rohan Kumar

A third-year undergraduate at Boston University, Rohan studies computer engineering. His academic experience has been in computer organization, algorithms and mathematics. His research experience thusfar has been in predictive models and optimization, both in topics regarding healthcare. He hopes to continue learning about the applications of AI in medicine. His hobbies include playing violin, playing and watching soccer, and reading.

Yu Ma

Yu Ma

A fourth-year doctoral student, Yu has already been collaborating with Hartford Hospital teams in such areas as rapid response and cardiovascular surgery. She has authored and co-authored research articles in such publications as Nature, Journal of the American Medical Association (JAMA) Surgery, International Journal of Radiation Oncology and Journal of Clinical Oncology Informatics. She was part of the original HAIM team that earned the Cognex prize at MIT’s Machine Intelligence for Manufacturing and Operations Symposium, and was the graduate student representative on the MIT committee for the use of humans as experimental subjects. In addition to her research, she loves to scuba dive.

Gagik Magakyan

Gagik Magakyan

A native of Armenia, Gagik represented his country in numerous international math and computer science olympiads, earning several medals. He then pursued a bachelor's degree in mathematics from the University of Cambridge. His journey into AI began through internships and summer projects, including recent experience at Google X. He is in the first year of graduate studies at MIT, where he is excited to immerse in the field of operations research. Beyond academics, he is an avid soccer fan and long-standing supporter of FC Manchester City.

Catherine Ning

Catherine Ning

Born and raised in Luxembourg, Catherine earned an integrated four-year master’s degree in engineering science at Oxford University, graduating in July 2023. Her master’s thesis was on mixed-integer programming and she has worked on a team designing an autonomous hospital robot for medicine delivery to help reduce medical waste. As she starts as a doctoral student in operations research this fall, she is excited to take part in HAIM and contribute to a better healthcare system in the future.

Phevos Paschalidis

Phevos Paschalidis

Phevos is a junior at Harvard University studying computer science. He has been working with Hartford Hospital for more than a year to apply interpretable machine learning techniques to prescribe the optimal valve type to minimize post-operative pacemaker risk during Transcatheter Aortic Valve Replacement (TAVR) procedures. His other research interests include theoretical computer science and reinforcement learning.

Matthew Peroni

Matthew Peroni

Matthew is a second-year doctoral student. His research is focused on building interpretable, multi-modal, and robust machine learning systems for high-stakes decision making, especially in the healthcare domain. Prior to graduate school, he worked in industry as a machine learning engineer for a healthcare AI start-up. He received his bachelor’s in mathematics and computer science from Cornell University.

Vasiliki “Vassilina” Stoumpou

Vasiliki “Vassilina” Stoumpou

A second-year doctoral student, Vassilina grew up in Greece. She finished her five-year undergrad studies at the National Technical University of Athens in electrical and computer engineering, with a major in computer science. She worked at Tesla as an associate electrical design engineer before coming to MIT, where she is currently working on projects that focus on leveraging machine learning techniques in healthcare.

Karl Zhu

Karl Zhu

Karl is a first-year master's degree student specializing in operations research. Before coming to MIT, he was a graduate research assistant in the University of Auckland (New Zealand) Department of Engineering Science, where he built a strong foundation in applying statistical and optimization models. His work earned the Best Presentation of the Young Practitioner’s Prize at the 2022 Operations Research Society of New Zealand conference. Karl holds a bachelor’s degree in engineering science from the University of Auckland.

Here are the projects that will be studied

(along with the HHC clinical partner for each project)

Mapping Coercive Practices in Psychiatric Care and their Relationship with Racial Bias

Javeed Sukhera, MD, PhD, Behavioral Health Network

Prediction of Left Ventricular Ejection Fraction from a 12-Lead Electrocardiogram

Steven Zweibel, MD, Electrophysiology, Heart & Vascular Institute

Prediction of Mortality/Complications in TAVR and Other Cardiac Procedures

Howard Haronian, MD, Interventional Cardiology, Heart & Vascular Institute

Prediction of Chronic Subdural Hematoma Recurrence

Tapan Mehta, MD, Interventional Neuroradiology, Neuroscience Institute

Patient Screening for Early Detection and Guideline-Directed Management of Aortic Stenosis

Trevor Sutton, MD, Cardiac Anesthesia

Predicting Psychosis Relapse

Manu Sharma, MD, Behavioral Health Network

Determine Brain Age from T1 Images

Michal Assaf, MD, Behavioral Health Network

A Data-Driven Analysis Strategy to Identify Features of HHC Communities that Increase Risk for Weight Gain and Obesity

Dale Bond, PhD, Digestive Health Institute

Rapid Identification of Psychosis Biotypes from the BSNIP Project Using AI/ML

Godfrey Pearlson, MD, Behavioral Health Network and Olin Neuropsychiatry Research Center

Treatment of Fragility Hip Fractures

Heeren Makanji, MD, Bone & Joint Institute

Automated Care Reviews and Medical Necessity Denial Management

Swathi Rachoor MD, Integrated Care Partners

Using AI/ML to Diagnose AAST Injury Classification for Blunt Splenic and Liver Injuries on CT scan

Shea Gregg, MD, Trauma Surgery

Predicting which Patients Transitioning from an Inpatient Stay will have Poor Outcomes Due to Medication-Related Issues

Eric Arlia, MBA, RPh, Pharmacy

Predicting patients at risk for “Hospital Acquired Infections”

Adam Steinberg, MD, Vice President of Medical Affairs, Hartford Hospital

 
HAIM

What is "Holistic AI in Medicine"?

Holistic AI in Medicine (HAIM) is a new artificial intelligence framework that can integrate multiple data sources — including information stored in electronic health records (like Epic), clinical images, and readings from EKGs, pulse oximeters, blood pressure cuffs and other devices, physician notes, and radiology reports. HAIM uses AI to learn from the complex relationships between these data sources to make predictions about patient outcomes.

HAIM has proven to be effective in a variety of healthcare applications: predicting patient mortality, diagnosing diseases, determining the best course of treatment, monitoring patient progress and predicting drug interactions.

Hartford HealthCare and MIT believe that HAIM has the potential to revolutionize healthcare by providing doctors with more accurate information about their patients. MIT’s AI research students will collaborate with HHC’s renowned clinical experts and work to develop new treatments, strategies and diagnostic tools for a variety of health conditions and situations.

 

In the News

These CT doctors look to AI to advance medicine, patient health. ‘It’s that transformational.’

Ed Stannard
Hartford Courant: 09-11-2023
Fourteen Hartford HealthCare doctors have projects they think could be accomplished, or completed more efficiently, if only artificial intelligence were applied to them. They recently presented those projects to 14 doctoral students from the Massachusetts Institute of Technology’s business school, who are ready to take them on.
Read the story