The Suchi Saria Lab, part of the Institute for Computational Medicine, explores topics within the fields of machine learning and computational statistics, with a focus on computational solutions for problems in health informatics. Our team investigates the applications of machine learning and computational statistics to domains where one has to draw inferences from observing a complex, real-world system evolve over time. We use Bayesian and probabilistic graphical modeling approaches to address the challenges that emerge with modeling and prediction in real-world temporal systems.

The Mark Dredze Lab investigates topics such as natural language processing, speech, machine learning and intelligent user interfaces. Our team is currently exploring several key health information applications, including information extraction from social media and biomedical and clinical texts. Our recent research in these areas include vaccine communication during the Disneyland measles outbreak; the validity of online drug forums for estimating trends in drug use; and the use of Twitter to examine social rationales for vaccine refusal.

Directed by Debraj “Raj” Mukherjee, MD, MPH, the laboratory focuses on improving access to care, reducing disparities, maximizing surgical outcomes, and optimizing quality of life for patients with brain and skull base tumors.

The laboratory achieves these aims by creating and analyzing institutional and national databases, developing and validating novel patient-centered quality of life instruments, leveraging machine learning and artificial intelligence platforms to risk-stratify vulnerable patient populations, and designing novel surgical trials to push the boundaries of neurosurgical innovation.

Our research also investigates novel approaches to improve neurosurgical medical education including studying the utility of video-based surgical coaching and the design of new operative instrumentation.

The Mathioudakis lab is focused on developing and evaluating clinical decision support systems, technology, and mHealth for diabetes prevention and management. Our lab leverages large electronic medical record databases and uses machine learning algorithms and artificial intelligence to identify patterns in clinical care associated with optimal clinical outcomes. We are interested in understanding the role that advanced diabetes technologies can play in improving health outcomes for patients with diabetes. Our lab has published extensively on outcomes related to diabetes prevention and diabetes management and outcomes. Based on data from our long-term (over 10 year) clinic-based prospective cohort study from the Johns Hopkins Multidisciplinary Diabetic Foot and Wound Clinic, we have published extensively on clinical predictors and outcomes of patients with diabetic foot ulcers, focusing specifically on the role that glycemic control plays in patients with this complication. Healthcare disparities exist throughout medicine, but are particularly prominent in diabetes; our lab has evaluated healthcare inequities in diabetes outcomes and is developing and evaluating strategies to overcome them. In addition to identify optimal management approach to diabetes and its complications, our lab is interested in development and evaluation of innovative technology approaches to diabetes prevention.