We will build a user-friendly infrastructure that integrates and produces usable, diverse and disparate data. By including rich data resources from the population health, clinic and basic science domains (see diagram below), investigators will be able to address long-standing questions from completely new perspectives using cutting-edge machine learning and other big data techniques. We can then translate key insights into real, evidence-based solutions for the prevention, treatment and management of obesity.

The DataBridge

Program integration

The NYU Langone Comprehensive Program on Obesity will build upon our research strengths in three major spheres (pictured above):

  • Population Health
  • Clinical Medicine
  • Discovery Research (i.e., Basic Science)

The DataBridge will integrate data from these spheres to help us answer novel questions since many factors influence them, e.g.,:

  • Environmental determinants and the microbiome are featured prominently in our population health and discovery research 
  • Biosamples and clinical targets help shape our clinical and discovery research
  • Clinical data and social determinants affect the direction of our clinical and population health research 

Translational Research: Weight and Metabolic Change

Determine predictors of weight and metabolic outcomes in bariatric surgery patients

The dramatic weight loss during bariatric surgery serves as a powerful tool to study mechanisms of weight and metabolic change in obesity and diabetes. Comprehensive sample collection, combined with individual, clinical and population health data and big data analytics, will allow us to better understand these mechanisms and their implications, and improve our ability to predict the outcomes of the surgery. 

The Plan
We will recruit bariatric surgery patients and follow them longitudinally for 2 years after surgery. In addition to collecting stool, blood and adipose tissue samples, we will conduct continuous glucose monitoring several times during pre- and post- bariatric surgery.

Enriched Database Research: Pediatric Obesity

Standard methods allow us to predict 7-10% of the variance in childhood obesity. Our goal is to predict 75% of variance with currently available data.

We are currently unable to accurately predict which children or small communities will experience problems with obesity later in life. We will combine comprehensive clinical and population health data, along with advanced analytic techniques, to better calculate obesity risk for children. This will set the stage for prevention and treatment of the disease for individuals and communities.

The Plan
We will gather a diverse set of data on a large sample of children in Brooklyn over time and, using machine learning, employ big data techniques to predict obesity at the individual level and population level.  

Program Integration


Program Integration

Academic Partners

Bariatric Surgery

Department of Medicine – Division of General Internal Medicine and Clinical Innovation

Department of Medicine – Division of Endocrinology, Diabetes and Metabolism

Department of Population Health – Section on Health Choice, Policy and Evaluation