Data and Research Core

The mission of the AIM-AHEAD Data and Research Core (DRC) is to broaden the diversity and representation of healthcare data, including electronic health record (EHR) data, in AI/ML and expand its availability to diverse teams of researchers to address health disparities.

The DRC is designed to support the development of a workforce capable of conducting high-quality AI/ML for health equity-focused research, with an emphasis on supporting persons from groups underrepresented in biomedical research (UBR) and/or building capacity at Minority-Serving Institutions (MSIs) to work with healthcare data, starting with electronic health record (EHR) and relevant linked social and environmental determinants of health (SEDOH) data and expanding to data from imaging, genomics, wearable devices and text such as clinical notes.

The DRC is not a single database. Instead, AIM-AHEAD seeks to catalyze an ecosystem of datasets to help address the lack of population diversity in data used in most AI/ML models and applications.

AIM-AHEAD Data Partners

AIM-AHEAD-funded projects may apply to receive facilitated access and data concierge services from AIM-AHEAD data partners that emphasize historically under-resourced and under-represented populations:

OCHIN, a nonprofit health care innovation center with a core mission to advance health equity, operates the most comprehensive database on primary healthcare and outcomes of safety net patients in the U.S., connected by OCHIN’s Epic EHR system, representing 6 million patients from 170 health systems and 1,600 clinic sites across 33 states. LEARN MORE

The MedStar Health System and the MedStar Health Research Institute (MHRI) include an extensive network of clinical facilities in the mid-Atlantic region, including 10 hospitals and over 200 affiliated facilities connected by MedStar’s Cerner EHR system, representing 5 million unique patients, approximately 31% of whom are African American. LEARN MORE

The DRC and Infrastructure Core also collaborate to assist AIM-AHEAD awardees in locating other data sources to support their projects. As part of its mission to diversify datasets used in AI/ML, AIM-AHEAD has conducted a landscape survey to raise awareness about datasets that may be of interest to the research community. Each dataset has its own governance process and rules for access.

View the landscape survey of datasets Apply to Include a Dataset in Landscape List

How AIM-AHEAD Data Partners Expand Representation


Race	People who select a single race other than White, or who select more than one race	2,620,875	2,467,865
Ethnicity	People who select an ethnicity other than those listed under the race of White	2,224,756¹	2,233,438
Age	<18 years old and 65 years and above	1,337,789	1,424,082
Sexual and Gender Minority	People who self-report intersex as their sex at birth People who select any sexual orientation choice other than straight People who select any gender identity choice other than man or woman	249,978²	Not well-captured
Income	Annual household income < $25,000	4,129,925	Not well-captured
Education	People without a high school diploma or GED	Not well-captured but FQHCs generally higher than general population	2,560
Access to Care	Needed a medical visit in the past 12 months but cannot readily use the health care system or pay for needed care	Not well-captured but FQHCs generally higher than general population	Not well-captured
Geography	Residents of established rural and non-metropolitan zip codes, based on the HRSA Federal Office of Rural Health Policy data files	1,072,666	36,345³
Disability	People with a physical, functional, cognitive, or other condition that substantially limits one or more life activities	632,574⁴	266,226⁴

Source: All of Us reference UBR categories

¹People with Hispanic ethnicity and any race

²People with ‘other’ sex or non-straight sexual orientation or non-male, non-female gender identity

³Based on rural and suburban hospital discharges

⁴Based on ICD codes for disability in study by Clark et al , including physical, visual, hearing, intellectual/developmental disabilities

Other Dataset Options for AIM-AHEAD-funded Projects

Data Set	Brief Description	Data Allowed	Size	Analysis Platform Tools
60+ studies from NHLBI BioData Catalyst	Selected large-scale cohorts related to heart, lung, blood and sleep disorders. Includes both prospective clinical studies and associated genomic TOPMED data.	De-identified dataset. Including individual level genomic (TOPMED full genomes) and clinical datasets.	Additional description List of studies : 60+ studies are available to choose from	NHLBI BioData Catalyst PIC-SURE and Seven Bridges Platforms
Selected 15 Open datasets on AWS	A variety of datasets available including clinical and genomic data	Public data, and controlled access data (depends on dataset)	Selected 15 Open datasets on AWS	AIM-AHEAD Service Workbench
NIH All of Us	The All of Us Research Program is building one of the largest biomedical data resources of its kind.	The All of Us Research Hub stores health data from a diverse group of participants from across the United States.	Additional descriptions 616,000+ participants, 360,000+ electronic health records, 444,000+ biosamples	All of Us Researcher Workbench