Artificial Intelligence Ready and Equitable Atlas for Diabetes Insights (AI-READI) Research Internship Program: Recruiting and Developing a Diverse AI Workforce

Abstract

Purpose : To report recruitment efforts of a new program aiming to promote workforce diversity at the nexus of AI and biomedical research.

Methods : AI-READI is part of the National Institutes of Health (NIH) Bridge2AI Consortium. This structured, yearlong mentored research program aims to facilitate exposure to AI and data science among diverse individuals. Recruitment efforts included outreach to affinity organizations (e.g. Student National Medical Association, Latino Medical Students Association), NIH networks, diversity and inclusion offices at higher education institutions, specialty societies (e.g. the American Academy of Ophthalmology), social media, and print media (e.g. USA Today). A partnership was also formed with the NIH Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity (AIM-AHEAD) program, which facilitated additional referrals. A multi-disciplinary selection committee, which included under-represented minority (URM) faculty, selected applicants. A curricular framework was developed to include a summer bootcamp, yearlong didactic programming, and individual mentored research (Fig. 1).

Results : In this inaugural cycle, out of 26 applications, 13 (50%) were female (Table 1). One applicant (4%) identified as Hispanic/Latino. The most represented racial group was Asian (12/26, 46%), followed by Black/African American (6/26, 23%). Half (13/26, 50%) identified as URM based on NIH criteria. The mean (SD) age of applicants was 35.4 (8.9) years (range 18-53). The inaugural cohort consisted of 10 interns who accepted offers. The majority (7/10, 70%) were female. Demographics were similar to the overall applicant pool; almost one-third (3/10, 30%) of the cohort is Black, and half (5/10, 50%) are URM (Table 1). Interns were successfully recruited from multiple sites nationwide. All interns completed a 2-week in-person programming bootcamp and are currently undergoing didactic lectures and mentored AI research (Fig. 1).

Conclusions : Active outreach through multiple channels resulted in successful nationwide recruitment of a diverse cohort in the AI-READI internship program, with strong representation of women and Black individuals. Ongoing outreach is needed to successfully recruit other under-represented groups, such as Hispanics/Latinos and American Indians/Indigenous Americans.

This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.

 

 

Curriculum