Evaluation of AIML + HDR-A Course to Enhance Data Science Workforce Capacity for Hispanic Biomedical Researchers

Area of Science:

• Health Informatics
• Data Science
• Medical Research

Background:

• Artificial intelligence (AI) and machine learning (ML) offer revolutionary medical advancements.
• Existing AI/ML methods perpetuate health inequities due to inherent biases.
• Training a diverse workforce is a key strategy to mitigate these biases.

Purpose of the Study:

• To introduce Data Science (DS) approaches to health disparities research.
• To emphasize the application of AI/ML in research focusing on Hispanic populations.
• To develop a curriculum addressing both technical AI/ML skills and health disparities concepts.

Main Methods:

• Developed and delivered a course titled "Artificial Intelligence and Machine Learning applied to Health Disparities Research (AIML + HDR)".
• Covered technical topics: Jupyter Notebook, R and Python for data manipulation, ML libraries.
• Included health disparities topics: Electronic Health Records, Social Determinants of Health, Bias in Data.

Main Results:

• The course successfully trained 34 Hispanic participants.
• Over 80% of participants expressed high satisfaction with course organization, activities, and content.
• Participants strongly agreed that activities were relevant (3.71 ± 0.21) and promoted learning.
• Participants strongly agreed the course aided professional development (3.76 ± 0.18).
• Quantitative analysis of open-ended feedback showed 75% of comments indicated high satisfaction.

Conclusions:

• The AIML + HDR course effectively trained a diverse cohort in AI/ML for health disparities research.
• The curriculum successfully integrated technical DS/AI/ML skills with critical health disparities topics.
• High participant satisfaction and perceived value for professional development suggest the course's success in addressing workforce diversity for equitable AI in health.