Building Diversity, Equity, and Inclusion Within Radiology Artificial Intelligence: Representation Matters, From Data...

Area of Science:

• Radiology Artificial Intelligence
• Healthcare Equity
• Medical Informatics

Background:

• Diversity, equity, and inclusion (DEI) are essential for advancing radiology artificial intelligence (AI).
• Current AI development and deployment in healthcare risk perpetuating existing health inequities.
• A lack of diverse representation in AI leadership and research is a significant concern.

Purpose of the Study:

• To highlight the critical role of DEI in the development and deployment of radiology AI.
• To emphasize the need for integrating DEI principles throughout the AI lifecycle.
• To propose solutions for fostering a diverse and inclusive AI-driven healthcare future.

Main Methods:

• Analysis of DEI principles in AI design and data sets.
• Examination of DEI's impact on AI model accuracy and patient care.
• Review of representation in radiology AI leadership and research.

Main Results:

• Biased data sets can lead to inaccurate and discriminatory AI models.
• Diverse leadership and research teams are necessary to mitigate health inequities.
• Targeted DEI training programs can cultivate future leaders in radiology AI.

Conclusions:

• Integrating DEI into radiology AI is a moral and practical imperative.
• Proactive measures are needed to ensure AI benefits all patient populations equitably.
• Developing a diverse pipeline of leaders is key to an inclusive AI-enabled healthcare system.