Reducing Racial Bias in AI Models, with Nawsabah Noor, MBBS

Artificial intelligence (AI) is reshaping clinical diagnostics, but its promise is often limited by biased training data. In skin disease detection, most AI models rely on datasets dominated by lighter skin tones—leaving patients with darker skin at risk of misdiagnosis.

Nawsabah Noor, MBBS, assistant professor of medicine at Popular Medical College in Bangladesh, and colleagues are working on developing an AI model with novel training to address racial bias when diagnosing mpox. She presented findings from the model and app that they are developing at the American College of Physicians (ACP) Internal Medicine (IM) Meeting 2025, held April 3-5, in New Orleans, Louisiana.

HCPLive spoke with Noor to learn more about the AI model and the team’s novel strategy for reducing racial bias by altering the saturation of pictures. She shared further research the investigators are pursuing to further validate the app. She also urged clinicians to try and adapt to the inevitable incorporation of AI into everyday practice and research.

“AI is coming, and it will help us in various way. So, we need to cooperate with AI. We need to think about how AI can help us in clinical practices, so that we can collaborate… and develop some solutions like this easy solution. It sounds very simple, right? We need to incorporate AI in our day-to-day practice, and also we need AI in our research as well,” Noor told HCPLive during the meeting.

Noor has no relevant disclosures to report.

REFERENCE

Noor N. Reducing racial bias in Artificial Intelligence (AI) models for mpox skin lesion detection: A colorspace augmentation scheme based on the Fitzpatrick skin-tone scale. Presented at: ACP IM Meeting 2025; April 3-5; New Orleans, Louisiana.