Artificial intelligence (AI) is revolutionizing the field of clinical diagnostics, offering unprecedented opportunities to enhance accuracy and efficiency. However, a significant challenge remains: the bias inherent in training data, particularly in the detection of skin diseases. Most AI models are trained on datasets predominantly featuring lighter skin tones, which can lead to misdiagnosis in patients with darker skin.
Enter Nawsabah Noor, MBBS, an assistant professor of medicine at Popular Medical College in Bangladesh. Noor and her team are at the forefront of developing an AI model designed to address this racial bias, specifically in diagnosing mpox. Their innovative approach involves a novel saturation strategy that alters the color saturation of images to create a more balanced dataset.
Noor presented her groundbreaking findings at the American College of Physicians (ACP) Internal Medicine Meeting 2025 in New Orleans, Louisiana. Her work highlights the importance of inclusive datasets in AI training and the potential for AI to transform clinical practices.
In an interview with HCPLive, Noor emphasized the necessity of integrating AI into everyday medical practice and research. "AI is coming, and it will help us in various ways. We need to cooperate with AI," she stated. "We need to think about how AI can help us in clinical practices, so that we can collaborate and develop solutions like this easy solution."
Noor's research is a call to action for clinicians to embrace AI's potential and adapt to its integration into healthcare. Her team's work not only aims to improve diagnostic accuracy but also to ensure equitable healthcare outcomes for all skin tones.
Key Takeaways
- AI models often suffer from racial bias due to unbalanced training datasets.
- Nawsabah Noor's team is developing a model to address this bias in mpox diagnosis.
- Their strategy involves altering image saturation to create a more inclusive dataset.
- Noor advocates for the integration of AI into daily medical practice and research.
- The ultimate goal is to improve diagnostic accuracy and equity in healthcare.
By addressing these biases, Noor and her team are paving the way for more equitable healthcare solutions, ensuring that AI can serve all communities effectively.