Researchers from Fox Chase Cancer Center, Temple University’s College of Engineering, and the Lewis Katz School of Medicine at Temple University have introduced a new artificial intelligence method to improve skin cancer detection in people with darker skin tones.
The findings are detailed in the study “MST-AI: Skin Color Estimation in Skin Cancer Datasets,” published in the Journal of Imaging.
Hayan Lee, corresponding author on the study and assistant professor in the nuclear dynamics and cancer research program as well as a member of the Cancer Epigenetics Institute at Fox Chase, explained some limitations with current technology. “The biggest issue with current AI cancer detection models is that they are more effective at detecting melanoma in lighter skin tones and often have difficulty detecting it in darker skin tones. As a result, when melanoma is detected in patients with darker skin, those patients tend to be diagnosed at later stages,” said Lee.
Researchers point out that many existing AI models rely on limited data sets that are not representative of all populations. Most training data come from only a few locations or time periods—often within one country—and do not include diverse patient types. This lack of diversity can cause bias, making these tools less accurate for people with dark skin compared to those with light skin.
By expanding the range of skin tones included in training data, the team’s approach aims to address this gap and improve early diagnosis for all individuals. The researchers say their new method will help doctors use smarter diagnostic tools that move beyond one-size-fits-all solutions.
