Harald Kittler, MD, Medical University of Vienna, Vienna, Austria, highlights biases that could interfere with accurate diagnosis of skin cancers using artificial intelligence (AI) models. Systemic bias can lead to the underdiagnosis or overdiagnosis of a disease. Training data sets for AI diagnosis models may exclude underrepresented patient cohorts based on age, sex, or ethnicity, or contain the histopathology of specific disease subtypes, which are not well defined and lead to misdiagnosis. A potential solution would be to incorporate a more diverse and inclusive patient cohort in the training datasets for AI diagnosis of skin cancers. This interview took place at The European Congress on Dermato-Oncology 2024 in Vienna, Austria. These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.

Harald Kittler, MD, Medical University of Vienna, Vienna, Austria, highlights biases that could interfere with accurate diagnosis of skin cancers using artificial intelligence (AI) models. Systemic bias can lead to the underdiagnosis or overdiagnosis of a disease. Training data sets for AI diagnosis models may exclude underrepresented patient cohorts based on age, sex, or ethnicity, or contain the histopathology of specific disease subtypes, which are not well defined and lead to misdiagnosis. A potential solution would be to incorporate a more diverse and inclusive patient cohort in the training datasets for AI diagnosis of skin cancers. This interview took place at The European Congress on Dermato-Oncology 2024 in Vienna, Austria. These works are owned by Magdalen Medical Publishing (MMP) and are protected by copyright laws and treaties around the world. All rights are reserved.