AI assistance may enhance the accuracy of skin cancer diagnoses in clinical practice, according to a study in NPJ Digital Medicine. Isabelle Krakowski, MD, and colleagues conducted a systematic review and meta-analysis comparing clinicians’ diagnostic performance with and without AI assistance, with the intention to evaluate how AI assistance affects the accuracy of skin cancer diagnoses. They identified 2,983 studies, of which 10 met the criteria for meta-analysis. Without AI assistance, clinicians had a pooled sensitivity of 74.8% (95% CI, 68.6-80.1) and specificity of 81.5% (95% CI, 73.9-87.3). With AI assistance, sensitivity improved to 81.1% (95% CI, 74.4-86.5) and specificity to 86.1% (95% CI, 79.2-90.9). AI assistance benefitted clinicians across all experience levels, with non-dermatologists showing the most significant improvement. The researchers detected no publication bias, and sensitivity analysis con – firmed the robustness of the findings. They recommended further research to validate these benefits in real-world clinical environments.