Site Editor
Soo Park, MD
Soo Park, MD
Posted: Tuesday, May 28, 2024
A new artificial intelligence (AI) skin cancer detection tool appears to be able to successfully identify melanoma lesions, according to an article published in the British Journal of Dermatology. Of 253 skin lesions of concern in 228 patients, 11 malignant melanomas and 10 precancerous melanoma lesions were found by regular diagnostic procedures. The app correctly identified all 21 lesions as melanomas, except for 1 of the precancerous lesions. These findings were reported by Magnus Falk, MD, PhD, of Linköping University, Sweden, and colleagues, and the study spanned 36 primary care centers in Sweden.
“Our study is the first in the world to test an AI-based mobile app for melanoma in primary care in this way,” the authors said in a Swedish Research Council press release. “A great many studies have been done on previously collected images of skin lesions, and those studies relatively agree that AI is good at distinguishing dangerous from harmless ones.”
In this prospective multicenter clinical trial, the app demonstrated a 95.2% sensitivity and an 84.5% specificity in diagnosis. Additionally, nonmelanoma lesions were correctly identified with a 99.5% probability. The app double-checked its results by following the normal physician procedure, referring the patient to a dermatologist, or removing the lesion for tissue analysis.
According to the researchers, the next step is to expand a follow-up study in several countries to ensure the reliability of this instrument.
Disclosure: For full disclosures of the study authors, visit academic.oup.com.
British Journal of Dermatology
JNCCN Spotlights
