The accuracy of the preliminary diagnosis of benign and malignant melanocytic skin tumors using the “Derma Onko Check” AI-based smartphone application

Abstract

Objective. To evaluate the diagnostic accuracy of an artificial intelligence (AI)-based smartphone application “Derma Onko Check” for differentiating benign and malignant melanocytic skin neoplasms during primary care visits. Material and methods. The efficacy of the AI-based “Derma Onko Check” software for visual recognition and differential diagnosis of benign versus malignant melanocytic tumor of skin during primary care visits was evaluated in 224 patients aged 22–90 years. The AIgenerated medical conclusions were compared against both preliminary clinical diagnoses and final diagnoses established by experienced dermatovenereologists. Results. “Derma Onko Check” achieved high diagnostic accuracy for benign and malignant cutaneous neoplasms during preliminary assessment. Performance metrics (accuracy: 93.3%; sensitivity: 94.7%; specificity: 93.0%) were comparable to these of dermatovenereologists (accuracy: 95.6%; sensitivity: 100%; specificity: 100%). Conclusion. The AI-based smartphone application “Derma Onko Check” is a practical and efficient tool for differential diagnosis between benign and malignant melanocytic neoplasms, demonstrating high diagnostic accuracy during primary care examinations. This AI-based software helps optimize patient triage pathways with malignant skin lesions.

Ähnliche Arbeiten