Evaluation of an Artificial Intelligence Technology for the Diagnosis of Pigmented Skin Lesions in Egyptian Patients | ||
| Al-Azhar International Medical Journal | ||
| Volume 2025, Issue 5, May 2025, Pages 40-47 PDF (685.85 K) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/aimj.2025.446551 | ||
| Authors | ||
| Amr Mohamed Zaky; Hassan Mamdoh Abd-Alaziz; Mohamed Ahmed Sayed Abd-Alaziz | ||
| Dermatology, Venereology and Andrology, Faculty of Medicine for Boys, Al-Azhar University, Cairo, Egypt | ||
| Abstract | ||
| Background: Pigmented skin lesions are that lesions which cause colored spots and may classified to Benign or malignant lesions. It is very important to differentiate between them to develop a suitable treatment. Previous literature reported a new technology that could be used to diagnosis the pigmented skin lesion based on artificial intelligence(AI). Aim and objectives: To assess the efficacy of artificial intelligence-based tools for the evaluation of pigmented skin lesions, both malignant and benign, and their ability to predict possible differential diagnoses. Subjects and methods: This cross-sectional study was carried out on 200 participants with different pigmented skin lesions who underwent skin biopsy after taking a photo session of lesions. The photograph was assessed by AI tools, and the results were compared with the results of the biopsy. The result of AI tools was evaluated depending on two aspects: the ability to differentiate between benign and malignant lesions, and the ability to predict the correct diagnosis within the predicted differential diagnosis. Results: Biopsy results revealed 50% benign and 50% malignant lesions, with seborrheic keratosis(48%) and basal cell carcinoma(51%) being the most common. All AI tools showed strong agreement with histopathology(k-values ranging between 0.64-0.89), particularly in identifying pigmented skin lesion. Conclusion: AI tools for screening pigmented skin lesions improve diagnostic accuracy, helping healthcare providers, especially those with less experience, to achieve expert-level assessments. They are transforming dermatology by enhancing diagnostic precision and improving patient outcomes through timely interventions and efficient care. | ||
| Keywords | ||
| Pigmented skin lesions; Artificial intelligence | ||
|
Statistics Article View: 1 |
||
