VGG19-ResSE: an optimized hybrid model for accurate segmentation and classification of vitiligo lesions

Skin diseases pose a significant public health challenge, impacting millions of people globally with a wide range of phenomena, from mild conditions to life-threatening. In skin diseases, people are well known with an autoimmune disorder, which results from abnormal responses of the adaptive body’s immune system. Vitiligo is also a part of chronic autoimmune disorder, where a particular area of human skin loses its original colors, and people are affected in different ratios worldwide. This study investigates the effectiveness of deep learning models for detecting and segmenting vitiligo lesions in dermatological images. Two models, VGG19-ResSE and YOLOv8, were evaluated for their performance. The VGG19-ResSE model achieved an accuracy of 95.5% and a Dice coefficient of 0.92 in segmentation tasks. It enhances performance by using advanced features, including squeeze-and-excitation blocks and residual connections. The YOLOv8 model showed strong real-time detection capabilities, achieving an F1-score of 0.92. However, it could have been more effective in segmentation compared to VGG19-ResSE. The results indicate that while YOLOv8 is suitable for rapid detection, the VGG19-ResSE model is more appropriate for detailed analysis of vitiligo lesions. This study highlights the potential of deep learning in improving diagnostic accuracy in dermatology.