SkinHealthMate app: An AI-powered digital platform for skin disease diagnosis

Abstract

Accurate diagnosis of skin diseases remains a significant challenge due to the inherent limitations of traditional visual and manual examination methods. These conventional approaches, while essential to dermatological practice, are prone to misdiagnoses and delays in treatment, particularly for conditions like skin cancer. To address these gaps, this paper presents the SkinHealth App, an innovative AI-driven solution that enhances the accuracy and efficiency of skin disease diagnosis. The app integrates a robust ensemble learning model, combining the strengths of EfficientNetB1 and EfficientNetB5 architectures. This ensemble model improves disease classification performance through advanced image processing techniques such as noise reduction and data augmentation. The key contributions of this work include the development of a scalable and secure server-side structure that ensures the safe handling of patient data and efficient processing of diagnostic queries. Experimental results on the HAM10000 dataset demonstrate the model's superior performance, achieving an accuracy of 93%, along with high precision and recall scores, thereby reducing false positives and false negatives. These outcomes clearly establish the app's potential to enhance dermatological diagnosis by providing timely and accurate disease identification. Ultimately, this work bridges the gap between traditional diagnostic methods and modern AI-driven technology, offering a transformative tool for improving patient care in dermatology.

Keywords

• Artificial intelligence
• Dermatology ensemble learning
• Skin disease classification
• Digital health platforms