AI-Driven Skin Disease Classification Using Deep LSTM Networks Optimized by Tunicate Swarm Algorithm

Abstract

This study presents an innovative methodology for skin disease classification by integrating Deep Long Short-Term Memory (Deep LSTM) networks with the Tunicate Swarm Algorithm (TSA). The proposed hybrid model significantly enhances the accuracy of dermatological diagnosis using medical images. Deep LSTM networks are employed to effectively capture complex spatial and temporal patterns inherent in dermatological data, while TSA is utilized to fine-tune feature extraction and optimize model parameters, leading to improved performance and computational efficiency. This fusion of deep learning and swarm intelligence introduces a powerful framework for intelligent, automated skin disease detection in clinical settings. The experimental results demonstrate that the synergistic combination of Deep LSTM and TSA outperforms traditional classification approaches, underscoring the potential of AI-driven models in advancing precision healthcare.