Scalp Disease Detection Using Deep Learning: A Hybrid Ensemble Approach with EfficientNetB0, DenseNet121 and MobileNetV2
TLDR The hybrid ensemble model improves scalp disease detection accuracy and consistency.
The study presents a hybrid ensemble approach using deep learning models EfficientNetB0, DenseNet121, and MobileNetV2 for detecting scalp diseases. This method leverages transfer learning and combines the outputs of these models through a soft voting mechanism to improve prediction accuracy and consistency. The ensemble model outperforms individual models, especially in scenarios with varied or limited data, making it suitable for real-world applications. It aims to enhance early detection of scalp conditions, supporting medical decisions in areas with limited resources.
