Classification of Skin Disease Using Deep Learning Neural Networks with MobileNet V2 and LSTM.

Area of Science:

• Artificial Intelligence
• Dermatology
• Medical Imaging Analysis

Background:

• Deep learning models excel at identifying complex patterns in data.
• Accurate and early skin disease diagnosis is crucial for effective treatment and patient outcomes.
• Existing diagnostic methods can be time-consuming and require specialized expertise.

Purpose of the Study:

• To develop and evaluate a deep learning-based system for automated skin disease classification.
• To leverage MobileNet V2 and Long Short Term Memory (LSTM) for precise image analysis and stateful predictions.
• To compare the proposed model's performance against established deep learning architectures.

Main Methods:

• Utilized a deep learning approach combining MobileNet V2 for feature extraction and LSTM for sequential data processing.
• Employed a grey-level co-occurrence matrix to monitor disease progression.
• Trained and validated the model on the HAM10000 dataset, comparing it with Fine-Tuned Neural Networks (FTNN), CNN, and VGG models.

Main Results:

• The proposed MobileNet V2-LSTM model achieved over 85% accuracy in skin disease classification.
• Demonstrated superior performance compared to FTNN, CNN, and VGG models.
• Showcased computational efficiency, requiring approximately half the computations of the conventional MobileNet model for faster recognition.

Conclusions:

• The developed deep learning system provides an accurate and efficient method for skin disease diagnosis.
• The system's robustness and speed, coupled with a mobile application, facilitate early detection and management of skin conditions.
• This technology can significantly aid general practitioners in diagnosing skin diseases, reducing complications and improving patient morbidity.