Medical Image Classifications for 6G IoT-Enabled Smart Health Systems.

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Healthcare
• 6G Technology Applications

Background:

• Massive data generation in 6G-enabled Internet of Medical Things (IoMT) necessitates improved medical diagnosis.
• Current diagnostic methods require enhancement for accuracy and real-time processing.

Purpose of the Study:

• To present a novel framework for real-time medical diagnosis within the 6G-enabled IoMT.
• To improve prediction accuracy using integrated deep learning and optimization techniques.

Main Methods:

• Preprocessing computed tomography images and converting them into feature vectors using a neural network.
• Employing the MobileNetV3 architecture for feature extraction.
• Enhancing the Arithmetic Optimization Algorithm (AOA) with Hunger Games Search (HGS) operators, creating the AOAHG algorithm for feature selection and classification.

Main Results:

• The AOAHG algorithm demonstrated superior performance in feature selection and model classification.
• Experiments on ISIC-2016, PH2, WBC, and OCT datasets showed significant improvements over existing methods.
• Achieved high performance metrics, including accuracy, precision, recall, and F1-score, with specific examples like 99.69% for OCT classification.

Conclusions:

• The proposed 6G-enabled IoMT framework effectively enhances medical diagnosis accuracy and real-time capabilities.
• The AOAHG optimization technique significantly boosts classification performance in medical image analysis.
• The framework shows strong potential for advancing healthcare diagnostics through advanced AI and communication technologies.