Superlative Feature Selection Based Image Classification Using Deep Learning in Medical Imaging.

Area of Science:

• Medical imaging analysis
• Artificial intelligence in healthcare
• Diagnostic technology

Background:

• Medical image recognition is crucial for disease forecasting and early detection.
• Accurate image categorization is challenging but vital for patient health management.
• Deep learning (DL) offers potential for improving diagnostic accuracy.

Purpose of the Study:

• To develop an optimized feature extraction model for effective medical imaging categorization.
• To enhance a clinical classifier using deep learning for improved diagnostic performance.
• To increase the precision, recall, F1 score, and accuracy of medical image detection.

Main Methods:

• Utilized a deep learning (DL) model incorporating preprocessing, feature extraction, and classification.
• Employed the pretrained EfficientNetB0 model for optimal feature extraction.
• Developed an enhanced classifier based on Feature Selection oriented Clinical Classifier.

Main Results:

• The optimized features significantly improved classifier performance.
• Achieved high precision, recall, F1 score, and accuracy in medical image detection.
• The presented approach demonstrated superior performance, reaching 98% accuracy.

Conclusions:

• The proposed DL-based approach effectively enhances medical image categorization.
• Optimized feature extraction using EfficientNetB0 leads to improved diagnostic accuracy.
• This method holds promise for advancing early disease identification and patient care.