Multi-modal medical image classification using deep residual network and genetic algorithm

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Deep Learning

Background:

• Accurate medical image classification is crucial for diagnosis and treatment planning.
• Conventional methods struggle with the semantic gap, requiring manual feature extraction.
• Deep learning, particularly Convolutional Neural Networks (CNNs), shows promise in overcoming these limitations.

Purpose of the Study:

• To bridge the semantic gap in medical image classification.
• To enhance classification performance for multi-modal medical images using deep learning.
• To evaluate the efficacy of the ResNet50 model for this task.

Main Methods:

• Utilized a deep learning-based model, ResNet50.
• Trained and validated the model on a dataset of 28,378 multi-modal medical images.
• Evaluated performance using accuracy, precision, recall, and F1-score.

Main Results:

• The proposed ResNet50 model achieved an overall accuracy of 98.61%.
• The model demonstrated superior classification performance compared to other state-of-the-art methods.
• Key evaluation parameters (accuracy, precision, recall, F1-score) confirmed the model's effectiveness.

Conclusions:

• Deep learning, specifically the ResNet50 model, effectively bridges the semantic gap in medical image classification.
• The developed model offers a significant advancement in accurate and consistent diagnostic decision-making.
• This research directly benefits healthcare services through improved diagnostic tools.