Brain Tumor Classification Using Deep Neural Network and Transfer Learning.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Histopathological analysis for brain tumor classification is time-consuming and labor-intensive.
• Deep learning, particularly Convolutional Neural Networks (CNNs), offers a promising alternative for medical image analysis.

Purpose of the Study:

• To develop and evaluate a novel CNN model for accurate classification of brain tumors as benign or malignant using MRI images.
• To leverage transfer learning to enhance the performance of the CNN model.

Main Methods:

• A novel Convolutional Neural Network (CNN) architecture was designed, incorporating transfer learning.
• The proposed model was trained and validated on MRI images for brain tumor classification.
• Performance was benchmarked against established pre-trained networks: Res-Net, Alex-Net, U-Net, and VGG-16.

Main Results:

• The proposed CNN model, utilizing an improved Res-Net 50 architecture, achieved high classification accuracies: 99.30% for benign and 98.40% for malignant tumors.
• Significant improvements were observed in prediction accuracy, precision, recall, and F1-score compared to existing methods.
• The system demonstrated enhanced image fusion quality.

Conclusions:

• The novel CNN model provides a highly accurate and efficient method for brain tumor classification from MRI data.
• This deep learning approach has the potential to significantly aid clinicians in achieving more accurate and timely diagnoses.
• The developed system represents a substantial advancement over traditional histopathological methods and existing deep learning models.