Brain tumor detection empowered with ensemble deep learning approaches from MRI scan images

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Early brain tumor detection is crucial for effective treatment and improved patient outcomes.
• Existing artificial intelligence (AI) methods for brain tumor detection face challenges in accuracy and reliability.
• Accurate classification of brain MRI scans into types like pituitary, meningioma, and glioma is vital.

Purpose of the Study:

• To investigate a potent artificial intelligence (AI) technique for brain tumor detection using MRI scans.
• To enhance the accuracy and dependability of AI-based brain tumor classification.
• To evaluate a novel AI approach combining two distinct deep learning models.

Main Methods:

• A novel AI technique was developed by combining two deep learning models: InceptionV3 and Xception.
• The combined AI model was trained using a dataset of brain MRI scan images.
• Performance was assessed using key metrics including accuracy, precision, and dependability.

Main Results:

• The combined InceptionV3 + Xception AI model achieved a training accuracy of 98.50% and a validation accuracy of 98.30%.
• The integrated AI approach demonstrated superior performance compared to individual deep learning models.
• The system effectively identified and categorized different types of brain tumors from MRI scans.

Conclusions:

• The study highlights the potential of advanced AI techniques in medical imaging for brain tumor detection.
• Concurrent use of multiple AI models significantly enhances the accuracy and reliability of brain tumor diagnosis.
• This combined AI approach offers a promising tool for improving early detection and classification of brain tumors.