Computer-Aided Brain Tumor Diagnosis: Performance Evaluation of Deep Learner CNN Using Augmented Brain MRI

Area of Science:

• Neurology
• Medical Imaging
• Artificial Intelligence

Background:

• Brain tumors are a serious neurological disease with increasing mortality.
• Manual analysis of Magnetic Resonance Images (MRIs) is insufficient for accurate and timely brain tumor diagnosis.
• Early detection is crucial for effective treatment and improved patient survival rates.

Purpose of the Study:

• To develop and evaluate a Convolutional Neural Network (CNN) for the early and accurate diagnosis of brain tumors using MRI.
• To enhance the state-of-the-art in computer-aided diagnosis (CAD) for brain tumors.
• To improve the sustainability of diagnostic models for unseen data.

Main Methods:

• A CNN model was trained on the BR35H benchmark dataset of brain tumor MRIs.
• Geometric data augmentation and statistical standardization were employed to improve model performance and generalization.
• The model's performance was rigorously evaluated on six diverse datasets: BMI-I, BTI, BMI-II, BTS, BMI-III, and BD-BT.

Main Results:

• The proposed CNN-based CAD system achieved an average accuracy of approximately 98.8% and a specificity of 0.99.
• The system demonstrated 100% diagnostic accuracy on two specific datasets: BTS and BD-BT.
• Comparative analysis confirmed that the proposed system outperforms existing brain tumor diagnosis systems.

Conclusions:

• The developed CNN model shows significant promise for accurate and early brain tumor diagnosis.
• The use of data augmentation and standardization enhances the model's robustness for real-world applications.
• This AI-driven approach has the potential to improve patient outcomes by enabling timely treatment decisions.