Brain Tumors Classification for MR images based on Attention Guided Deep Learning Model.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Neuro-oncology

Background:

• Magnetic Resonance Imaging (MRI) is crucial for high-resolution brain tumor imaging.
• Manual analysis of MRI scans is time-intensive and prone to human error.
• Existing deep learning models excel at tumor detection but often fail to differentiate primary from secondary intracranial tumors.

Purpose of the Study:

• To develop an advanced deep learning model for automated brain tumor diagnosis.
• To enhance tumor detection accuracy and efficiency in MRI scans.
• To accurately distinguish between primary and secondary intracranial tumors.

Main Methods:

• An attention-guided deep Convolution Neural Network (CNN) was developed.
• The model was trained and validated using Magnetic Resonance Imaging (MRI) data.
• Ten-fold cross-validation was employed to assess model performance.

Main Results:

• The proposed CNN model achieved an average accuracy of 99.18% for brain tumor detection.
• The model demonstrated an average accuracy of 83.38% in distinguishing primary from secondary intracranial tumors.
• Performance metrics indicate the model's superiority over existing methods.

Conclusions:

• The attention-guided CNN model offers a highly accurate and efficient solution for brain tumor diagnosis from MRI.
• The model's ability to differentiate tumor origins shows significant clinical potential.
• This AI-driven approach shows promise in assisting medical experts in neuro-oncology.