Efficient Framework for Identifying, Locating, Detecting and Classifying MRI Brain Tumor in MRI Images

Area of Science:

• Medical Imaging
• Computer Vision
• Image Processing

Background:

• Brain tumors necessitate precise identification and removal.
• Magnetic Resonance Imaging (MRI) is crucial for visualizing brain structures.
• Accurate segmentation of MRI slices is challenging but vital for tumor characterization.

Purpose of the Study:

• To present a novel Adaptive Convex Region Contour (ACRC) algorithm for improved brain tumor segmentation.
• To enable accurate 3D reconstruction of tumors from 2D MRI slices.
• To enhance tumor volume estimation for surgical guidance.

Main Methods:

• Utilizing Support Vector Machine (SVM) for classifying MRI slices as normal or abnormal.
• Applying the Adaptive Convex Region Contour (ACRC) algorithm for segmenting abnormal slices.
• Employing the Rapid Mode Image Matching (RMIM) algorithm for 3D reconstruction from segmented slices.

Main Results:

• The ACRC algorithm effectively segments abnormal brain tissues from MRI slices.
• 3D reconstruction using RMIM provides a clear visualization of tumor shape and size.
• The proposed method demonstrated superior accuracy in tumor volume estimation compared to existing techniques.

Conclusions:

• The ACRC algorithm combined with 3D reconstruction offers a significant advancement in brain tumor analysis.
• Accurate tumor segmentation and volume estimation are crucial for effective surgical intervention.
• This approach aids neurosurgeons by providing precise anatomical information for treatment planning.