Brain Tumor Detection Using Depth-First Search Tree Segmentation

Area of Science:

• Medical Image Processing
• Computational Anatomy
• Graph Theory Applications

Background:

• Medical image processing enables visual perception of anatomical abnormalities.
• Image segmentation is crucial for identifying substances and their margins in medical images.
• Accurate brain tumor diagnosis via Magnetic Resonance Imaging (MRI) necessitates precise segmentation due to large data volumes.

Purpose of the Study:

• To present an automated segmentation technique for precise brain tumor detection in MRI.
• To introduce a novel Depth-First Search (DFS) segmentation algorithm based on graph theory.
• To compare the proposed algorithm's performance with existing systems and classifiers.

Main Methods:

• Pixels in MRI scans are organized into a tree-like structure based on proximity using a graph theory approach.
• A Depth-First Search (DFS) algorithm is employed for image segmentation.
• Performance evaluation includes comparison with other systems and assessment of Adaptive Neuro-Fuzzy Inference System (ANFIS) and Support Vector Machine (SVM) classifiers.

Main Results:

• The proposed DFS segmentation algorithm effectively distinguishes healthy cells from brain tumor-affected cells.
• Experimental results demonstrate reduced computational complexity compared to existing methods.
• Enhanced accuracy in tumor segmentation was achieved with the proposed approach.

Conclusions:

• The developed DFS-based segmentation algorithm offers an accurate and computationally efficient solution for brain tumor analysis in MRI.
• This method aids in obtaining precise information crucial for effective treatment planning.
• The study highlights the potential of graph theory-based image processing for medical diagnostics.