Brain tumor segmentation on Multimodal MRI scans using EMAP Algorithm.

Area of Science:

• Medical Imaging
• Computer Vision
• Artificial Intelligence

Background:

• Digital image processing is crucial for clinical applications and machine vision.
• Manual tumor segmentation in brain MRI is time-consuming and prone to inaccuracies.
• Automated segmentation methods are needed to handle the increasing volume of MRI data.

Purpose of the Study:

• To propose a fully automated method for brain tumor segmentation in volumetric MRI images.
• To accurately identify whole tumor and sub-tumor regions (core, enhancing, non-enhancing).
• To segment both high-grade glioma (HGG) and low-grade glioma (LGG) without a training database.

Main Methods:

• Utilizes image processing techniques based on expectation maximization and K-mean clustering.
• Estimates tumor regions independently, requiring no training data.
• Applies the algorithm to sequences of MRI volumes for comprehensive analysis.

Main Results:

• Achieved an average DICE score of 0.92 on the BRATS 2015 dataset.
• Demonstrated comparable performance to state-of-the-art, computationally expensive algorithms.
• Successfully segmented both HGG and LGG tumors and their sub-regions.

Conclusions:

• The proposed automated method offers an efficient and accurate solution for brain tumor segmentation in MRI.
• The algorithm's independence from training databases makes it broadly applicable.
• This technique provides a valuable tool for clinical diagnosis and research in neuro-oncology.