Comparative Study of Probabilistic Atlas and Deep Learning Approaches for Automatic Brain Tissue Segmentation from MRI Using N4 Bias Field Correction and Anisotropic Diffusion Pre-processing Techniques

TL;DR

This study compares traditional probabilistic atlas and modern deep learning models like nnU-Net for brain tissue segmentation from MRI, demonstrating nnU-Net's superior performance when combined with advanced pre-processing techniques.

Contribution

It provides a comprehensive comparison of segmentation methods using pre-processing techniques, highlighting the effectiveness of nnU-Net models for brain tissue segmentation.

Findings

3D nnU-Net achieved highest Dice score (0.937)

2D nnU-Net had lowest Hausdorff Distance (5.005 mm)

nnU-Net models outperform traditional methods with pre-processing

Abstract

Automatic brain tissue segmentation from Magnetic Resonance Imaging (MRI) images is vital for accurate diagnosis and further analysis in medical imaging. Despite advancements in segmentation techniques, a comprehensive comparison between traditional statistical methods and modern deep learning approaches using pre-processing techniques like N4 Bias Field Correction and Anisotropic Diffusion remains underexplored. This study provides a comparative analysis of various segmentation models, including Probabilistic ATLAS, U-Net, nnU-Net, and LinkNet, enhanced with these pre-processing techniques to segment brain tissues (white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF)) on the Internet Brain Segmentation Repository (IBSR18) dataset. Our results demonstrate that the 3D nnU-Net model outperforms others, achieving the highest mean Dice Coefficient score (0.937 +- 0.012), while the 2D nnU-Net model recorded the lowest mean Hausdorff Distance (5.005 +- 0.343 mm) and the lowest mean Absolute Volumetric Difference (3.695 +- 2.931 mm) across five unseen test samples. The findings highlight the superiority of nnU-Net models in brain tissue segmentation, particularly when combined with N4 Bias Field Correction and Anisotropic Diffusion pre-processing techniques. Our implemented code can be accessed via GitHub.