Automated Feature Extraction in Brain Tumor by Magnetic Resonance Imaging Using Gaussian Mixture Models.

Ahmad Chaddad

doi:10.1155/2015/868031

Abstract

This paper presents a novel method for Glioblastoma (GBM) feature extraction based on Gaussian mixture model (GMM) features using MRI. We addressed the task of the new features to identify GBM using T1 and T2 weighted images (T1-WI, T2-WI) and Fluid-Attenuated Inversion Recovery (FLAIR) MR images. A pathologic area was detected using multithresholding segmentation with morphological operations of MR images. Multiclassifier techniques were considered to evaluate the performance of the feature based scheme in terms of its capability to discriminate GBM and normal tissue. GMM features demonstrated the best performance by the comparative study using principal component analysis (PCA) and wavelet based features. For the T1-WI, the accuracy performance was 97.05% (AUC = 92.73%) with 0.00% missed detection and 2.95% false alarm. In the T2-WI, the same accuracy (97.05%, AUC = 91.70%) value was achieved with 2.95% missed detection and 0.00% false alarm. In FLAIR mode the accuracy decreased to 94.11% (AUC = 95.85%) with 0.00% missed detection and 5.89% false alarm. These experimental results are promising to enhance the characteristics of heterogeneity and hence early treatment of GBM.

Highlights

Providing quantitative and accurate information for medical diagnosis, Magnetic Resonance Imaging (MRI) plays an essential role in medical imaging [1]
We focused on features derived from Gaussian mixture model (GMM) analysis on both weighted T1 and T2 and Fluid-Attenuated Inversion Recovery (FLAIR) sequences
The schematic of the proposed method is shown in Figure 2: (1) preprocessing to normalize grayscales and filtering to remove the noise from images in the three MRI sequences, T1-WI, T2-WI, and FLAIR; (2) tumor (GBM) areas detection by multithresholding segmentation and normal areas determined from the normal brain material; (3) feature extraction from the GMM curve fitting of the grayscale histogram on T1WI, T2-WI, and FLAIR images; (4) applying three classifier techniques to discriminate between the tumor areas and normal areas based on GMM features; and (5) validating the effect of GMM features by comparative study with principal component analysis (PCA) and wavelet features

Summary

Introduction

Providing quantitative and accurate information for medical diagnosis, Magnetic Resonance Imaging (MRI) plays an essential role in medical imaging [1]. A bottleneck of MR image processing arises from variations in intensity due to B1 and B0 field inhomogeneity [4, 5] This is manifested by the nonuniform appearance even of a single tissue which may mislead image analysis algorithms, which enhance abnormality area detection by a segmentation model [2, 6]. Despite the ongoing research and clinical trials, GBM remains one of the most aggressive malignant tumors with less than 5% of patients surviving five years after diagnosis [10]. This is attributed to the highly infiltrative nature and the heterogeneity that Glioblastoma exhibits on molecular and genomic levels which lead to differences in individual treatment response and prognosis [11]

Objectives

Methods

Results

Conclusion