Segmentation of Brain Tumors in MRI Images Using Three-Dimensional Active Contour without Edge

Ali Hasan,Farid Meziane,Hamid Jalab,Rob Aspin

doi:10.3390/sym8110132

Ali Hasan, Farid Meziane + Show 2 more

Open Access

https://doi.org/10.3390/sym8110132

Copy DOI

Abstract

Brain tumor segmentation in magnetic resonance imaging (MRI) is considered a complex procedure because of the variability of tumor shapes and the complexity of determining the tumor location, size, and texture. Manual tumor segmentation is a time-consuming task highly prone to human error. Hence, this study proposes an automated method that can identify tumor slices and segment the tumor across all image slices in volumetric MRI brain scans. First, a set of algorithms in the pre-processing stage is used to clean and standardize the collected data. A modified gray-level co-occurrence matrix and Analysis of Variance (ANOVA) are employed for feature extraction and feature selection, respectively. A multi-layer perceptron neural network is adopted as a classifier, and a bounding 3D-box-based genetic algorithm is used to identify the location of pathological tissues in the MRI slices. Finally, the 3D active contour without edge is applied to segment the brain tumors in volumetric MRI scans. The experimental dataset consists of 165 patient images collected from the MRI Unit of Al-Kadhimiya Teaching Hospital in Iraq. Results of the tumor segmentation achieved an accuracy of 89% ± 4.7% compared with manual processes.

Highlights

Brain tumors are relatively less common than other neoplasms, such as those of the lung and breast, but are considered highly important because of prognostic effects and high morbidity [1].Clinical diagnosis, predicted prognosis, and treatment are significantly affected by the accurate detection and segmentation of brain tumors and stroke lesions [2].The Iraqi Ministry of Health reported that the use of depleted uranium and other toxic substances in the first and second Gulf Wars had increased the average annual number of registered cancerous brain tumor cases and birth defects since 1990 [3]
This study was conducted in collaboration with magnetic resonance imaging (MRI) units in Iraqi hospitals that have witnessed the high numbers of these cases
This research aimed to develop an automated method that can locate the initial contour of brain tumor segmentation across all axial slices of volumetric MRI brain scans

Summary

Introduction

Brain tumors are relatively less common than other neoplasms, such as those of the lung and breast, but are considered highly important because of prognostic effects and high morbidity [1].Clinical diagnosis, predicted prognosis, and treatment are significantly affected by the accurate detection and segmentation of brain tumors and stroke lesions [2].The Iraqi Ministry of Health reported that the use of depleted uranium and other toxic substances in the first and second Gulf Wars had increased the average annual number of registered cancerous brain tumor cases and birth defects since 1990 [3]. Image processing was embedded in medical systems and used widely in medicine, from diagnosis to therapy. This approach demonstrated an enhanced classification of pathological tissues that. Prior using knowledge suggests thatimage the background intensity values of MRI brain slices often can be achieved the unmodified [13]. Segmentation much higher number of pixels than that of the brain region but without meaningful information [13]. Histogram thresholding wasbackground used as a segmentation method to isolate the background. This approach is based on the thresholding of intensity values by a specific T value. A much higher number of pixels than that of the brain region but without meaningful information [13]

Objectives

Methods

Results

Conclusion