An Innovative Model for Detecting Brain Tumors and Glioblastoma Multiforme Disease Patterns

An Innovative Model for Detecting Brain Tumors and Glioblastoma Multiforme Disease Patterns

Peifang Guo (Independant Researcher, Montreal, Canada) and Prabir Bhattacharya (Morgan State University, Baltimore, MD, USA)
DOI: 10.4018/IJSSCI.2017100103

Abstract

In this article, an innovative model is proposed for detecting brain tumors and glioblastoma multiforme disease patterns (DBT-GBM) in medical imaging. The DBT-GBM model mainly includes five steps, the image conversion in the L* component of the L*a*b* space, an image sample region selection, calculation of the average values of colors, image pixel classification using the minimum distance classifier and the segmentation operation. In the approach, the minimum distance classifier is used to classify each pixel by calculating the Euclidean distance between that pixel and each color marker of the pattern. In the experiments, the authors implement the DBT-GBM model into real-time data, the samples of three anatomic sections of a T1w 3D MRI (axial, sagittal and coronal cross-sections) on the GBM-3D-Slicer datasets and the CBTC datasets. The implementation results show that the proposed DBT-GBM robustly detects the GBM disease patterns and cancer nuclei (involving the omics indicative of brain tumors pathologically) in medical imaging, leading to improved segmentation performance in comparison.
Article Preview

1. Introduction

In clinical studies, it has been found that brain tumors in adults occupy the 13th place in frequency of all cancers, and they are the fifth most common cause of cancer death in the under 65-year-old population because of their particularly poor prognosis in brain. In addition, the cells of brain tumors could grow quickly, and could spread throughout the brain (Bakas et al., 2017A; Bauer et al., 2013; Tasic et al., 2015; Pepe et al., 2013). Among brain tumors, the glioblastoma multiforme (GBM) is one of the most common and most aggressive malignant primary tumors in brain, where the incidence of the GBM is 2 to 3 per 100,000 people in the United States and Europe. The GBM accounts for 12% to 15% of all intracranial tumors and 50% to 60% of astrocytic tumors. In clinical, the treatments for the patients with brain tumors typically include maximum safe resection, percutaneous radiation, chemotherapy, surgical operation etc. (Menze et al., 2015; CBTC, 2015; Egger et al., 2013; Duyn, 2012).

Studies show that patients with brain tumors may benefit from early diagnoses in order to receive primary care or undergo surgical procedures to address their progressive symptoms (Bakas et al., 2017B; Iturria-Medina et al., 2016; Wang, 2013; Egger et al., 2013; Ahmed et al., 2015). Some modalities of biomarkers have been proved to be sensitive to brain tumors in the detection, including the measure of the volume of brain tumors through the MRI scans. Therefore, a robust model which could offer the analysis of volumes of disease patterns with systematic quantitative measurements is important to help to pursue the major objective of carrying out mass screening campaigns of medical imaging in the early detection clinically (Guo, 2017A; Tohka, 2014; Wang et al., 2011; Masood et al., 2013).

Image segmentation refers to the partitioning of an image into disjoint regions with respect to a chosen property, for example, brain tissues, image patterns, textures of colors in MRI scans (Guo, 2017B; West et al., 2012; Yushkevich et al., 2006). Since the segmentation of cancer nuclei and detection of the GBM disease patterns is crucial early steps on supporting of brain tumor diagnosis, it has become one of the important areas of research in assess patients with brain tumor in medical community (Martono et al., 2016; Ahmed et al., 2015; Paul et al., 2015). This study seeks to address the challenge for providing a model for detecting brain tumors involving ‘omics’ indicative of brain tumors pathologically in digital pathology imaging, as well as the quantitative analysis of the GBS disease patterns in the samples of the anatomic sections of a T1w 3D MRI, in order to assist physicians in quantifying volumes of brain tumors in clinical.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2019): Forthcoming, Available for Pre-Order
Volume 10: 4 Issues (2018): 2 Released, 2 Forthcoming
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing