Content area
Full Text
Received Jan 5, 2018; Revised Mar 6, 2018; Accepted Apr 2, 2018
This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
1. Introduction
Gliomas are the most common primary brain tumors, characterized by the uncontrolled proliferation of abnormal brain cells. This disease is one of the most common causes of cancer death in men and women [1]. According to classification of the World Health Organization (WHO), gliomas can be subdivided by their malignancy into grade II (lower grade) to grade IV (high grade) [2]. The clinical patients with high-grade gliomas, such as glioblastomas, have the median overall survival rate of 23.1 months, the 2-year survival rate of 47.4%, and the 4-year survival rate of 18.5% [3]. On the contrast, the slower growing low-grade gliomas, such as astrocytomas and oligodendrogliomas, come with an overall 10-year survival rate of 57% [4]. Therefore, early detection is considered as an effective way to get a hopeful prognosis.
Modern imaging techniques allow clinicians and radiologists to evaluate the progression of tumors and choose optimal treatment strategy, without invasive neurosurgery. There are many imaging modalities that can be used to study the brain, such as computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI). These imaging modalities can provide effective and reliable information about brain tissues. In view of the advantages of high soft tissues contrast and high spatial resolution, MRI is widely used to evaluate the tumor heterogeneity [5]. However, there are several common MRI modalities including T1-weighted (T1), T2-weighted (T2), gadolinium enhanced T1-weighted (T1c), and Fluid-Attenuated Inversion Recovery (FLAIR), which generate a large number of medical images. This has become a huge burden for radiologists, resulting in inaccurate detection or misinterpretation. Therefore, with the development of computer technology, there is an increasing demand for computer-aided diagnosis.
Recently, computer-aided detection diagnosis (CAD) has gradually become a research hotspot in the area of medical imaging. The main idea of CAD is to assist radiologists in making clinical decision by using the report of computer system as a “second opinion” [6]. El-Dahshan et al. [6] developed a CAD system that used feedback pulse-coupled neural network for image segmentation,...
| |