CAPSOCA: hybrid technique for nosologic segmentation of primary brain tumors

Abstract

Detection of primary brain tumors is inspired by the necessity of high accuracy as it deals with human life. Various imaging modalities techniques have incarnated as a tool in diagnosis and treatment domain. Yet, experienced and competent medical practitioners for the proper interpretation are still required. Thus, the involvement of information technology is highly demanded in introducing reliable and accurate computer systems. This study presents an algorithm for nosologic segmentation of primary brain tumors on Magnetic Resonance Imaging (MRI) brain images. Nosologic refers to the classification of diseases that can facilitate the diagnosis of neurological diseases. The purpose of segmentation is to highlight the tumor areas, whereas classification is used to identify the type of the primary brain tumors. For this purpose, an algorithm which hybridized the Grey Level <span style="text-decoration: underline;">C</span>o- occurrence Matrices (GLCM), Intensity Based Analysis (IBA), <span style="text-decoration: underline;">A</span>daptive Network-based Fuzzy Inference System (ANFIS) and <span style="text-decoration: underline;">P</span>article <span style="text-decoration: underline;">S</span>warm <span style="text-decoration: underline;">O</span>ptimization (PSO) <span style="text-decoration: underline;">C</span>lustering <span style="text-decoration: underline;">A</span>lgorithm (CAPSOCA) is proposed. The combination of several computer vision techniques is aim to deliver reproducible nosologic segmentation of primary brain tumors which are gliomas and meningiomas. The performance of the CAPSOCA is quantified by two measurements which are segmentation and classification accuracy. The segmentation accuracy is evaluated using comparison with ground truth approach. On the other hand, the classification accuracy is quantified using a truth table by comparing the classification outcomes with histopathology diagnosis. Upon the testing conducted, the CAPSOCA was proven to be an effective algorithm for nosologic segmentation of primary brain tumors. It appeared to return 88.09% of overall mean accuracy for gliomas segmentation, 86.92% of overall mean accuracy for meningiomas segmentation. In another note, 83.72% and 85.19% of classification accuracy for gliomas and meningiomas were observed.

Comparison of meta-heuristic algorithms for fuzzy modelling of COVID-19 illness’ severity classification

“…Since its inception in 1995, PSO has been successfully employed as a solution to a variety of function optimization problems or problems that able to be turned into function optimization problems [116]. Due to its lower memory needs and superior performance in offering solutions that are closer to the optimal for a variety of benchmarks and technical challenges such as computer vision [72], [93], [114], [120], [125]- [158]. PSO has become one of the most popular methods for tackling optimization problems [116][159]- [161].…”

Pre-trained classification of scalp conditions using image processing

“…The GLCM technique is found to be the most appropriate technique to be used due to its speed and less complexity. The four suitable GLCM texture properties extracted are contrast, correlation, energy and homogeneity [35].…”

Nearest-neighbor field algorithm based on patchMatch for myocardial perfusion motion estimation/correction

“…Therefore, and because of the slow acquisition nature of DMRI, it usually compromises the quality of output images. In other words, for accurate quantification of myocardial perfusion in MRI data, it is essential to have an extended slice, a suitable spatiotemporal resolution, coverage, and a high signal to noise ratio (SNR) [3], [4].…”