Shafaf Ibrahim

Seed-based region growing study for brain abnormalities segmentation

This paper proposes an empirical study of the efficiency of the Seed-Based Region Growing (SBRG) in segmentation of brain abnormalities. Presently, segmentation poses one of the most challenging problems in medical imaging. Segmentation of Magnetic Resonance Imaging (MRI) images is an important part of brain imaging research. In this paper, we used controlled experimental data as our testing data. The data is designed in such a way that prior knowledge of the size of the abnormalities are known. This is done by cutting various shapes and sizes of various abnormalities and pasting it onto normal brain tissues, where the tissues and the background are divided into different categories. The segmentation was done with twenty data of each category. The knowledge of the size of the abnormalities by the number of pixels were then used as the ground truth to compare with the SBRG segmentation results. The proposed SBRG technique was found to produce potential solutions to the current difficulties in detecting abnormalities in the human brain tissue area.

Empirical study of brain segmentation using particle swarm optimization

This study uses an empirical study of the efficiency of Particle swarm optimization (PSO) in segmentation of brain abnormalities. Presently, segmentation poses one of the most challenging problems in medical imaging. Segmentation of Magnetic Resonance Imaging (MRI) images is an important part of brain imaging research. In this study, we used controlled experimental data as our testing data. The data is designed which that prior knowledge of the size of the abnormalities are known. This is done by cutting various shapes and sizes of various abnormalities and pasting it onto normal brain tissues, where the tissues and the background are divided into different categories. The segmentation is done with twenty data of each category. The knowledge of the size of the abnormalities by number of pixels are then used as the ground truth to compare with the PSO segmentation results. The proposed PSO technique is found to produce potential solutions to the current difficulties in detecting abnormalities in human brain tissue area as it produced promising segmentation outcomes for light abnormalities. Nevertheless, the PSO produced poor performance in dark abnormalities segmentation as it produces low correlation values in all conditions.

Adaptive Neuro-Fuzzy Inference System for brain abnormality segmentation

This paper studies the application of the Adaptive Neuro-Fuzzy Inference System (ANFIS) for segmentation of brain abnormality in MRI images. Segmentation of MRI image is an important part of brain imaging research. In this study, 150 MRI images were used as testing data for the system. The data was created by combining the shapes and size of various abnormalities and pasting it onto normal brain image. Several types of backgrounds were tested — low, medium and high grey levels. The experimental results show good segmentation for medium and low background levels value for both light and dark abnormality levels over different backgrounds.

Qualitative Analysis of Skull Stripping Accuracy for MRI Brain Images

Skull stripping isolates brain from the non-brain tissues. It supplies major significance in medical and image processing fields. Nevertheless, the manual process of skull stripping is challenging due to the complexity of images, time consuming and prone to human errors. This paper proposes a qualitative analysis of skull stripping accuracy for Magnetic Resonance Imaging (MRI) brain images. Skull stripping of eighty MRI images is performed using Seed-Based Region Growing (SBRG). The skull stripped images are then presented to three experienced radiologists for visual qualitative evaluation. The level of accuracy is divided into five categories of “over delineation”, “less delineation”, “slightly over delineation”, “slightly less delineation” and “correct delineation”. Primitive statistical methods are calculated to examine the skull stripping performances. In another note, Fleiss Kappa statistical analysis is used to measure the agreement among radiologists. The qualitative performances analysis proved that the SBRG is an effective technique for skull stripping.

Radiographers agreement on skull stripping accuracy for MRI brain images

Skull stripping is the process of isolating brain from non-brain tissues. It supplies major significance in both medical and image processing fields. Nevertheless, the manual process of skull stripping is challenging due to the complexity of images, time consuming and prone to human errors. This paper proposes a qualitative analysis of skull stripping accuracy for Magnetic Resonance Imaging (MRI) brain images. The skull stripping of eighty MRI images is performed using Seed-Based Region Growing (SBRG). The skull stripped images are then presented to three experienced radiographers to visually evaluate the level of skull stripping accuracy. The level of accuracy is divided into five categories which are “over delineation”, “less delineation”, “slightly over delineation”, “slightly less delineation” and “correct delineation”. Primitive statistical methods of mode, mean and standard deviation are calculated to examine the qualitative performances of skull stripping capability. In another note, Fleiss Kappa statistical analysis is used to measure the agreement among the radiographers. The qualitative performances analysis proved that the SBRG is an effective technique for skull stripping. The reliability of agreement significances among the radiographers is found to be substantial.