Aboul Ella Hassanien (Abo)

The objective of this paper is to evaluate a new combined approach intended for reliable CT liver image segmentation, to separate the liver from other organs, and segment the liver into a set of regions of interest (ROIs). The approach combines the level set with watershed approach used as post segmentation step to produce a reliable segmentation result. Features of first order statistics and grey-level cooccurrence matrix, are calculated and passed to an artificial neural network, to be trained and to classify infected regions. Filtering is used before the segmentation approach to enhance contrast, remove noise and emphasize certain features, as well as connecting ribs around the liver. To evaluate the performance of presented approach, we performed many tests on different CT liver images. The experimental results obtained, show that the overall accuracy offered by the proposed approach is 92.1% in segmenting CT liver images into set of regions even with noise, and 88.9% average accuracy for neural network classification.

The aim of geophysical inverse problem is to determine the
spatial distribution and depths to buried targets at a variety of scales;
it ranges from few centimetres to many kilometres. To identify ore bodies,
extension of archaeological targets, old mines, unexploded ordnance
(UXO) and oil traps, the linear geomagnetic inverse problem resulted
from the Fredholm integral equation of the first kind is solved using
many strategies. The solution is usually affected by the condition of
the kernel matrix of the linear system and the noise level in the data
collected. In this paper, regularized Kaczmarz method is used to get a
regularized solution. This solution is taken as an initial solution to bat
swarm algorithm (BA) as a global swarm-based optimizer to refine the
quality and reach a plausible model. To test efficiency, the proposed hybrid
method is applied to different synthetic examples of different noise
levels and different dimensions and proved an advance over using the
Kaczmarz method.

Orphan drugs are a treatment for rare diseases. From that, comes the importance of orphan drug development and discovery. For an orphan drug to be approved by the FDA, it does not have to be similar to any approved orphan drug. So chemists opinions are important to determine the probability of similarity. It is too hard to check all orphan drugs for any rare disease. It takes a long time and big effort, so we introduce in this study a system that classifies the orphan drugs according to their probability of structural similarity. It also compares between them and the unauthorized orphan drug to determine the closest orphan drug to it. That system helps chemists to study a certain orphan database using the five features. That system provides better results. It provides chemists with the clusters of orphan drugs after adding the drug that needs to be authorized to its cluster.

n/a

Fish image segmentation can be considered an essential process in developing a system for fish recognition. This task is challenging as different specimens, rotations, positions, illuminations, and backgrounds exist in fish images. In this research, a segmentation model is proposed for fish images using Salp Swarm Algorithm (SSA). The segmentation is formulated using Simple Linear Iterative Clustering (SLIC) method with initial parameters optimized by the SSA. The SLIC method is used to cluster image pixels to generate compact and nearly uniform superpixels. Finally, a thresholding using Otsu’s method helped to produce satisfactory results of extracted fishes from the original images under different conditions. A fish dataset consisting of real-world images was tested. In experiments, the proposed model shows robustness for different cases compared to conventional work.

Face extraction is considered a very important step in devel-
oping a recognition system. It is a challenging task as there are dierent
face expressions, rotations, and artifacts including glasses and hats. In
this paper, a face extraction model is proposed for thermal IR human face
images based on superpixel technique. Superpixels can improve the com-
putational eciency of algorithms as it reduces hundreds of thousands of
pixels to at most a few thousand superpixels. Superpixels in this paper
are formulated using the quick-shift method. The Quick-Shift's superpix-
els and automatic thresholding using a simple Otsu's thresholding help
to produce good results of extracting faces from the thermal images. To
evaluate our approach, 18 persons with 22,784 thermal images were used
from the Terravic Facial IR Database. The Experimental results showed
that the proposed model was robust against image illumination, face
rotations, and dierent artifacts in many cases compared to the most
related work.

Illumination factors such as shading, shadow, and highlight observed from object surfaces affect the appearance and analysis of natural color images. Invariant representations to these factors were presented in several ways. Most of these methods used the standard dichromatic reflection model that assumed inhomogeneous dielectric material. The standard model cannot describe metallic objects. This chapter introduces an illumination-invariant representation that is derived from the standard dichromatic reflection model for inhomogeneous dielectric and the extended dichromatic reflection model for homogeneous metal. The illumination color is estimated from two inhomogeneous surfaces to recover the surface reflectance of object without using a reference white standard. The overall performance of the invariant representation is examined in experiments using real-world objects including metals and dielectrics in detail. The feasibility of the representation for effective edge detection is introduced and compared with the state-of-the-art illumination-invariant methods.

In this paper, we investigate the eect of using an optimum
number of clusters with Fuzzy C-Means clustering, for Liver CT image
segmentation. The optimum number of clusters to be used was measured
using the average silhouette value. The evaluation was carried out using
the Jaccard index, in which we concluded that using the optimum number
of clusters may not necessarily lead to the best segmentation results.