Journal of Education for Pure Science-University of Thi-Qar

In this paper we introduce presents a numerical approach, based on radial basis function networks
(RBFNS), for the approximation of a function and its derivatives (scattered data interpolation), The
proposed approach here is called the indirect radial basis function network (IRBFN) to solve second
and fourth order differential equations the procedure can start with the second derivative. First, the
second order derivative is approximated by a RBFN, then the first order derivative is obtained by
integration. Finally the original function is similarly obtained, i.e. by integrating the first derivative
function. This second method is here referred to as the second indirect method or IRBFN2 likewise
fourth -grade derivative IRBFN4

Breast cancer is the most frequently diagnosed cancer and is a common cause of cancer related death in
women, accounting for 25% of cancer cases and 15% of cancer-related deaths worldwide The present study
was designed to determine and compare the levels of Adiponectin hormone, Vitamin D3, Calcium, Alkaline
phosphatase, and C- reactive protein among women with Invasive ductal carcinoma &healthy individuals.
The study was designed to determine and compare the levels of serum Adiponectin hormone, Vitamin D3,
Calcium, Alkaline phosphatase, and C- reactive protein levels were measured in eighty -five patient’s
Invasive ductal carcinoma and seventy-five supposed healthy subjects. The levels of serum Adiponectin
hormone, Vitamin D3, were reveled significant decrease among patients with Invasive ductal carcinoma as
compared to control group whereas the levels of Calcium, Alkaline phosphatase, and C- reactive protein
showed a significant increase in Invasive ductal carcinoma patients in comparison to control subjects (P ≤
0.05). This present study indicated that low blood adiponectin level was associated with the risk of obesity-
related malignancies, including breast cancer, we can conclude that deficient of vit. D3 may contribute to
the process of carcinogenesis among the breast cancer patients, and our findings revealed significant in
Calcium, Alkaline phosphatase, and C- reactive protein levels among patients with Invasive ductal
carcinoma.

In this work, a batch adsorption study was conducted to investigate the removal efficiency of lead (II),
Copper (II) and Cadmium (II) from aqueous solutions by (poly (thiourea-Formaldehyde-Epoxy resin)
nanomagnetic Full-IPN's (NM Full-IPN's). The NM Full-IPN's was synthesized by sequential
polymerization with presence Fe3O4 nanoparticles. The chemical structure and surface morphology of NM
Full-IPNS resin nanoparticles were characterized by Fourier transform infrared spectroscopy (FTIR),
Scanning electron microscopy (SEM) and Transmission electron microscopy (TEM). The chemical
structure and surface morphology of NM Full-IPNS resin nanoparticles were characterized by Fourier
transform infrared spectroscopy (FTIR), Scanning electron microscopy (SEM) and Transmission electron
microscopy (TEM). The thermal properties of (NM Full- IPN's) has been evaluated by Thermogravimetric
analysis (TGA) and Differential Scanning Calorimetric (DSC). Adsorption of Pb (II), Cu(lI) and Cd(II)
onto NM Full- IPN's agreed well with the Langmuir model, as revealed by the higher values of correlation
coefficients. The results indicate that NM Full- IPN's could be used as efficient adsorbent for the removal
of Pb (II), Cu(lI) and Cd(II) from aqueous solution

The present study aimed at determining the effect polyphenolic extract for tubers cyperus rotundus on
myocardial infarction in female rats. In this study, 36 female rats were used and the animals were divided
into 6 groups of each group containing (6 rats). Myocardial infarction was induced by injection of
isoproterenol at (85 mg/kg) in the intraperitoneally of female rats at an interval of 24 h for two days. The
results showed that the injection of isoproterenol in the intraperitoneally of female rats has caused a
significant increase in levels of Troponin-I, lactate dehydrogenase(LDH), creatine phosphokinase (CPK),
alanine transferase (ALT), aspartic transferase (AST) and necrosis in the second group compared to the
first group. It was found in the current study that treatment of female rats with myocardial infarction with
phenolic extract of tubers of cyperus rotundus has caused a significant decrease in levels of Troponin-I,
lactate dehydrogenase (LDH), creatine phosphokinase (CPK), alanine transferase (ALT), aspartic
transferase (AST). Necrosis decreased in the group treated with phenolic extract of tubers of cyperus
rotundus. From this we conclude that the phenolic extract of cyperus rotundus of tubers has protective role
against myocardial infarction.

Emotion is a mental and physiological state associated with a wide variety of feelings, thoughts and
behaviors. Emotions are fundamental in the daily life of human beings as they play an important role in
human cognition, namely in rational decision-making, perception, human interaction, and human
intelligence.
This paper investigates the effect for the emotion-discriminating precision of Different wave levels of
EEG signals and a particular number of channels.
It using various sets of EEG channels, the proposal classified affective states in the equivalence and
excitability dimensions. To begin, DEAP normalized the pretreated hypothetical data. Following that,
discrete wavelet transduction was used to divide the EEG into four bands, The scales used were the features
of the K-nearest neighbor Algorithm entropy and energy algorithm.
The highest classification accuracy was using the K-NN algorithm for channels (4-10-14-18, 32) in the four
dimensions (valence, arousal, dominance, and liking). They are channel 18 (99.7656%, 99.7656%,
99.7656%, 99.7656%) respectively. While the highest classification accuracy for the frequency bands is
the gamma frequency greater from beta and alpha and theta frequency for the four dimensions is
(99.7656%).

Journal of Education for Pure Science- University of Thi-Qar
Vol.11, No.2 (Nove, 2021)
Website: jceps.utq.edu.iq Email: jceps@eps.utq.edu.iq
68
DOI: http://doi.org/10.32792/utq.jceps.11.01.07
Deep Learning-Based Face Detection and Recognition System
Jamal M. Alrikabi1 Kadhim H. Alibraheemi2
1’2Computer Science Department, College of Education for Pure Science, Thi-Qar University, Thi-Qar,
Iraq.1
Received 09/05/2021 Accepted 14/09/2021 Published 11/11/2021
This work is licensed under a Creative Commons Attribution 4.0 International License.
Abstract:
Face detection and recognition systems have recently achieved encouraging results using deep
learning especially Convolutional Neural Network (CNN). Face detection and recognition system have
many challenges in unconstrained environments that decrease the accuracy, for overcoming these
challenges a deep learning-based features combination has been proposed for face recognition. The scheme
performs feature-level combination by applying two pre-trained InceptionNet-v1 and VggNet-16 models
as deep feature extractors. First, faces are detected and aligned using Multi-Task Convolutional Neural
Networks (MTCNN) face detector then the deep features are extracted from a face image using each
individually pre-trained CNN. Second, features obtained from InceptionNet-v1 and VggNet-16 models are
combined using the serial-feature combinations method. Finally, a classification task is perform using a
multiclass Support Vector Machine (SVM) classifier. Experiments on the following datasets: VggFace2,
LFW, Essex, and ORL, indicate the efficacy of the proposed system as the combination of the two pre-
trained CNN models improves performance. The combination strategy, in particular, yields an accuracy of
95.33% to 99.29% on all datasets.

Wireless Sensor Network (WSN) is a network made out of huge number of minimal effort, low power
and multifunctional sensor hubs that are sent over an unattended zone either near or inside the objectives
to be noticed. These sensor hubs are little in size, however are outfitted with sensors, implanted chip and
radio handsets and in this manner have detecting ability, yet additionally information handling and
imparting capacities. Every single sensor hubs in the organization intermittently sense the states of the
objective, measure the information lastly send the detected information back to a Base Station (BS) or sink
either in single bounce or in multichip correspondence. On the off chance that immediate correspondence
is utilized, hubs which are distant from the sink need more transmission ability to communicate their
detected information to sink hub and consequently they drain their energy quicker when contrasted with
hubs closer to the sink. In multi bounce correspondence, energy opening shows up close to the sink hub in
light of the fact that the hubs closer to the sink hub will convey hefty traffic when contrasted with different
hubs. Consequently, no more information can be conveyed to the sink after an energy opening shows up.
Thus, a lot of energy is squandered and the organization lifetime closes rashly. To defeat the energy opening
issue, homogeneous group based WSN engineering are utilized. The essential thought is to gather hubs
around a Cluster Head (CH) that is dependable inter cluster availability. This paper gives an outline of
different grouping procedures utilized in WSN in order to keep up network adaptability, load adjusting,
and inertness decrease.

In recent years, the number of people having diabetes mellitus is steadily increasing in countries known by
low and middle income. Millions of them have been diagnosed with high blood sugar. Diabetes mellitus
is related to irregular carbohydrates metabolism with difficulty managing blood glucose which by the time
leads to serious damage to nervous system or even macrovascular, or blood vessels. Therefore, the demand
for advanced devices for glucose monitoring is highly growing. A bio-sensor is a signal detecting device
generated from reactions either biological or chemical. They can be used for many purposes, for example,
detection of biological molecules using in vitro and in vivo samples. Electrochemical sensors represent
one of the technologies could be used for this purpose. It shows high sensitivity and mechanical strength
while utilizing in physiological conditions which is a promising advantage for glucose determination. The
aim of current study is to review electrodes generations for glucose detection, as well as, commonly
prepared and investigated electrochemical electrodes for glucose determination. In addition, types of
glucose electrodes have been mentioned here based carbon nanotubes and/or graphene as excellently
conductive, stable, reproducible, and sensitive materials. The aim of current study is listing and discussing
the progression of glucose generations as well as the development of glucose sensors in recent years.
Enzymatic and non-enzymatic nanocarbon sensors were mainly studies and classified as glucose sensors
also explained with further details regarding limit of detection.

Accurate estimation of software development effort has become a crucial for effective projects
planning. It is a very challenging task for project teams to predict the development effort required in the
initial phases of a software project. Software estimation prior to development process can decrease the risk
and enhance the project's success rate. Although numerous traditional and machine learning models have
been proposed for software effort estimation over the past decade, the level of accuracy is not satisfactory
enough. The objective of this study is to assess the capability of using two different soft computing methods
namely artificial neural networks (ANNs) and adaptive neuro-fuzzy inference system (ANFIS) to estimate
the software effort. COCOMO dataset is used to test and train the proposed models. Mean magnitude of
relative-error (MMRE) and correlation coefficient (R) were used as assessment criteria. It was concluded
that both models can satisfactorily estimate software development efforts, but ANFIS model has
outperformed the ANN model in two statistical indicators: MMRE and correlation R. It is recommended
that ANFIS can be used as a predictive model for software effort estimation.

Image filters has taken attention last few years due to its importance in terms of image processing
and applications. Applying image filters on images elements can be affected by the values of image
parameters, which resulted from any processing tasks. By applying image filters, we can extent the image
processing methods to present higher productivity. In this paper, we compare the effect of applying five
image filters on the statistical distribution, which are (Laplacian, Differentiation, LOG, Sharpening, and
Gaussian). Our method has been applied for a number of textural images (water texture, wool texture, and
wood texture), the images has been divided for three groups according to the texture type. The result of our
method proved that some of image filter affects the statistical distribution of image elements which are:
(Differentiation, LOG, Sharpening) while other do not affect the parameter distribution (Laplacian,
Gaussian). We evaluate our method by calculating the value of (MSE). The method opens the door in front
of extending such technique with other image processing aspects.
Keywords: Images processing, Image filtering, Statistical distribution, Textural Images, Probability
Density Function.

Image filters has taken attention last few years due to its importance in terms of image processing
and applications. Applying image filters on images elements can be affected by the values of image
parameters, which resulted from any processing tasks. By applying image filters, we can extent the image
processing methods to present higher productivity. In this paper, we compare the effect of applying five
image filters on the statistical distribution, which are (Laplacian, Differentiation, LOG, Sharpening, and
Gaussian). Our method has been applied for a number of textural images (water texture, wool texture, and
wood texture), the images has been divided for three groups according to the texture type. The result of our
method proved that some of image filter affects the statistical distribution of image elements which are:
(Differentiation, LOG, Sharpening) while other do not affect the parameter distribution (Laplacian,
Gaussian). We evaluate our method by calculating the value of (MSE). The method opens the door in front
of extending such technique with other image processing aspects.
Keywords: Images processing, Image filtering, Statistical distribution, Textural Images, Probability
Density Function.

In this paper, the homotopy analysis method(HAM) has been applied for solving time-delay
Burgers equation. HAM contains the auxiliary parameter ћ, which provides us with a simple way to control
the convergence region of solution series. In this study, we compare obtained results through HAM with
the exact solutions to show the accuracy of HAM