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

Attention deficit hyperactivity disorder (ADHD) is a behavioral problem that can last into adulthood
and affect children. Because it can show complicated brain activity, electroencephalography (EEG) plays
a key role in determining the neurophysiology of ADHD. several statistical features are extracted from
five frequency bands by using a discrete wavelet transform. The proposed system is evaluated by using
K-means-based feature selection and 5 machine learning methods (Least-square support vector machine,
k-nearest neighbor, Decision tree, and naive-Bayes classifier, support vector machine), so this system
developed using a ten-fold cross-validation strategy and showed the testing accuracy for each classifier as
( 96.49%, 92.66%, 88.08%, 68.39%,53.71% ), respectively

Classification of non-diatomic algae in the agricultural soils of Shatt Al-Arab district were studied for
an entire year starting from November 2020 until September2021.Diifferent environmental
parameters (Temperature of soil and air, salinity, pH and some nutrients) were studied. The current
study identified 72 taxonomic units belonging to two divisions, the blue-green algae and the green
algae, the appearance of algal species was highest in the spring months and the lowest in July. The
study identified five new species of blue-green algae represented by Gloeocapsopsis chroococcaides,
Lyngbya regalis, Oscillatoria salina , Phormidium breve, Phormidium tinctorium. The results also
showed the dominance of blue–green algae at the expense of green algae, and the genus Oscillatoria
recorded dominace over the rest of the genera, and the results of the statistical analysis showed that
temperature, salinity and conductivity were inversely associated with the number of algae species,
while the nutrients represented by nitrogen, phosphorus, and available potassium were directly
associated with the number of algal species.

The aim of this study was to evaluate the efficacy of compounds and the antifungal vitality of
Sargassumsp. against pathological fungi which are Candidaalbicans , Candida krusei , Aspergillusniger .
as the effective compounds were extracted from this algae solvents were used (hot water and ethanol
alcohol) with the qualitative detection of secondary metabolites from the aqueous extract, which
indicated the presence of chemical compounds such as flavonoids, phenols, glycosides, terpenes,
carotenoids and saponins. The chemical compounds were determined using gas chromatography - mass
spectrometry (Gc-mass) of the ethanolic extract of seaweed. These compounds are represented , n-
Hexadecanoic acid , Oleic acid , Fucostero , Cholesta-5,20,24-trien-3-ol, (3.beta.)- , Tetradecanoic acid ,
Palmitoleic acid and Pentane,2-(1-methylethyl)thio] . The aqueous extract of the seaweed did not show
any biological activity against all the fungal species under study. While it was noted that the ethanolic
extract was more effective in its effect on pathogenic fungal species. As the rates of inhibition zones of
the extract (22 , 21.2 , 20.76) against C.albicans, A.niger, and C.krusei, respectively.

The objective of this study was to determine bioactive compounds and antifungi of extracts of
Laminaria sp. against pathological fungi, which are Aspergillusniger, Candida albicans , and Candida
krusei, as the effective compounds extracted from this algae solvents were used (hot water and ethanol
alcohol) and the products of secondary metabolism from the aqueous extract were qualitatively detected,
which indicates existence of chemical compounds such as flavonoids and phenols Glycosides, terpenes
and carotenoids, Identified chemical compounds in the ethanolic extract of seaweed using gas
chromatography-mass spectrometry (Gc-mass). These compounds were Oleic acid, n-Hexadecanoic acid
, Fucosterol , Cyclohexane, 1,5-diethenyl-3-methyl-2-methylene-, (1.alpha.,3.alpha.,5.alpha.)- ,
Tetradecanoic acid and D-Mannitol. The aqueous extract of Laminaria sp. did not appear any biological
activity versus all the studied fungal species. While the ethanolic extract was found to be more effective
in its effect on pathogenic fungal species. The average zones of inhibition of the extract were (21.53,
21.46, 20.00) against C.albicans , A.niger and C.krusei , respectively.

Driver drowsiness is one of the top causes of road fatalities and transportation industry risks.
Electroencephalography (EEG) has been regarded as one of the most reliable physiological signs for
detecting drowsiness in drivers since it directly assesses neurophysiological brain activity. If possible, a
system that monitors driver drowsiness and warnings can prevent the greatest number of traffic incidents.
This study presents a simple and inexpensive method for detecting driver drowsiness or sleep onset with
single-channel EEG signal processing. The primary contribution of this study is the identification of
sleepiness detection from a publicly available graph signal dataset using only two discriminated features
(standard deviation and entropy) and a filter that is easily implementable in any microcontroller device or
smartphone. This study employed the least square support vector machine to categorize driver status into
two groups (awake or drowsy). The proposed method can be implemented in real-time with high
efficiency and precision. Furthermore, this method can be readily applied on a smartphone to create and
expand a sleepiness detection and alert system for vehicle drivers. The trial findings indicate an accuracy
of 95.5 percent.

The neurological condition known as epilepsy affects the brain in a way that causes recurrent problems.
As a result, seizure identification is crucial to the clinical treatment of epileptic patients. The
electroencephalogram (EEG) has become a crucial tool most of the time, a small number of very
competent professionals manually identify the epileptic EEG signal. Detecting and evaluating epileptic
seizure activity in humans. In this paper, we attempt to automate the detecting procedure. We extract
features using the wavelet transform and derive statistical parameters from the wavelet coefficients that
have been decomposed. Artificial neural network (ANN) is used for the classification. By utilizing the
University of Bonn's benchmark database.

Finding motifs in DNA sequences is a current challenge and an essential step in bioinformatics.
Processing these issues needs considerable data analysis due to technical advancements in the
industry. Artificial Neural Networks (ANNs) are increasingly used, particularly for motif
identification and genomic analysis. In order to find motifs in DNA sequences, this work proposed a
supervised learning algorithm for feed-forward neural networks called Scaled Conjugate Gradient
(SCG) algorithm. The SCG algorithm utilizes a step-size scaling mechanism that is fully automated to
minimize time-consuming row searches during each training iteration. This algorithm was used in this
work for motif discovery to train code patterns and to reduce a multivariate global error function
dependent on the network weights. It trains many code patterns of lengths between 4 to 509 bases to
find them in a database with 2,227,382 bases; many experiments were done with different numbers of
hidden layers; our finding ten hidden layers provide the best results, with training percentage is 100%.
Compared to the other supervised learning neural network algorithms, One Step Secant, Gradient
Descent, Bayesian Regularization, and BFGS Quasi-Newton; our find SCG algorithm produced
higher accuracy (100%) and less time during the training and testing phases.

The diagonally implicit Runge-Kutta method (DIRKT), fourth-and fifth-order, is developed in this study
to solve general fourth-order ordinary differential equations (ODEs). DIRKTO5 and DIRKTO4,
respectively, are the names of these methods. Both methods consist of three stages. A collection of test
problems were utilized to support the methodologies. The numerical outcomes demonstrated that the
suggested approaches performed better than the implicit Runge-Kutta (RK) methods in terms of precision
and the quantity of the evaluations of function.

polycystic ovarian syndrome, which affects between 5 and 10 percent of adolescent women, is one of
the most prevalent endocrine system conditions (PCOS). Infertility and failure to ovulate are symptoms,
as are cardiovascular conditions, type 2 diabetes, etc. PCOS can be found by biochemical, clinical, and
ultrasonographic techniques. It is well recognized that early detection and intervention can lower the risk
of developing PCOS. Therefore, it is essential to understand which classification model and features
contribute significantly to the prediction of disease, which is the goal of this study. Despite the
employment of several tools, Naïve Bayes exhibits accuracy performances of 89.51% with 6 chosen
features.

PCOS (Polycystic Ovary Syndrome) is an endocrine disorder that affects women of reproductive age.
Once diagnosed, the condition cannot be cured, but treatment can help relieve its symptoms. Although
the exact cause of PCOS is unknown, certain risk factors do exist. Obesity, insulin resistance, blood
pressure, depression, and inflammation are all factors that contribute to this syndrome. While the
symptoms include hirsutism, oligo-ovulation, acne, heavy bleeding, and skin darkening. Using the causes
and symptoms, a model is created that accepts them as features and outputs the presence or absence of
this condition. Machine learning models used for classification include Linear SVM, Gaussian SVM,
Decision Tree, Nave Bayes, KNN1, KNN3, and KNN5. The goal of building multiple models is to find
the best one for a given dataset within the known scope of knowledge.

Electroencephalography signals derived from electrical activity in the brain are commonly used to
diagnose neurological diseases. These signals reveal electrical activity in the brain and provide
information about the brain. One of the most severe brain conditions, epilepsy is brought on by a group of
neurons in the brain that exhibits abnormally pathological oscillatory activity. Automated methods that
evaluate and identify epileptic episodes using electroencephalography data are currently being created.
This study's goal is to evaluate how well the ensemble approach-based model can foretell whether or not
an epileptic seizure would occur. The proposed mode was evaluated using the benchmark clinical dataset
provided by Bonn University. In this research, a reliable method based on temporal frequency image
(TFI) and fractal dimension(FD) is proposed for epilepsy detection -based EEG signals. The ensemble
method which consists of classifier KNN, SVM, and Tree is used to distinguish patients with epilepsy.
The outcomes demonstrated that the suggested strategy produces a high-accuracy clustering and aids
neurologists in making an epileptic condition diagnosis and subsequently recommending the proper
treatment. The suggested strategy is a promising one. approach for analyzing EEG signals and providing
reliable and accurate clustering to patients who suffer from epilepsy.