Wasit Journal for Pure Sciences

In this research, the finite-time stability of Multi-Composition Caputo-Katugampola fractional Integro- Differential Nonlinear system with many values of fractional derivatives is studied with some sufficient and necessary conditions as Lipchitz conditions for nonlinear functions involving Riemann– Katugampola fractional integral and the formulas derived from them with their respective bounded value as well as finding the solution under these conditions that contains Mittag Leffler functions which appeared through the use of the generalized Laplace formula which suitable with Caputo-Katugampola fractional derivative. Therefore, important conditions appeared that contain the parameters that played a good role in finding and computing the stability as in the attached tables of illustrative examples that explain the necessary time requirements for it.

In this paper, fractional diffusion equations of Caputo-Hadamard presented as interesting classes of fractional differential equations have not studied before. For the first time, the solution to a diffusion equation of this type was calculated. Therefore, it is considered a new addition to the diffusion equation with the presence of this form of fractional derivatives, with its initial conditions. Additionally, this paper presents a suitable Laplace logarithmic formula for the given fractional equation, derived from the generalized Laplace formula and its established conditions. This also relies on the formula for the fractional derivative as a transformation. The first time this type of fractional derivative was established, it helped in finding an approximate formula for the solution. The residual power series method used is effective in solving many fractional equations and was even more effective when used with the Laplace logarithmic formula. The Log(t+1) –Laplace residual power series method combines the two concepts, formulating and evaluating the method, shown that from any changing of the values of the parameters involving in formulation of derivative as well as the values of fractional order which found in tables and figures for illustrative examples which illustrates the presented method.

This study examines the distributional properties of exchange rate volatility for major oil-exporting developing economies from 1990-2022. Three probability distributions - Normal, Fréchet, and Log Normal - are compared utilizing the Moments Method and Maximum Likelihood estimate approaches. The Bayesian Information Criterion (BIC) and the Akaike Information Criterion (AIC) are used to assess goodness-of-fit. The results indicate that the Normal distribution tends to provide the greatest fit for many countries, including Iraq, Kuwait, Libya, Iran, Algeria, Oman, and Bahrain. The Log Normal distribution is found to be optimal for Saudi Arabia, Qatar, and Egypt, while the Fréchet distribution fits best only for the United Arab Emirates. The findings highlight significant variation in exchange rate volatility patterns across these oil-dependent economies. This analysis provides insights into the underlying distributional characteristics of exchange rate returns in these countries, which can inform appropriate model selection for volatility forecasting and risk management applications. The research adds to existing literature by conducting a entire distributional comparison across a broader set of oil-exporting developing economies over an extended time.

With the proliferation of Internet of Things (IoT) devices, ensuring robust security has become a critical challenge due to limited computational resources, heterogeneous traffic, and evolving cyber threats. Intrusion Detection Systems (IDS) must therefore achieve not only high accuracy but also efficiency and transparency to be deployable in real-world IoT environments. This study introduces a Marine Predator Algorithm (MPA)-optimized Bidirectional Long Short-Term Memory (BiLSTM) framework for real-time IoT intrusion detection. The MPA simultaneously performs feature selection and hyperparameter tuning, reducing dimensionality and optimizing network configuration. Evaluations on the CICIoT2023 and TON_IoT datasets show that the proposed framework achieves 99.52% detection accuracy, with an inference latency of 37 ms on Jetson Nano and a compact TensorFlow Lite model size under 210 MB. Compared to baseline BiLSTM and other deep learning models, the proposed approach reduces memory usage by 38% and maintains real-time responsiveness on Raspberry Pi 4 and Jetson Nano. Furthermore, SHAP-based interpretability is integrated to identify and explain the most influential features, enhancing trust and usability of IDS outputs. These results demonstrate that the proposed MPA-BiLSTM achieves an effective balance of accuracy, efficiency, and interpretability, making it a strong candidate for deployment in resource-constrained edge-based IoT security systems.

Emotional face recognition is one of the most important tasks in the medical sphere and security. The paper is concerned with the problem of effective emotion detection by means of a Convolutional Neural Network (CNN) algorithm that was selected due to its capacity to handle spatial hierarchies of images data successfully. The study will attempt to create a user-friendly and real-time system that has the capacity to verify emotional scores based on facial expressions. The proposed methodology includes three major steps that include data preprocessing, classification and evaluation. The preprocessing of the image data is done by software and after that a CNN model is trained on the processed image data. The system was tested on several datasets, with a lightweight CNN giving an accuracy of 88.67 percent in the training and testing cases. Moreover, the model was compared as far as current emotion recognition techniques are concerned, showing better results in real-time tasks. The paper emphasizes the possibility of the CNN model being applicable to the real-world context especially in medical and safety sectors. Nevertheless, the study also defines the issues of data imbalance and the lack of multimodal methods that might be considered in the future research to increase the system strength and flexibility.

The synthesis of proteins, DNA, and RNA as well as the metabolism of methionine and folate, depend on the enzyme MTHFR, a crucial regulator of the one-carbon cycle. This enzyme is necessary for the remethylation of homocysteine to methionine and transforms 5,10-methylenetetrahydrofolate into its active form, 5-methyltetrahydrofolate. Determining the degree of polymorphism in the gene for the enzyme methylenetetrahydrofolate reductase (MTHFR) in patients with beta-thalassemia major and comparing it to the control group, as well as sequencing, analyzing, and detecting mutations in the DNA nucleotides of the gene for the enzyme MTHFR, were the objectives of the study. 90 samples were taken from patients at the Al-Ramadi Teaching Hospital for Women and Children, both male and female. The study included 60 samples of individuals with β-thalassaemia major. The patients were between the ages of 3 and 52. The control sample consisted of 30 samples of each sex, ages ranging from 16 to 65. For every sample in the investigation, the MTHFR C677T gene band was amplified using the polymerase chain reaction (PCR). When compared to the 100 bp marker, all samples had a band size of 198 bp. The CHROMAS was used to examine the sequence data. The BLAST SEARCH tool was used to record the data, which had the same gene, C677T, between 9800 and 9930, and showed no alterations from the reference gene (NG_013351). These findings show that the participants in this research were wild type for this location, meaning they did not carry the C677T mutation

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease characterized by persistent synovial inflammation that primarily targets the joints and can involve extra-articular organs, including the cardiovascular system. Although its precise etiology remains unclear, RA is believed to result from a complex interplay of genetic susceptibility (e.g., HLA-DR alleles), environmental factors, and microbial triggers. RA affects all age groups and both sexes, but it is approximately three times more prevalent in women than in men. Initial symptoms typically present as symmetric polyarthritis, affecting the small joints of the hands and feet, with progression to larger joints such as the knees and elbows. This study aimed to evaluate cardiovascular risk markers among RA patients in Wasit Governorate, Iraq. A total of 90 individuals aged 30–60 years were enrolled, comprising 60 RA patients (diagnosed based on clinical and laboratory criteria) and 30 age- and sex-matched healthy controls. Among RA patients, 73.3% were female, and the majority were either obese (70.0%) or overweight (21.7%). RA patients exhibited significantly elevated ESR values, along with variations in hematological indices such as reduced RBC count and hemoglobin levels. Moreover, serum lipid analysis revealed significantly elevated levels of TC, TG, LDL, and VLDL, accompanied by decreased HDL levels.

Low river water levels in summer and low rainfall in winter have led to desertification, leading to widespread use of groundwater in many areas of Iraq, especially in the southern and central regions. Groundwater is considered an alternative solution to address desertification, but in reality, many health problems have emerged due to the ions present in groundwater, which cause numerous diseases. Among these ions studied in this research is the nitrate ion (NO3⁻). This spectroscopic method was proposed to determine the concentration of nitrate ions in groundwater. Eriochrome Black T (EBT) colorimetric detector was used as a spectroscopic detector after complexing the detector with silver nitrate to determine the nitrate ion in groundwater. Several parameters and optimal conditions for this reaction were studied, and the assessment of nitrate levels was found to be consistent with the linear equation of the Beer-Lambert law in the range of 0.5–5.0 mg/L at a wavelength of (530) nm. When plotting the calibration curve, good linearity was recorded (R2 = 0.9979). The limits of detection (LOD) and quantification (LOQ) of the method were determined to be 0.269 and 0.816 mg/L, respectively, based on the standard deviation values. The precision measured in terms of relative standard deviation (RSD) over the specified time period at five concentration levels was less than 6.7%. The recovery rate was satisfactory, ranging from 96 to 103, Sandell's sensitivity equal to (0.0263) μg.cm-2. The rationale for using spectrophotometric methods lies in their simplicity of operation, low cost, and applicability in laboratories with limited resources, as opposed to more advanced techniques such as HPLC or LC-MS. Based on these results, the method was approved for the determination of nitrate ions.