Samarra Journal of Pure and Applied Science

The aqueous environment contamination by pseudomonas has become a great challenge for human beings. Walnut shells were washed by distilled water, mixed with KOH, burned at 650°C, and treated ultrasonically to obtain carbon nanoparticles. The bacterial isolates were from different sites of domestic water in hospitals and characteristic the pseudomonas. After that the minimum inhibitory concentration (MIC) was done to determine the optimum concentration that affected on p.aeruginosa and it was 0.006 mg/l , then using of polymerase chain reaction (PCR) to detect virulence factor genes (copB and czcA) and they characterized by biochemical and morphological diagnosis of p.aeruginosa in five isolates. Following treatment, the copB gene expression was 0.211 for isolate 2 and 0.047 for isolate 3, while the czcA gene expression was 0.882 for isolate 10 and 0.392 for isolate 11. The capacity of CNPs to reduce the gene expression in virulence factor genes was tested by using quantitative PCR (qPCR), as a result that affects bacteria, the CNPs at certain concentrations has high capacity to identify the virulence factors genes. These results were supported by XRD examination (Panalytical ‘X’ Pert Pr, UK), which indicated the presence of two Bragg diffraction peaks in the carbon nanoparticles, TEM (Zeiss LEO 912 AB/ Germany) results also showed the presence of inhomogeneous particles, as well as the irregular shape of the surface of the carbon nanoparticles with a large surface area, according to SEM examination (Zeiss Sigma VP/ Germany).

Candida is a natural microscopic organism are yeasts that are found in the normal microorganisms of the oral cavity, opportunistic that cause diseases when conditions are created, overgrowth of yeasts leads to tissue invasion and causes diseases. One of the diseases caused by yeast overgrowth in children is oral thrush, this study was conducted at the Women's and Children's Hospital in Ramadi, where 100 oral swabs were collected from infants with oral thrush aged 1 month-2 years and then morphological and biochemical tests were conducted on positive growth isolates. The results showed that 77 isolates were positive growth by 77% and 23 isolates by 23% showed no growth. Four types of Candida were identified, Candida albicans with 49.35%, followed by Candida tropicalis with 28.57%, Candida krusei with 14.28% and Candida glabrata with 7.79%. The results also showed a statistically significant difference between the infection and the age group, where the most infected age group between 1 month-1 year by 74.36%, and the results also showed that the cases of oral fungal infection in infants are associated with artificial feeding by 90.91%.

This study was conducted during the period from 2022 to 2023 on soils contaminated with organic pollutants, where samples of contaminated soil were taken from different intersections of the Sharqat District, Salah al-Din Governorate, as well as the Rumana area, Qayyarah District, with the aim of isolating filamentous bacteria. Streptomyces from the sources of contaminated soils and their diagnosis for the purpose of using them later in the bioremediation of organic pollutants. Seven isolates were obtained from Streptomyces. From soils contaminated with organic pollutants and 8 isolates from contaminated garden soils, where all isolates were positive for Gram stain and produced amylase and catalase enzymes, and also possessed distinctive morphological characteristics on the surface of the culture media, as their colonies appeared in different shapes, the most common of which are chalky and circular, and there is also variation in the shapes and colors of the hyphae, and this bacteria is characterized by its ability to produce distinctive odors due to its ability to produce volatile compounds, and this bacteria is also characterized by its ability to grow slowly on the culture media. The treatment process using this type of bacteria showed high effectiveness in getting rid of organic pollutants. The chemical and physical properties of the polluted soil were examined before the treatment process. The percentage of organic matter was 6.899% before the treatment process. The percentage of electrical conductivity was 33407 micro siemens/cm, and the pH was 7.9. The percentages changed after the treatment process and under two conditions, one of which was controlled in the incubator and at a suitable temperature. The percentage of organic matter reached 3.381%. There was also a decrease in the value of electrical conductivity, which reached 4208 micro siemens/cm, and an increase in the pH value of 10.04. This indicated that the bacteria Streptomyces The organic matter was exploited, and under natural conditions there was more decomposition of pollutants as the percentage of organic matter reached 3.10%, and there was also a decrease in electrical conductivity of 4439 microsiemens/cm, and there was also an increase in the pH value, evidence that bacteria, Streptomyces, It has the ability to grow in harsh conditions and works to decompose and exploit organic pollutants and rid the environment of these pollutants.

Reactive hyperemia (RH) is a temporary increase in blood flow after ischemia, associated with reduced arterial stiffness and used as a marker of vascular health. Arterial stiffness, a key cardiovascular risk factor, is commonly assessed using pulse transit time (PTT), a non-invasive method. This study examines the effect of occlusion site (arm vs. forearm) on RH magnitude and its relationship with vascular stiffness.Ten healthy males (23.1 ± 5.0 years) participated. PTT and digital pulse amplitude (DPWA) were measured before and after RH induction at baseline and one-minute intervals for five minutes post-induction. Arm occlusion led to a post-occlusion reactive hyperemia index (RHI) increase of 23% at 2 minutes (P < 0.03). Forearm occlusion resulted in RHI increases of 36%, 27%, 27%, and 18% at 2, 3, 4, and 5 minutes, respectively, compared to 1 minute. No significant differences in RHI were found between arm and forearm occlusion, with RH sustained up to 5 minutes after forearm occlusion.PTT increased significantly by 6% and 4% at 1 and 2 minutes after arm occlusion release. Following forearm occlusion, PTT was significantly higher (5%) than baseline only at 1-minute post-release. The peak RHI occurred in the second minute after occlusion release, with forearm RH lasting longer (up to 5 minutes), though magnitudes were similar. Peak PTT, indicating maximum vascular stiffness reduction, occurred in the first minute post-release and may extend into the second minute.

The current study aimed to focus on the biosynthesis method of
Nickel Oxide nanoparticles (NiONPs) using the cell- free culture
supernatant (CFCS) from Staphylococcus aureus as a stabilizing and
reducing agent, the nanoparticles of Nickel Oxide were synthesized
using Nickel nitrate Ni(NO3)2, in addition to the therapeutic
applications of Nickel Oxide nanoparticles that will be used against A.
baumannii MDR bacteria, the Nickel Oxide nanoparticles were
characterized using UV-VIS spectroscopy, atomic force microscopy
(AFM), X-ray diffraction (XRD), field emission scanning electron
microscopy (FE-SEM), energy-dispersive X-ray spectroscopy (EDX)
and Fourier-transform infrared spectroscopy (FTIR). AFM analysis
revealed that NiONPs have an average diameter of 63.81 nanometers
and the FE-SEM image showed a spherical shape, Additionally, the
most resistant bacterial isolate was identified, selected and tested for
antibacterial activity against MDR-A. baumannii using different
concentrations of Nickel Oxide nanoparticles (500μg/ml, 1mg/ml,
2mg/ml, and 4mg/ml), the results of the biogenic Nickel Oxide
nanoparticles showed a promising activity against MDR-A.
baumannii, with an inhibition zone diameter of 22 mm at a
concentration of 4 mg/ml, while the inhibition zone diameter was 18
mm at 2 mg/ml, which is the minimum inhibitory concentration
(MIC), the biogenic synthesis of NIONPs showed promising
antibacterial activity against biofilm production.The results of the
anti-biofilm activity using Nickel nanoparticles (NiONPs) against
multidrug-resistant bacteria (MDR-A. baumannii) at different
concentrations (500 μg/ml, 1 mg/ml, 2 mg/ml, 4 mg/ml and 4
mg/ml), respectively, against biofilm production showed that the
best biofilm inhibitory concentration is 8 mg/ml of Nickel Oxide
nanoparticles, which may have applications as a

This study was conducted in Baghdad, Iraq, over a period of time extending from the 1 / November / 2024 to 25 /February /2025 among individuals within the community in Baghdad, Iraq. This study takes Ninety blood sample were collected and divided into three groups (30 sample from the control group, 30 sample from the skinny group and 30 sample from the obesity group) from the community with attached personal information. All laboratory tests were analyzed in the laboratories of the College of Medical Technology/ Al-Farahidi University/Baghdad. The thyroid hormone analyses were performed using Fine-care technique. The results appear a significant increase in BMI (P > 0.05) and Thyroid-stimulating hormone (P > 0.05), while there were no significant differences with triiodothyronine (T3) hormone (P <0.05) between the study groups. Age showed a significant increase with (BMI) with (P >0.05) and TSH hormone (P >0.05), while age hasn\'t statistically significant with T3 hormone (P <0.05).

This study, conducted at Bakrajo Technical Institute (2021–2022), examines factors affecting bioremediation of lead (Pb²⁺) using two sunflower species, Lemon Queen and Mammoth Russian. A randomized complete block design (CRD) and analysis of variance (ANOVA) were used to evaluate the effects of species, water types, moisture depletion, and their interactions on Pb²⁺ bioremediation and translocation. Results show significant effects of species (p = 0.00047), water types (p = 0.001), and their interaction (p = 0.001) on the bioremediation factor (BAF Pb²⁺), highlighting their importance in improving Pb²⁺ removal by sunflowers. Moisture depletion also significantly impacted Pb²⁺ remediation (p = 0.036), but higher-order interactions, such as species × moisture depletion and water types × moisture depletion, were not significant. The study also found that species and water types significantly influenced Pb²⁺ translocation (F = 6071.850, p = 0.00016), root Pb²⁺ concentration (p = 0.002), and soil Pb²⁺ levels (F = 451.077, p = 0.002). Additionally, species and water types significantly affected potassium (K⁺) uptake in roots (F = 601664.1, p < 0.0001), while moisture depletion had little effect. These findings emphasize the importance of selecting appropriate species and water management for optimizing Pb²⁺ bioremediation.

The urgent demand for effective and intelligent intrusion detection systems (IDSs) to address this problem has grown in tandem with the increasing complexity and diversity of cyberattacks. With deep learning (DL) methods, the presented study aims to build an integrated framework that raises the intrusion detection accuracy and generalizability across several data sources. In this work, three well-known datasets, UNSW-NB15, NSL-KDD, and CICIDS2017, have been utilized. The suggested approach uses convolutional neural networks (CNNs) to extract deep features from such datasets. Dimensionality is reduced and irrelevant features are eliminated using principal component analysis (PCA). Combining the three datasets using a multimodal deep autoencoder (MDAE) extracts a new dataset. Final classification utilizes random forest (RF) technology, efficiently and precisely classifying network traffic using combined data. Supported by high precision, F1 score, and recall of 99.9%, evaluation findings show an excellent accuracy of the model of 99.9%. These results show how well modern classification algorithms, feature extraction, and multi-modal data fusion methods can be combined to create a sophisticated intrusion detection model capable of efficiently and consistently defending against challenging attacks

The effect of quenched disorder on the critical behavior of the twodimensional
Ising model has been systematically investigated using
extensive Monte Carlo simulations and finite-size scaling analysis. To
verify the distinctive critical exponents of the pure universality class, the
clean 2D Ising system was initially examined. A binary distribution of
exchange couplings with different disorder concentrations was then used
to introduce disorder, enabling a thorough analysis of its effects on the
scaling characteristics close to the critical temperature. Thermodynamic
parameters such as correlation length, magnetization, and susceptibility
are analyzed, and the results show that disorder contributes significant
logarithmic corrections to the scaling laws while the leading critical
exponents stay close to those of the clean Ising model. These findings
verify that, despite disorder-dependent logarithmic changes to its scaling
behavior, the quenched binary Ising model maintains the Ising
universality class. Further evidence that weak disorder in two
dimensions causes marginally irrelevant perturbations that provide
logarithmic corrections to critical scaling is provided by the current
results, which are in agreement with theoretical expectations from
renormalization-group analysis and earlier numerical investigations.

An imbalance between free radicals and the antioxidant defense systems responsible for neutralizing them results in oxidative stress. Antioxidants are essential because they limit the accumulation of free radicals, thereby reducing oxidative stress withinthe body. Type 1 diabetes mellitus arises from the pancreas’ inability to produce insulin due to autoimmune destruction of insulin-secreting β-cells. In contrast, type 2 diabetes develops through two main stages: it begins with insulin resistance in peripheral tissues, which stimulates increased insulin secretion as a compensatory response, followed by a gradual decline in β-cell function, leading to inadequate insulin production and impaired blood glucose regulation. Both types of diabetes are characterized by chronic hyperglycemia, which, if poorly controlled, can result in serious complications. Persistent high glucose levels promote the generation of free radicals through several metabolic pathways, including advanced glycation end product (AGE) formation, activation of the hexosamine and polyol pathways, and glucose auto-oxidation. These pathways collectively intensify oxidative stress and contribute to the progression of diabetes-related microvascular and macrovascular complications. Antioxidants therefore play an important supportive role by scavenging free radicals, reducing oxidative damage, and enhancing the body’s antioxidant capacity. However, they are not considered a primary therapeutic treatment for diabetes; rather, they serve as adjunctive agents that help limit cellular injury associated with oxidative stress.

Metaverse-based learning environments generate large volumes of multimedia data from learner interactions, including eye-tracking, movement patterns, and auditory and behavioural cues. This high-dimensional data presents a computational challenge that impacts the speed and efficiency of adaptive learning systems. This study aims to improve feature selection by proposing a hybrid algorithm that combines genetic algorithms (GA) and hill-climbing algorithms (HCA) to achieve an effective balance between granular search and local optimisation. HCA enhances GA's ability to avoid premature convergence towards local solutions and improves the quality of selected features. The proposed algorithm was evaluated on six standard datasets, and the results showed a significant improvement in classification accuracy with fewer selected features than several established algorithms. In the context of metaverse-based learning, the proposed model reduces the volume of real-time data processed, enables immediate feedback, and improves the efficiency of adaptive analysis systems within immersive learning environments.

This study proposes an intelligent framework for automated brain tissue segmentation in MRI scans using U-Net-based convolutional neural networks, aiming to improve accuracy over existing methods. The methodology involves implementing both 2D and 3D U-Net architectures, trained and validated on public datasets (MRBrainS13, MRBrainS18, IBSR) with manual expert segmentations. Performance was compared against established tools (FSL, Dipy) using Dice Coefficient, Jaccard Index, AUC, and SSIM metrics. Results demonstrate the superiority of U-Net models, with Dice scores exceeding 0.9 for gray matter and showing high consistency, while 2D U-Net slightly outperformed the 3D variant in reducing false positives and negatives. The conclusion confirms that U-Net-based segmentation significantly enhances precision and reliability for distinguishing gray matter, white matter, and cerebrospinal fluid, offering a robust alternative to conventional techniques

Higher education institutions in developing countries continue to struggle with certificate authenticity due to the limitations of centralized verification systems. Although blockchain-based certificate verification offers a secure and transparent alternative, adoption intention remains low because behavioral determinants are insufficiently understood. This study develops and tests an integrated model combining Technology Readiness Index dimensions, diffusion of innovation factors, perceived complexity, compatibility, and resistance to change to explain intention to adopt blockchain-based certificate verification. Survey data were collected from 586 student record officers across 64 Iraqi universities and analyzed using PLS-SEM. Results indicate that innovativeness, optimism, discomfort, insecurity, and perceived compatibility significantly influence adoption intention, while perceived complexity is not significant. Resistance to change selectively moderates the effects of innovativeness and discomfort but no other predictors. The model demonstrates strong explanatory power, emphasizing the dominant role of psychological readiness and behavioral resistance over purely technical concerns. The study extends blockchain adoption literature by integrating multiple theoretical perspectives in an educational context and provides practical guidance for capacity building, training, and change management to support blockchain implementation in developing higher education systems.

In this work, isodecyl methacrylate monomer IDMA was was synthesized by esterification of methacrylic acid with isodecylalcohol. IDMA monomer was then copolymerized with acrylamide monomer AAM, in different feed compositions, using free radical polymerization technique. The synthesized IDMA and IDMA/AAM copolymer were characterized by employing proton nuclear magnetic resonance and Fourier transform infrared spectroscopy techniques. Composites of fabricated polymers with metallic mesh, having different pore size, were prepared to examine their performance for separation oil from water. Separation time, reclaimed oil, coating percentage with polymer, and water recovery were directly affected by the composition of copolymer and mesh pore size. The highest separation efficiency of copolymer was for 25 % IDMA based on 1000 micron mesh which was calculated as 99%. Water recovery was ranged from 75% to 92%. 75AAM/25IDMA copolymer, without crosslinker, separated oil from water faster than other compositions. Microscopic and photographic images of water, before and after separation by polymers, showed that oil was completely removed from water.

Schiff base ((E)-1,2-diphenyl-2-(pyridin-2-ylimino) nethan-1-ol) was synthesized from benzoin and 2-Aminopyridine, along with its Co (II), Ni (II), Cu (II), and Zn (II) complexes. synthesized compounds were characterized by ¹H-NMR, ¹³C-NMR, FT-IR, Uv-Vis, and molar conductivity measurements. Additionally, Cu (II) and Ni (II) nanocomplexes were prepared using Pulsed Laser Ablation in Liquid (PLAL) and characterized via FESEM, EDX, and XRD. Their antioxidant activities were evaluated using IC50 values, compared against standard antioxidants and their bulk counterparts, while molecular docking studies explored binding interactions with VEGFR2—a key target in liver and kidney cancer treatments due to its role in angiogenesis. Furthermore, Density Functional Theory (DFT) calculations were carried out at the B3LYP level using the 6-311G++(d,p) basis set to analyse the electronic properties of the compounds

A green electrochemical sensor was developed for the selective detection of sulfamethoxazole (SMX), a sulfonamide antibiotic, using silver nanoparticles (AgNPs) synthesized via an eco-friendly route employing lemon peel extract. The AgNPs were complexed with SMX to form ion-pair structures (SMX-AgNPs), which were subsequently incorporated into a graphite-based ion-selective electrode (ISE). Structural and morphological characterization of the synthesized nanomaterials was performed using UV–Vis spectroscopy, scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDX), and X-ray diffraction (XRD), confirming the formation of crystalline, nanoscale particles. The fabricated electrode demonstrated a wide linear response range (1×10⁻¹ to 1×10⁻⁷ M), fast response time, and high reproducibility, with operational stability maintained over a 65-day period. The sensor exhibited excellent sensitivity and selectivity, enabling accurate quantification of SMX in pharmaceutical formulations. These results underscore the potential of green-synthesized AgNPs in advancing low-cost, sustainable sensing technologies for drug monitoring.

Thyroid hormones influence the majority of bodily functions by directly impacting various physiological processes and the operation of numerous tissues. They are essential for the operation of other hormones. Aim of the study : Assessment level of glutathione (GSH), glutathione peroxidase (GSH-PX), and malondialdehyde (MDA), ,Thyroxine (T4) and triiodothyronine (T3),and lipid profile in hypothyroidism women before and after treatment with levothyroxine. Materials and methods: 60 women (before and after taking Levothyroxine treatment) and 30 control group were selected for the study. The study sample was selected from (15-54) years for the period from the end of September 2024 to the beginning of March 2025. Samples were taken from people who visit specialists in outpatient clinics and health centers in Samarra city. Result: The serum levels of GSH, GSH-PX, and MDA in hypothyroid women before and after treatment were significantly distinct from those in the control group (p < 0.001). The lipid profiles, including total cholesterol, triglycerides, low-density lipoprotein (LDL) cholesterol, and high-density lipoprotein (HDL) cholesterol, in the serum of hypothyroid women exhibited significant differences from those of the control group before and after treatment (p = 0.01). Thyroid hormone levels (TSH, T3, and T4) exhibited a substantial rise (p < 0.01) in the serum of hypothyroid women pre- and post-treatment in comparison to the control group. Conclusion: The current study concluded increase MDA in patient with hypothyroidism before treatment, while decrease after treatment. In contrast decrease both GSH, GSH-PX, and HDL in in women with hypothyroidism before treatment, while increase after treatment.

This study investigates the occurrence of liquid crystalline properties in chalcone esters with different hydrocarbon chain lengths. These compounds with the general formula [Br-Ph-CO-CH=CH-Ph-O-CO-CnH2n+1] were synthesized by the reaction of 4-bromoacetophenone with 4-hydroxybenzaldehyde, and then the product was reacted with carboxylic acid halides in the presence of suitable catalysts. The compounds were characterized by FT-IR, 1H-NMR and 13C-NMR. The occurrence of liquid crystalline properties was confirmed by differential scanning calorimetry (DSC), and the texture morphology was confirmed by polarized light microscopy (POM). Compounds A1, A7, and A9 were monomorphic, while compounds A8 and A10 were dimorphic, and the remaining compounds did not exhibit liquid crystalline properties.

Obesity is a medical condition characterized by the disproportionate storage of excessive body fat, which can negatively impact health. It is a critical public health problem. This study aimed to evaluate the levels of obestatin, adipsin, and neuropeptides, as well as some biochemical variables in obese males. The study was conducted on 90 obese males. Samples were collected from Samarra city and divided into groups based on body mass index (BMI): a control group, overweight group, and obese group. The samples were collected from males aged 25-45 years, between August 14, 2024, and September 30, 2024. The results showed an increase in BMI with overweight and obesity, reaching (28.12±0.91 and 34.64±2.83) kg/m2, respectively, compared to the ideal weight group (22.70±0.97). The current results showed a significant decrease in obestatin and glutathione levels at a probability value of (P>0.05) in the overweight group, being lower in the obese group compared to those of ideal weight. Meanwhile, adesipin, neuropeptide A, and malondialdehyde levels increased in the overweight and obese groups compared to those of ideal weight at a probability level of (P>0.05). We conclude from the results that levels of obestatin, glutathione, and high-density lipoprotein (HDL) cholesterol were lower in overweight and obese men compared to those of ideal weight. Levels of adesipin, neuropeptide A, total lipids, triglycerides, low-density lipoprotein (LDL) cholesterol, very-low-density lipoprotein (VLDL) cholesterol, glucose, and malondialdehyde were higher in overweight and obese men compared to those of ideal weight.

In this study, zinc oxide nanoparticles were
prepared using Tradescantia spathacea leaves extract; being
an environmentally friendly and inexpensive material that has
not been used previously for this purpose and characterizing
it with both (XRD), (EDX) and (FESEM); and working on
mixing (3 wt%) of the mixture Cetyltrimethylammonium
bromide/Sodium Dodecyle Sulphate (CTAB/SDS) as a
surfactant with ZnONPs as a tertiary mixture, with the aim of
making CTAB and SDS more effective and with fewer damages
by employment lower doses of them. effect of different
concentrations of the zinc oxide nanoparticle solution on the
ratio (2:8) was studied the most viscous of the binary
mixtures of surfactants (SDS/CTAB), which is classified as
containing more micelles than in other mixing ratios. The
results showed that the highest value (2705.223 Pa.s×102)
was at a temperature of 298 K, and the lowest value (1.506897
Pa.s×102 ) was at a temperature of 323 K, also showed that
∆Go values generally increased with increasing temperature
and the higher concentration of nanomaterial gave lower ∆Go
values. In addition, it was observed that the mixing exothermic
process and the entropy change was negative, meaning that
the system is in a state where the arrangement of molecules is
regular.

In this paper, a reliability formula was found for a cascade model
containing four units (the first unit ????1 is basic and the other three
units ????2, ????3, and ????4 are redundant standby). It was assumed that the
two variables strength and stress follow the exponential distribution
of the variables. The estimation of the exponential distribution
parameters was found using three estimation methods (Maximum
likelihood, Percentile, and Least Squares), these methods were used to
estimate the reliability of the model. MATLAB was used in the Monte
Carlo simulation and mean square errs was used to compare the
simulation results and find which estimation methods are the best for
reliability estimation. The simulation results proved that the
estimator ML is the best for estimating the model's reliability.

It is known that one of the most paramount and complicated
problems in variational inequality Theory (VIT) is the development
of an efficacious and implementable approximation schemes for
solving diverse categories of variational inequality. Thus, through
this study we introduce a new iterative scheme with errors to solve
generalized strongly nonlinear implicit quasivariational inequalities
(GSNIQVI) by fixed point defined on for a set-valued mapping ∅ ≠ ???? ̅
⊂????,???? ̅ :Ξ ⟶ 2Ξ with closed convex values in Hilbert space Ξ. These
iterative schemes are constructed for Lipschitz continuous mappings,
Γ-strongly monotone, and ℮-strongly monotone mappings under
some appropriate conditions. These iterative schemes is solving
special case of GSNIQVI with errors and without errors. We proved
an existence and uniqueness results for solution GNIQVIP and
convergence of new iterative algorithms with errors, and without
errors, for several mappings: the first Γ:Ξ ⟶ Ξ is a Lipschitz
continuous mapping, the second Â, Ĉ, Ê, Û:Ξ ⟶ Ξ are Lipschitz
continuous mappings, the third Ň:Ξ4 ⟶ Ξ is a mapping such that
Ň(Â,Ĉ,Ê,Û) is a strongly monotone with respect to  in Ň1, Γ
strongly monotone with respect to Ĉ in Ň2, ℮-strongly monotone
with respect to Ê in Ň3 and Lipschitz continuous with respect to
Ň1,Ň2,Ň3, Ň4and the fourth P???? ̅(ğ) :???? ̅ → Ξ is the metric projection
which is Lipschitz continuous. Our results is inspired and encouraged
for several research works in the literature.

This study was conducted to determine the effect of water stress on the
growth and chemical content of lemongrass plants and the efficiency of
spray fertilization and fertilizer addition in reducing the harmful effect
of drought stress under greenhouse conditions. The experiment was
designed according to the Complete Randomize Design (CRD), with
three replicates and two plants per experimental unit. The experiment
included two factors, the first was fertilization using the organic
fertilizer ORGEVIT after a stage of 5 leaves mixed with the soil, at two
levels of 250 g and 125 g, and the second factor was fertilizing by foliar
spraying using the organic stimulant Maxi-Grow, at two concentrations
of 500 mg/l and 250 mg/l. The third factor included irrigation
intervals, at a rate of every 10 and 15 days. The replicates and their
combinations resulted in 9 treatments for each replicate, including the
control treatment, which was placed in the designated place in the
greenhouse after cleaning and arranging it. The vegetative traits of
lemongrass plants were studied, and the results of the vegetative traits
showed that the highest value for leaf area and root length was in the
fertilization factor of 250 kg/ha and 250 mg/l, while the results
showed that the highest values for the fresh weight and dry weight
traits were in the fertilization factor of 250 mg/l. liter and 500 mg/liter

Optical Coherence Tomography (OCT) is a common technique that is used to
identify retinal caused diseases choroidal neovascularization (CNV), diabetic
macular edema (DME), and drusen-related degeneration. Nonetheless, there
are difficult aspects of training strong deep learning models to the
classification of OCT, such as privacy, multi-institutional (heterogeneous)
distributions of data, and the scarcity of labeled annotations. Our proposed
federated learning framework in this work consists of two stages of
classification-based privacy-preserving retinal disease classification self
supervised contrastive pretraining followed by variance-reduced federated
optimization frameworks. The first stage involves clients learning OCT
specific features through a contrastive learning method based on momentum
(MoCo v3) to do local self-supervised representation learning without the need
of labeled data. The second step involves fine-tuning of the pretrained encoder
with the SCAFFOLD algorithm of supervised federated learning to address
client drift when data are distributed by non-IID. EfficientNet-B4 is the
adopted backbone architecture to provide both a balance in the performance
of classification and communication efficiency. Multi-institution OCT
datasets are experimented on Kermany OCT and Duke OCT datasets and
further cross-domain testing is done on the OCTID dataset. The presented
framework gets 96.8 percent accuracy in classification and AUC-ROC of
0.994 on in-domain evaluation, surpassing a variety of centralized and
federated baselines. Moreover, cross-domain experiments are associated with
a stronger generalization to unseen OCT data.These findings indicate the
potential effectiveness of federated learning when combined with contrastive
pretraining to achieve greater robustness in the case of heterogeneous data
distribution and maintain the privacy of patient data.