Iraqi Journal for Computer Science and Mathematics

Basketball players in the NBA are renowned for their talent, athleticism, and commitment to the game. NBA players
may make enormous sums of money; however, they vary greatly. Rookie agreements begin at a lower price and go up
following performance. NBA players’ pays are influenced by several factors. Because they influence games and the success
of the club, exceptional players fetch larger compensation. This study employs Machine Learning (ML) techniques,
including Lasso Regression and Random Forest Regression (RFR) models to analyze wage trends, enhanced by the Slime
Mould Algorithm (SMA) and Artificial Rabbit Optimization (ARO) for accuracy. The goal is to forecast NBA player
salaries based on their performance. Upon analysis, the RFAR model emerged as the top-performing model across the
train, test, and validation phases, showcasing impressive R2 values of 0.987 for both the train and test phases and 0.970
during validation. In contrast, the LASSO model was identified as the weakest performer, with R2 values of 0.906 during
training, 0.887 in the validation phase, and 0.910 R2 values during the testing phase. In conclusion, RFAR is the most
effective model for predicting NBA players’ salaries.

COVID-19 was diagnosed using deep learning models by a group of studies. Evaluating and benchmarking these models
are essential to achieving the most suitable model for diagnosing coronavirus. Objective: In this investigation, we offer
an inclusive valuation of several deep learning models to detect the maximum appropriate and active model which
gratifies doctors’ requirements and assessment criteria. Method: This study combines Fuzzy decision by the opinion
score method (FDOSM) and Fuzzy-Weighted Zero-Inconsistency (FWZIC). According to the advantage of Trapezoidal
Intuitionistic fuzzy, we developed FWZIC into Trapezoidal Intuitionistic fuzzy named (TrIF-FWZIC) for weighting criteria
and FDOSM into Trapezoidal Intuitionistic fuzzy FDOSM (TrIF-FDOSM) to evaluate and benchmark the effectively deep
learning models and tackle the issue of uncertainty. Fundamentally, the methodology of this study is presented in 2
phases; the 1st phase is related to identifying a new decision matrix containing 24 evaluation criteria to evaluate the ten
deep learning models. Furthermore, the 2nd phase is related to the development of TrIF-FWZIC and TrIF-FDOSM in two
main stages. Result: The findings of this study were: (1) For the individual decision-maker, the best one was Xception for
the first decision-maker with a score (i.e., 0.267510407). The optimal algorithm for the 2nd and 3rd decision-makers was
ResNet-101 with scores (i.e., 0.316710828, 0.457770263), respectively. (2) The best deep learning model, depending
on the group decision-making, was ResNet-101 with a score (i.e., 0.32574743). Conclusion: The proposed methodology
undergoes validation, sensitivity analysis, and comparative evaluation. This research enhances the selection of effective
models for COVID-19 diagnosis, catering to individual and collective decision-making scenarios.

Intracranial hemorrhage (ICH) denotes bleeding inside the skull, which can occur in or around the brain. Computed
tomography (CT) has been used to detect ICH due to its high efficiency and accuracy. Nowadays, deep learning model
design is introduced to allow an accurate and efficient classification of ICH in CT images. This work focused on developing
U-Net-based models for the segmenting of ICH. Furthermore, the proposed model employed two transfer learning models,
MobileNet and Xception, as the backbones of the U-Net topology. This approach aims to establish metrics that improve
ICH treatment through precise segmentation techniques. A free dataset from the Radiological Society of North America
(RSNA) was fed to the proposed model. This dataset comprises 5996 annotated CT images divided into four groups, with
1499 images per group. Three of these groups are for hemorrhage, and one is for the normal group. The average detection
accuracy rates were impressive, with Intraparenchymal Hemorrhage (IPH) at 98%, Intraventricular Hemorrhage (IVH) at
98%, and Subdural Hemorrhage (SDH) achieving accuracy of 97%. This approach could assist radiologists in overloaded
medical centers in precisely detecting ICH, especially in the local healthcare centers.

This paper involves adopting a well-known generalization method in the branches that dealing with fixed points via
weakening the suppositions. As appearing in previous sources, letting (, Sb, K) be a strong b-MS, and F, H be two
multi-valued maps on equipped with a graph σ s.t the set of vertices of σ , 3(σ ) = and the set of edges of σ , 4(σ )
⊆ × . The acceptable assupmtions have been adopted to finding a common fixed point for in . Our result is a
generalization to other statuses.

COVID-19 is a highly contagious viral infection that primarily affects the respiratory system, causing symptoms such
as high fever, cough, and severe respiratory distress. Early detection of the disease is of utmost importance to control the
spread and severity. Common diagnostic methods include Reverse Transcription Polymerase Chain Reaction (RT-PCR),
antigen tests, chest X-rays, and computed tomography (CT) scans. Similarly, viral pneumonia, another severe lung
infection, leads to fluid or pus accumulation in the lungs, causing symptoms such as chest pain, fatigue, excessive
sweating, and nausea. The elderly and young children are particularly vulnerable to severe complications. Similarly,
viral pneumonia, another severe lung infection, leads to fluid or pus accumulation in the lungs, causing symptoms such as
chest pain, fatigue, excessive sweating, and nausea. The elderly and young children are particularly vulnerable to severe
complications. This paper aims to get the bottom of the ResNet-34 model, for detecting COVID-19 and Viral Pneumonia
using radiography images and to build a web application using a Streamlit framework for detecting the presence of the
disease. The ResNet-34 is a Convolutional Neural Network (CNN) model used to classify COVID-19, Viral Pneumonia,
and normal (negative) among 15000 Chest X-ray images. In this work, datasets from Kaggle repository have considered
that help to acquire Chest X-ray images as these are extensively used to diagnose COVID-19 and Viral Pneumonia as
they give clear insights into the lungs. The presented resNet-34 model, helps to discriminate between COVID-19 and
Viral Pneumonia with high accuracy and precision.

Reliability-based design optimization (RBDO) determines optimal design parameters by incorporating reliability
constraints. This paper presents a RBDO approach using differential-algebraic equations (DAEs) for modeling and
constraints. DAEs provide an accurate representation of dynamic engineering systems with coupled differential and
algebraic equations. However, the nonlinearity and implicit nature of DAEs pose challenges for uncertainty propagation
and optimization. This study proposes an efficient RBDO methodology based on stochastic collocation to quantify
uncertainty in DAEs. The DAEs are transformed into an explicit ODE system to enable direct uncertainty analysis via
sampling. Optimization under reliability constraints is achieved using a sequential approximate programming strategy.
The approach is demonstrated through application to optimal design of a chemical reactor system. The results indicate
the DAE-based RBDO framework provides an efficient way to optimize design reliability. This methodology enables
reliable design optimization for complex coupled systems across energy, aerospace, chemical, and other DAE-based
engineering applications.
The abstract highlights the key points of using DAEs in RBDO, the proposed methods to handle uncertainty analysis
and optimization for DAEs, and the advantages of a DAE-based approach. Let me know if you would like me to modify
or expand the abstract further.

Image denoising plays a vital role in enhancing visual quality by effectively suppressing noise while retaining
critical image structures and textures. Traditional Block-Matching and 3D (BM3D) Filtering techniques, although widely
adopted, often encounter challenges in achieving an optimal trade-off between noise reduction and feature preservation
due to limitations in fixed-thresholding strategies and suboptimal block matching. To address these shortcomings,
this study introduces a novel Cosine Adaptive BM3D (CA-BM3D) approach, which integrates cosine similarity for
more accurate block matching and incorporates adaptive thresholding to enhance denoising efficiency. The proposed
method was evaluated on six standard 8-bit grayscale images such as Leena (512*512), Barbara (512*512), Zelda
(512*512), peppers (512*512), Cameraman (512*512), House (512*512) contaminated with Additive White Gaussian
Noise (AWGN) Results demonstrate that the Cosine Adaptive BM3D algorithm consistently outperforms conventional
BM3D across all test cases, achieving an average PSNR of 31.28 dB and SSIM of 0.678. Notably, for the “Leena” image, the
cosine-based approach attained a PSNR of 31.42 dB and SSIM of 0.8921, surpassing alternative distance metrics such
as Euclidean (30.55 dB, 0.8782), Manhattan (29.96 dB, 0.8657), Jaccard (28.78 dB, 0.8425), and Minkowski (30.02
dB, 0.8691). Similar performance gains were observed for the other images: “Barbara” (30.33 dB, 0.8856), “Zelda”
(32.15 dB, 0.9012), “Peppers” (31.78 dB, 0.8967), “Cameraman” (30.02 dB, 0.8823), and “House” (32.89 dB, 0.9105).
These findings substantiate the effectiveness of the cosine similarity measure in enhancing both noise suppression and
structural fidelity in the denoising process.

This article investigates the M-fractional complex Ginzburg-Landau (CGL) equation by utilizing the modified simplest
equation method and the modified G′
G -expansion method. The CGL holds significant importance in the realm of oscillatory
phenomena and optical fibers. By implementing a fractional complex transformation, the M-fractional CGL equation is
transformed into an ordinary differential equation (ODE). By employing these methodologies, bright periodic solutions,
periodic solutions, and dark wave solutions relevant to phenomena witnessed in nonlinear optics or plasma environments
are ascertained. It is demonstrated that these methods are very powerful and efficient to solve many fractional differential
equations.

This study presents an innovative application of Feedforward Neural Networks ‘FNNs’ to solve Variable-Order
Fractional Partial Differential Equations ‘VO-FPDEs’ with time delays. Utilizing the Caputo definition, the variable-
order fractional derivatives are approximated in terms of integer-order derivatives. The problem is reformulated as
a system of partial differential equations with delay terms, which is then addressed using ‘FNNs’ to achieve explicit
approximate solutions. Comprehensive error and convergence analyses validate the method’s precision and reliability.
The effectiveness of the proposed approach is highlighted through numerical examples, with graphical and tabular
representations showcasing minimal absolute errors and robust convergence. These results demonstrate the proposed
method’s efficiency and simplicity, establishing it as a powerful tool for addressing complex fractional delay problems.

Parkinson’s disease (PD) is a progressive neurological disorder that primarily affects individuals over the age of 55.
It is characterized by a range of motor and non-motor symptoms that can significantly impact various aspects of daily
life. Despite notable advancements in medical science, there is currently no permanent cure or definitive treatment
for PD. This therapeutic gap underscores the critical importance of early diagnosis, which remains a major focus of
ongoing research. Due to the disease’s gradual progression, PD symptoms may take years to fully develop, making early
detection essential for improving patient outcomes and quality of life. Moreover, the clinical presentation of PD overlaps
with several other neurological conditions, highlighting the need for accurate and reliable diagnostic methods. This study
proposes a novel framework for analyzing auditory data to enhance the precision of Parkinson’s disease diagnosis. By
integrating quantum computing with neural network architectures, the proposed approach aims to improve diagnostic
accuracy and support early detection and intervention. The quantum component facilitates more precise and efficient
computations, while the neural model enhances information flow and feature extraction. Specifically, a Quantum
Convolutional Neural Network (QCNN) is developed and evaluated using auditory data for PD diagnosis. The QCNN
demonstrated exceptional performance on Dataset C, which includes 1,000 samples, achieving an accuracy of 99%, a
precision of 99.5%, a recall of 99.5%, and an F1-score of 100%. Compared to classical convolutional neural networks,
the proposed QCNN offers several advantages. Quantum algorithms are capable of performing specific computations
significantly faster than their classical counterparts, leading to reduced processing time. Furthermore, they allow for
a more compact representation of features, which lowers model complexity while enhancing overall performance and
diagnostic accuracy.