Step Of Algorithm Research Articles

Modeling of turbulent fluid flow based on the solution of the Mathieu equation

Objective. This work examines the problem of fluid flow simulation in a turbulent regime. The main reasons why turbulent flow occurs are due to the existence of high velocities of movement in fluids; besides that, there may be obstacles or changes in the shapes of flows. Method. To determine the characteristics of a flow, it is proposed to use Mathieu’s equation. The main steps of the solution algorithm are presented for Mathieu functions that were used over the course of the computer program execution. Result. Upon evaluating eigenvalues, it is shown that it is necessary to rely on the corresponding transcendental equations. It is illustrated how modified Mathieu functions are calculated. It is indicated how calculations should be carried out in the case of a small parameter value in Mathieu functions. A block diagram of the executed turbulence modeling algorithm is presented. In order to conduct turbulence simulation, a GUI shell based on the Qt library was created in the Qt Creator environment. During the simulation process, the value of the Smagorinsky constant was chosen to be equal to 0.01. For the time step, a length value of 0.05 was selected. During the execution of the simulation, 2000 time steps were used. The simulation results were recorded every 10 steps. The results of the simulation are presented visually. Conclusion. A mathematical model has been created, on the basis of which there are possibilities for modeling turbulent media. The mathematical model has been constructed for various parameters during the flow of solid bodies by turbulent flows.

Real-Time Quantification of Gas Leaks Using a Snapshot Infrared Spectral Imager.

We describe the various steps of a gas imaging algorithm developed for detecting, identifying, and quantifying gas leaks using data from a snapshot infrared spectral imager. The spectral video stream delivered by the hardware allows the system to combine spatial, spectral, and temporal correlations into the gas detection algorithm, which significantly improves its measurement sensitivity in comparison to non-spectral video, and also in comparison to scanning spectral imaging. After describing the special calibration needs of the hardware, we show how to regularize the gas detection/identification for optimal performance, provide example SNR spectral images, and discuss the effects of humidity and absorption nonlinearity on detection and quantification.

Hepatocellular Carcinoma: Imaging Advances in 2024 with a Focus on Magnetic Resonance Imaging.

The EASL diagnostic algorithm for hepatocellular carcinoma, currently in use, dates back to 2018. While awaiting its update, numerous advancements have emerged in the field of hepatocellular carcinoma imaging. These innovations impact every step of the diagnostic algorithm, from surveillance protocols to diagnostic processes, encompassing aspects preceding a patient's inclusion in surveillance programs as well as the potential applications of imaging after the hepatocellular carcinoma diagnosis. Notably, these diagnostic advancements are particularly evident in the domain of magnetic resonance imaging. For example, the sensitivity of ultrasound in diagnosing very early-stage and early-stage hepatocellular carcinoma during the surveillance phase is very low (less than 50%) and a potential improvement in this sensitivity value could be achieved by using abbreviated protocols in magnetic resonance imaging. The aim of this review is to explore the 2024 updates in magnetic resonance imaging for hepatocellular carcinoma, with a focus on its role in surveillance, nodular size assessment, post-diagnosis imaging applications, and its potential role before the initiation of surveillance.

Conditional Generative Adversarial Net based Feature Extraction along with Scalable Weakly Supervised Clustering for Facial Expression Classification

Article Conditional Generative Adversarial Net based Feature Extraction along with Scalable Weakly Supervised Clustering for Facial Expression Classification Ze Chen 1, Lu Zhang 2, Jiaming Tang 3, Jiafa Mao 3, and Weiguo Sheng 1,* 1 Department of Computer Science, Hangzhou Normal University, Hangzhou, P.R. China 2 China Telecom Hangzhou Branch, Hangzhou, P.R. China 3 School of Computer Science and Technology, Zhejiang University of Technology, Hangzhou, P.R. China * Correspondence: w.sheng@ieee.org Received: 28 September 2023 Accepted: 30 June 2024 Published: 24 December 2024 Abstract: Extracting proper features plays a pivotal role in facial expression recognition. In this paper, we propose to extract facial expression features via a conditional generative adversarial net, followed by an algorithmic optimization step. These refined features are subsequently integrated into a scalable weakly supervised clustering framework for facial expression classification. Our results show that the proposed method can achieve an average recognition rate of 85.3%, which significantly outperforms related methods. Further, by employing a residual-based scheme for feature extraction, our method shows superior adaptability compared to algorithms based solely on weakly supervised clustering. Additionally, our method does not require high accurate annotation data and is robust to the noise presented in data sets.

Cost-Effectiveness of rTMS as a Next Step in Antidepressant Non-Responders: A Randomized Comparison With Current Antidepressant Treatment Approaches.

Although repetitive transcranial magnetic stimulation (rTMS) is an effective and commonly used treatment option for treatment-resistant depression, its cost-effectiveness remains much less studied. In particular, the comparative cost-effectiveness of rTMS and other treatment options, such as antidepressant medication, has not been investigated. An economic evaluation with 12 months follow-up was conducted in the Dutch care setting as part of a pragmatic multicenter randomized controlled trial, in which patients with treatment-resistant depression were randomized to treatment with rTMS or treatment with the next pharmacological step according to the treatment algorithm. Missing data were handled with single imputations using predictive mean matching (PMM) nested in bootstraps. Incremental cost-effectiveness and cost-utility ratios (ICERs/ICURs) were calculated, as well as cost-effectiveness planes and cost-effectiveness acceptability curves (CEACs). Higher QALYs, response, and remission rates were found for lower costs when comparing the rTMS group to the medication group. After 12 months, QALYs were 0.618 in the rTMS group and 0.545 in the medication group. The response was 27.1% and 24.4% and remission was 25.0% and 17.1%, respectively. Incremental costs for rTMS were -€2.280, resulting in a dominant ICUR for QALYs and ICER for response and remission. rTMS appears to be a cost-effective treatment option for treatment-resistant depression when compared to the next pharmacological treatment step. The results support the implementation of rTMS as a step in the treatment algorithm for depression. The trial is registered within the Netherlands Trial Register (code: NL7628, date: March 29, 2019).

K-hyperparameter tuning in high-dimensional genomics using joint optimization of deep differential evolutionary algorithm and unsupervised transfer learning from intelligent GenoUMAP embeddings

AbstractK-hyperparameter optimization in high-dimensional genomics remains a critical challenge, impacting the quality of clustering. Improved quality of clustering can enhance models for predicting patient outcomes and identifying personalized treatment plans. Subsequently, these enhanced models can facilitate the discovery of biomarkers, which can be essential for early diagnosis, prognosis, and treatment response in cancer research. Our paper addresses this challenge through a four-fold approach. Firstly, we empirically evaluate the k-hyperparameter optimization algorithms in genomics analysis using a correlation based feature selection method and a stratified k-fold cross-validation strategy. Secondly, we evaluate the performance of the best optimization algorithm in the first step using a variety of the dimensionality reduction methods applied for reducing the hyperparameter search spaces in genomics. Building on the two, we propose a novel algorithm for this optimization problem in the third step, employing a joint optimization of Deep-Differential-Evolutionary Algorithm and Unsupervised Transfer Learning from Intelligent GenoUMAP (Uniform Manifold Approximation and Projection). Finally, we compare it with the existing algorithms and validate its effectiveness. Our approach leverages UMAP pre-trained special autoencoder and integrates a deep-differential-evolutionary algorithm in tuning k. These choices are based on empirical analysis results. The novel algorithm balances population size for exploration and exploitation, helping to find diverse solutions and the global optimum. The learning rate balances iterations and convergence speed, leading to stable convergence towards the global optimum. UMAP’s superior performance, demonstrated by short whiskers and higher median values in the comparative analysis, informs its choice for training the special autoencoder in the new algorithm. The algorithm enhances clustering by balancing reconstruction accuracy, local structure preservation, and cluster compactness. The comprehensive loss function optimizes clustering quality, promotes hyperparameter diversity, and facilitates effective knowledge transfer. This algorithm’s multi-objective joint optimization makes it effective in genomics data analysis. The validation on this algorithm on three genomic datasets demonstrates superior clustering scores. Additionally, the convergence plots indicate relatively smoother curves and an excellent fitness landscape. These findings hold significant promise for advancing cancer research and computational genomics at large.

Accelerated enumeration of extreme rays through a positive-definite elementarity test.

Analysis of metabolic networks through extreme rays such as Extreme Pathways and Elementary Flux Modes has been shown to be effective for many applications. However, due to the combinatorial explosion of candidate vectors, their enumeration is currently limited to small- and medium-scale networks (typically less than 200 reactions). Partial enumeration of the extreme rays is shown to be possible, but either relies on generating them one-by-one or by implementing a sampling step in the enumeration algorithms. Sampling-based enumeration can be achieved through the Canonical Basis Approach (CBA) or the Nullspace approach (NSA). Both algorithms are very efficient in medium-scale networks, but struggle with elementarity testing in sampling-based enumeration of larger networks. In this paper, a novel elementarity test is defined and exploited, resulting in significant speedup of the enumeration. Even though NSA is currently considered more effective, the novel elementarity test allows CBA to significantly outpace NSA. This is shown through two case studies, ranging from a medium-scale network to a genome-scale metabolic network with over 600 reactions. Extreme Pathways are chosen as the extreme rays in this study, but the novel elementarity test and CBA are equally applicable to the other types. With the increasing complexity of metabolic networks in recent years, CBA with the novel elementarity test shows even more promise as its advantages grows with increased network complexity. Given this scaling aspect, CBA is now the faster method for enumerating extreme rays in genome-scale metabolic networks. All case studies are implemented in Python. The codebase used to generate Extreme Pathways using the different approaches is available at https://gitlab.kuleuven.be/biotec-plus/pos-def-ep. Supplementary data are available at Bioinformatics online.

Perspectives in the Development of Modern Bulgarianentrepreneurship

The present development is an attempt to critically understand the problems of Bulgarian entrepreneurship at the current historical moment. Processes and trends are reported, both in the country and on a wider scale. Attention is focused on delineating a vision for development consistent with contemporary priorities. An algorithm of key steps is marked, about the ultimate goal – achieving sustainable development. The aspiration is, by carefully analyzing the situation – opportunities, risks, and limitations, to propose a new, more realistic time horizon for the implementation of significant transformations in our country. Development perspectives are outlined and a system of mutually complementary models for sustainable development of Bulgarian entrepreneurship is proposed

Quantum implementation of bilinear interpolation algorithm based on NEQR and center alignment

Abstract In the field of quantum image processing, scaling techniques have been extensively studied as a critical subfield. Predominantly, these techniques leverage bilinear interpolation method. However, the current quantum bilinear interpolation algorithm remains in its initial version, resulting in a significant flaw: the centers of the source image and the interpolated image are misaligned. This flaw not only leads to results identical to nearest neighbor interpolation during integer multiple image scaling-down but also causes excessive qubit usage. To tackle these issues, our study introduces an innovative bilinear interpolation method with the following key advancements. Firstly, by integrating the NEQR representation with an enhanced bilinear interpolation from OpenCV, the central misalignment is effectively rectified. Secondly, resolving the central misalignment enables our method to prevent the degradation to nearest neighbor interpolation during integer multiple image reduction. Finally, based on the characteristics of optimization algorithms and quantum computing, along with simpler quantum modules and quantum convolution algorithms, the steps of the quantum image scaling algorithm have been optimized to reduce circuit complexity. Through a series of simulation experiments and complexity analyses, it has been demonstrated that our method, compared to the initial bilinear interpolation method, exhibits lower circuit complexity and significantly reduces qubit usage.

Evaluating a Child Presenting with Symptoms Concerning for Sjögren Disease: A Clinical Tool for Practice.

Childhood Sjögren disease (cSjD) is a rare disease. There are no widely accepted diagnostic or classification criteria for cSjD. To fill this gap, members from CARRA Sjogren Workgroup and the International cSjD Workgroup created a clinical diagnostic algorithm. This study evaluates the accuracy of this algorithm using an international cohort of participants with clinician diagnosed cSjD. First, experts developed a cSjD diagnostic algorithm through a series of virtual workgroup meetings. Using the adult classification criteria as a framework experts modified the algorithm through opinion and literature review. The group discussed and finalized each algorithm step by achieving majority rule. Then, R statistical software evaluated each participant's disease status in the diagnostic algorithm via an international cohort of 300 cSjD cases. The diagnostic algorithm has three distinct clinical pathways, which represents key clinical presentation in cSjD: parotitis, extraglandular manifestations, and sicca symptoms. The algorithm showed an overall sensitivity of 75% in the population that had enough data to complete at least 1 pathway of the algorithm (n = 100 filtered out of 300). Parotitis (70%) and sicca pathways (82%) had the highest sensitivity and extraglandular pathways (52%) had the lowest. As cSjD lacks a diagnostic strategy, this algorithm provides a clinical tool for evaluating children with cSjD-like symptoms. It performed well in an international cohort of cSjD, supporting the integration of this algorithm into clinical practice; however, its utility may be limited by low utilization of diagnostic testing in this population.

Space-time tree: a spatiotemporal construct for efficient similarity matrix calculations among network-constrained trajectories

Data mining of network-constrained trajectories has broad applications in the GIScience field. The calculation of a complete trajectory similarity matrix is a key step in various data mining algorithms. However, computing this matrix is computationally intensive for large datasets, as it involves numerous point-to-point shortest-path (PPSP) queries. To tackle this issue, we propose a new spatiotemporal construct called the space-time tree, which directly delineates the network distance from a query trajectory to any network space-time point. By constructing the space-time tree, we can efficiently compute the trajectory similarity matrix without additional PPSP queries. The space-time tree supports several similarity metrics, including closest pair distance, furthest pair distance, longest common subsequence (LCSS), and distance-weighted LCSS. It can further integrate with advanced spatiotemporal query techniques for scalable partial trajectory similarity matrix calculations. A case study using real datasets was conducted to apply the space-time tree in the trajectory clustering application. The results show that the space-time tree completed the clustering task on 0.5 million trajectories within 49 minutes, achieving a nearly 147-fold speedup compared to state-of-the-art methods.

Comparison of the STANDARD M10 C. difficile, Xpert C. difficile, and BD MAX Cdiff assays as confirmatory tests in a two-step algorithm for diagnosing Clostridioides difficile infection.

Current guidelines recommend a two-step algorithm rather than relying solely on a single test for diagnosing Clostridioides difficile infection. This algorithm starts with enzyme immunoassay (EIA) for detecting glutamate dehydrogenase (GDH) and toxins A/B, followed by nucleic acid amplification test (NAAT) for GDH-positive but toxin-negative cases. This study compared the performance of three commercial NAATs: the STANDARD M10 C. difficile, Xpert C. difficile, and BD MAX Cdiff assays, utilized as confirmatory testing of the two-step algorithm. Two hundred archived stool specimens, previously tested GDH-positive but toxin-negative by EIA, were analyzed in parallel with these NAATs and toxigenic culture, which served as the reference standard. Sensitivity, specificity, positive predictive value, and negative predictive value were 89.1%, 92.6%, 94.6%, and 85.2%, respectively, for the M10 assay; 95.8%, 86.4%, 91.2%, and 93.3%, respectively, for the Xpert assay; and 89.8%, 91.4%, 93.8%, and 86.0%, respectively, for the BD MAX assay. The rates of invalid results were 1.0%, 0.5%, and 1.0% for the M10, Xpert, and BD MAX assays, respectively. In conclusion, the M10 assay is a reliable diagnostic tool, performing comparably to the Xpert and BD MAX assays when used as confirmatory testing in the two-step algorithm.IMPORTANCEWhile numerous studies have assessed nucleic acid amplification tests (NAATs) as stand-alone tests for diagnosing Clostridioides difficile infection, limited research has compared their performance as confirmatory tests in a two-step algorithm. This study evaluated the performance of three commercial NAATs (M10, Xpert, and BD MAX assays) using 200 archived stool specimens initially tested as glutamate dehydrogenase (GDH)-positive but toxin-negative by GDH/toxin A/B enzyme immunoassay, the first step in the two-step algorithm. All three assays demonstrated high sensitivity (89.1% to 95.8%) and specificity (86.4% to 92.6%), with low rates of invalid results (≤1%). Our findings suggest that the M10 assay performs comparably to the Xpert and BD MAX assays when used as confirmatory testing in the two-step algorithm. Offering similar performance and turnaround time to these widely used assays at a slightly lower cost, the M10 assay serves as a practical alternative in this setting.

A Unified Approach to Aitchison’s, Dually Affine, and Transport Geometries of the Probability Simplex

A critical processing step for AI algorithms is mapping the raw data to a landscape where the similarity of two data points is conveniently defined. Frequently, when the data points are compositions of probability functions, the similarity is reduced to affine geometric concepts; the basic notion is that of the straight line connecting two data points, defined as a zero-acceleration line segment. This paper provides an axiomatic presentation of the probability simplex’s most commonly used affine geometries. One result is a coherent presentation of gradient flow in Aichinson’s compositional data, Amari’s information geometry, the Kantorivich distance, and the Lagrangian optimization of the probability simplex.

Clinical Characteristics of Persistent Hypophosphatasemia Uncovered in Adult Patients: A Retrospective Study at a Japanese Tertiary Hospital.

Background: Hypophosphatasemia is often overlooked despite its potential to indicate underlying pathologies. The aim of this study was to determine the prevalence of persistent hypophosphatasemia in a large, urban, multi-specialty hospital population and characterize the clinical and laboratory findings in adult patients with this condition. Methods: In this retrospective observational study, the results of 424,434 alkaline phosphatase (ALP) tests in 50,136 patients aged ≥18 years that were performed at Okayama University Hospital between July 2020 and October 2023 were analyzed. Persistent hypophosphatasemia was defined as consistently low ALP levels (≤40 IU/L) for 28 days with a minimum recorded level of ≤35 IU/L. Results: Persistent hypophosphatasemia was detected in 273 patients (0.54% of the tested patients), and the patients with persistent hypophosphatasemia included a higher proportion of females (72.5% vs. 52.9% in the people without persistent hypophosphatasemia; chi-squared test, p < 0.01) and had a younger median age (51 years vs. 63 years; Mann-Whitney U test, p < 0.01) than those in the overall tested population. The common causes of persistent hypophosphatasemia were cancer (30%), glucocorticoid use (21%), and immunosuppressants (16%). Notably, 38 patients (14%) had no apparent cause for low ALP values. These patients were categorized on the basis of their clinical characteristics, with some patients presenting symptoms potentially related to adult hypophosphatasia. Conclusions: This study provides prevalence and insights into the causes and characteristics of persistent hypophosphatasemia in a Japanese tertiary care setting. While most cases were associated with known causes, patients with unexplained hypophosphatasemia and symptoms such as chronic pain, muscle weakness, and general fatigue could have adult hypophosphatasia. In such cases, comprehensive evaluation and further investigation for hypophosphatasia should be considered. Persistent hypophosphatasemia of undetermined etiology could be a crucial initial step in diagnostic algorithms for this condition.

Accelerating radio astronomy imaging with RICK

This paper presents an implementation of radio astronomy imaging algorithms on modern High Performance Computing (HPC) infrastructures, exploiting distributed memory parallelism and acceleration throughout multiple GPUs. Our code, called RICK (Radio Imaging Code Kernels), is capable of performing the major steps of the w-stacking algorithm presented in Offringa et al. (2014) both inter- and intra-node, and in particular has the possibility to run entirely on the GPU memory, minimising the number of data transfers between CPU and GPU. This feature, especially among multiple GPUs, is critical given the huge sizes of radio datasets involved.After a detailed description of the new implementations of the code with respect to the first version presented in Gheller et al. (2023), we analyse the performances of the code for each step involved in its execution. We also discuss the pros and cons related to an accelerated approach to this problem and its impact on the overall behaviour of the code. Such approach to the problem results in a significant improvement in terms of runtime with respect to the CPU version of the code, as long as the amount of computational resources does not exceed the one requested by the size of the problem: the code, in fact, is now limited by the communication costs, with the computation that gets heavily reduced by the capabilities of the accelerators.

Binocular camera calibration using a BP algorithm optimized by an improved subtraction-average-based optimizer

The BP algorithm exhibits drawbacks such as reduced precision and extended iteration duration during the process of calibrating a binocular camera system. A technique employing an improved subtraction-average-based optimizer to optimize the BP algorithm (ISABO-BP) is introduced to address these limitations and complete the calibration of the binocular camera. The pixel values of the corner in 2D space and the coordinate values in 3D space are used as the input and output for the BP algorithm, respectively. First, the logistic chaotic mapping is used to initialize the population to increase the diversity of the initial population. The piecewise mapping is employed to generate random values to replace the coefficient values in the individual position update formula, making the population distribution more uniform. The golden sine strategy is adopted to help particles escape from local optima. The ISABO-BP algorithm is utilized to perform the process of calibrating a binocular camera. Then, through numerical experiments, the mean calibration precision for the BP algorithm is 0.1501 and 0.0445 cm for the ISABO-BP algorithm. The calibration precision has been enhanced by 70.4%. The iterative steps of the BP algorithm are 353 epochs, while the ISABO-BP algorithm are 76 epochs, and the iterative speed is increased by nearly four times. It is evident that the ISABO-BP algorithm enhances the calibration precision and expedites the convergence rate. Finally, other intelligent algorithms to optimize the BP algorithm for calibrating a binocular camera are compared, and the superiority of this algorithm is also verified.

GPU based discrete element modeling for convex polyhedral shape particles: Development and validation

Particle dynamics simulations face a significant challenge in understanding the intricate behaviors of convex polyhedral particles due to their complex geometries and interactions. DEM emerges as a key method, illuminating the concealed intricacies of these geometric entities. Traditional algorithms often require cumbersome processes to check each type of contact individually. However, Gilbert–Johnson–Keerthi’s (GJK) and the expanding polytope algorithm (EPA) provide efficient numerical solutions for polyhedral contact detection and contact resolution. These Minkowski difference-based methods streamline contact detection and overlap computation, paving the way for deeper exploration of three-dimensional contact theory within DEM simulations. By leveraging GPU computational power, this paper outlines key algorithmic steps and verifies the solver’s accuracy through comparison with simulated and experimental data, with an average deviation of less than 5%. This study explores the impact of particle shape on the dynamics and mechanical behavior of densely packed systems, particularly in hoppers and tumblers. Spherical particles discharge faster but mix more slowly than polyhedral shapes, with icosahedrons achieving quicker full mixing. These results align with experimental findings, further validating the simulation approach.

Hybrid Spotted Hyena based Load Balancing algorithm (HSHLB)

In this paper, we delve into the core of our proposed dynamic load balancing mechanism, the Hybrid Spotted Hyena based Load Balancing algorithm (HSHLB). We begin by presenting a comprehensive overview of the Spotted Hyena Optimization Algorithm (SHOA) and the fundamental behaviors it encompasses, namely migration and attacking. We then introduce the Load Balancing Algorithm (LB) and outline its principles, emphasizing its distinct characteristics. Recognizing the strengths and weaknesses of both SHOA and LB, we embark on a journey to hybridize these two optimization techniques, resulting in the potent HSHLB. We elucidate the intricate process of combining these algorithms, elucidating how HSHLB harnesses their respective strengths while mitigating their limitations. This hybridization is pivotal to the overarching goal of achieving dynamic load balancing within cloud computing environments. As we progress through this section, we provide invaluable insights into the inner workings of HSHLB, offering readers a comprehensive understanding of its algorithmic steps, parameters, and intricacies. For clarity and enhanced comprehension, we incorporate pseudocode or flowcharts to illustrate the practical implementation of HSHLB. In sum, Section 3 lays the foundation for the practical application of HSHLB in achieving dynamic load balancing, setting the stage for subsequent sections where we delve into implementation details, results, and analysis. HSHLB emerges as a promising solution to the multifaceted challenges of load balancing in cloud computing, leveraging the unique strengths of SHOA and LB to optimize resource allocation and enhance Quality of Service (QoS).

A variable step and multi-constraint vehicle’s trajectory generation algorithm based on deep deterministic policy gradient network

Abstract Aiming at the problem that the traditional vehicle’s trajectory generation method takes a long time and is difficult to calculate in real time, a variable step and multi-constraint trajectory generation algorithm based on a deep deterministic policy gradient (DDPG) network is proposed. Firstly, the dynamic model and constraint conditions of the vehicle are analyzed. On this basis, the reinforcement learning training model is constructed based on DDPG, and the state, action, and reward design of the training model are defined. At the same time, the variable step is introduced into the action model to improve the generated trajectory’s navigation precision. Finally, the proposed method is verified under different cases, and according to the results, the proposed method can realize quick generation of multi-constraint guidance trajectories.

Entomopathogenic nematode detection and counting model developed based on A-star algorithm

Entomopathogenic nematodes are soil-dwelling living organisms widely employed in the biological control of agricultural insect pests, serving as a significant alternative to pesticides. In laboratory procedures, the counting process remains the most common, labor-intensive, time-consuming, and approximate aspect of studies related to entomopathogenic nematodes. In this context, a novel method has been proposed for the detection and quantification of Steinernema feltiae isolate using computer vision on microscope images. The proposed method involves two primary algorithmic steps: framing and isolation. Compared to YOLO-V5m, YOLO-V7m, and YOLO-V8m, the A-star-based developed network demonstrates significantly improved detection accuracy compared to other networks. The novel method is particularly effective in facilitating the detection of overlapping nematodes. The developed algorithm excludes processes that increase space and time complexity, such as the weight document, which contains the learned parameters of the deep learning model, model integration, and prediction time, resulting in more efficient operation. The results indicate the feasibility of the proposed method for detecting and counting entomopathogenic nematodes.