Transactions on Machine Intelligence2821-16934420211201Improvement of the Performance of Cell Counter Devices Using a Proposed Model Combining Multilayer Perceptron Neural Networks18219021871310.47176/TMI.2021.182ENN.AhmadiMaster's Degree, Department of Biomedical Engineering-Biomechanics, Faculty of Mechanical Engineering, Iran University of Science and Technology, Tehran, IranS.AsyabiAssistant Professor, Department of Mechanical Engineering, Iran University of Science and Technology, Tehran, IranJournal Article20210804In automated cell counting devices, the performance of the counting channel can be significantly affected under problematic conditions, such as platelet aggregation, leading to inaccuracies in key blood parameter measurements. Given the limitations of existing algorithms in addressing these challenges, this study proposes an enhanced algorithm to improve the counting accuracy of critical blood components, particularly platelets and hematocrit, within the counting chamber. To achieve this, a hybrid approach integrating two computational models was implemented, demonstrating an improvement in overall counting performance. Among the tested optimization techniques, the Satin Bowerbird Optimization (SBO) Algorithm yielded superior results, outperforming other methods in terms of prediction accuracy. While the Biogeography-Based Optimization (BBO) Algorithm and the Teaching-Learning-Based Optimization (TLBO) Algorithm exhibited higher accuracy for certain blood parameters compared to the SBO Algorithm, the SBO Algorithm achieved the highest number of correct predictions across all parameters. In contrast, the Particle Swarm Optimization (PSO) and Firefly (FA) Algorithms failed to produce reliable results. The findings highlight the effectiveness of the proposed algorithm in enhancing the robustness and precision of blood parameter quantification, making it a promising approach for improving automated cell counting in clinical and laboratory applications.https://www.tmachineintelligence.ir/article_218713_a9e5f1a36d42c3c8800c23c5802c4fdc.pdfTransactions on Machine Intelligence2821-16934420211201A Review Study on the Use of Dynamic Complex Networks in Combating the COVID-19 Pandemic19120021871610.47176/TMI.2021.191ENA. M.KaramzadehMaster’s Student, School of Computer Engineering and Sciences, Shahid Beheshti University, Tehran, IranN.GhasriPhD Student, School of Computer Engineering and Sciences, Shahid Beheshti University, Tehran, IranJournal Article20210705The COVID-19 pandemic has profoundly affected multiple facets of society, including social interactions, financial activities, international relations, economic stability, and education. The unprecedented scale of the pandemic and the rapid transmission of the virus have necessitated multidisciplinary collaborations, bringing together researchers from diverse fields such as medicine, epidemiology, and computer science to develop predictive models and preventive strategies. Within this interdisciplinary landscape, the study of dynamic complex networks and artificial intelligence (AI) has emerged as a critical tool complementing traditional medical approaches. In the realm of computer science, network theory offers a powerful framework for modeling and simulating the spread of infectious diseases, providing valuable insights into transmission dynamics and intervention strategies. Additionally, AI-driven technological tools have been instrumental in facilitating COVID-19 control measures, including contact tracing, real-time monitoring of infected individuals, and predictive analytics for outbreak management. Many countries have leveraged network science methodologies to track viral spread and assess the broader societal impacts of the pandemic. This study systematically reviews and categorizes key research contributions in the field of dynamic complex networks applied to COVID-19. By analyzing these works, the study highlights major trends, methodological advancements, and challenges in the application of network science and AI. The findings provide a foundation for future research directions, emphasizing the potential of computational methods in enhancing pandemic preparedness and response strategies.https://www.tmachineintelligence.ir/article_218716_2712f7c68e4eef641343a39ee283416a.pdfTransactions on Machine Intelligence2821-16934420211201Design of Optimal Linear Feedback Controller for HIV Treatment20121521871710.47176/TMI.2021.201ENF.MostafaviDepartment of Electrical Engineering, Faculty of Engineering, Khomeini Shahr Branch, Islamic Azad University, Khomeini Shahr, IranH.KhodadadiAssistant Professor, Control Engineering Department, Islamic Azad University, Khomeini Shahr Branch, Khomeini Shahr, IranJournal Article20210414The design of optimal linear feedback controllers for HIV treatment is a multidisciplinary endeavor that combines principles from control theory, systems biology, and medical science. The goal is to create adaptive treatment strategies that can effectively manage the dynamics of HIV infection and optimize patient outcomes. Given the rapid advancements in control methods and the growing use of modern computers in recent years, these tools have also been applied in the field of medical processes. In this paper, after reviewing the model presented for HIV, a suitable model to describe the disease dynamics is selected. Then, using the linear feedback control method and considering the effect of antiretroviral drugs, an attempt is made to properly treat HIV. The six-variable HIV virus infection model is the basis of this study. Furthermore, considering the effect of RTIs and PIs drugs as control inputs, it is observed that the system is a nonlinear, multi-input, multi-output system. The stability of the system's internal dynamics is examined, and an optimal control method is used to optimize the dosage of injectable drugs for the patient, aiming to reduce side effects while effectively controlling the disease.https://www.tmachineintelligence.ir/article_218717_9d1a7772f929c79189a368304cf75509.pdfTransactions on Machine Intelligence2821-16934420211201Determination of Chemical Exchange Rate in Chemical Exchange Saturation Transfer (CEST) Phenomenon in Magnetic Resonance Imaging through Analytical Solution of Bloch-McConnell Equations21622521871810.47176/TMI.2021.216ENM. R.RezaeianAssistant Professor, Department of Biomedical Engineering, Hamedan University of Technology, Hamedan, IranJournal Article20210703Magnetic Resonance Imaging (MRI), by enabling the non-invasive measurement of certain physiological markers, facilitates the study and tracking of molecular states, ultimately leading to the early diagnosis of diseases. Chemical Exchange Saturation Transfer (CEST) serves as a novel contrast mechanism in MRI for molecular studies. This contrast depends on multiple parameters, including relaxation times, the chemical exchange rate between water molecules and the contrast agent, the concentration of the contrast agent, and the properties of the applied radiofrequency (RF) pulse. The chemical exchange rate is a crucial parameter, as it correlates with various clinical indicators such as pH, temperature, and metabolite concentration. However, its direct measurement remains challenging. In this study, a method for determining this rate using RF pulse width is proposed. First, the CEST effect is expressed in an analytical relationship, demonstrating that this contrast reaches its maximum at a specific RF pulse width. This analytical expression distinguishes the CEST effect from the magnetic transfer effects caused by macromolecules in biological tissues, which act as confounding factors. Assuming a known contrast concentration, the chemical exchange rate can be determined through an analytical relationship based on the RF pulse width that maximizes the CEST effect. The validity of this analytical relationship is confirmed by comparison with widely accepted definitions of the CEST effect. Furthermore, by applying a Taylor series expansion to the analytical expression, relevant formulas from reputable publications are derived. The study employs validated data from a three-pool structure, which is consistent with biological tissue models referenced in authoritative sources. It is recommended that the findings of this study be practically implemented on MRI scanners.https://www.tmachineintelligence.ir/article_218718_47f3b83bc8dcaf6412f99bf87005c369.pdfTransactions on Machine Intelligence2821-16934420211201Analysis of the Functionality of New Techniques in Ultrasonic Systems and the Use of Array Transducers for Parallel Processing and Real-time 3D Imaging and Related Applications22623721872010.47176/TMI.2021.226ENJ.BagheriAssistant Professor, Faculty of Electrical and Computer Engineering, Shahid Chamran University of Ahvaz, IranF.AhmadiMaster's Degree in Electrical Engineering - Control, Faculty of Engineering, Islamic Azad University, Science and Research Branch, Tehran. IranE.Asad SamaniGraduate in Mechanical Engineering, Faculty of Engineering, Shahrekord University, Shahrekord. IranJournal Article20210522The optimal use of the functionalities of ultrasonic waves and the rational utilization of the capabilities of various multidimensional ultrasonic parameters have presented a unique ability in the fields and techniques related to the measurement of various physical quantities. Different types of probes, such as linear and phased arrays, are considered indispensable components in most ultrasonic imaging systems for performing many imaging processes and operations. The use of two-dimensional ultrasonic arrays is one of the solutions to prevent a decrease in contrast levels. A limiting factor in multidimensional arrays is the additional time required for data acquisition and signal processing. This paper, while refining and explaining the above-mentioned issues, analyzes the "functionality of new techniques in ultrasonic systems and the use of array transducers for parallel processing and real-time 3D imaging and related applications." In this context, other effective approaches, such as parallel processing, types and advantages of multidimensional (two-dimensional) arrays for achieving real-time 3D imaging, micro-machined ultrasonic transducers (cMUT), pyramid scanning, and straight-line 3D off-line imaging, are also analyzed in this paper.https://www.tmachineintelligence.ir/article_218720_628f5067e0954f279ce6e0ccc3bef30e.pdfTransactions on Machine Intelligence2821-16934420211201Finding the Potential Accepted Answer on Stack Overflow: a Text Mining Approach23824421872110.47176/TMI.2021.238ENM.Jamshidiyan TehraniFaculty of Informatics, Università della Svizzera Italiana, Lugano, SwitzerlandP.ArjomandDepartment of Computer Engineering, Salman Farsi University of Kazerun, Taleghani, Kazerun, 73175-457, Fars, IranS.HaghighatDepartment of Computer Engineering, Salman Farsi University of Kazerun, Taleghani, Kazerun, 73175-457, Fars, IranJournal Article20210617Stack Overflow serves as a widely-used, community-driven platform where developers seek assistance with programming-related issues. While the platform allows users to post questions and receive multiple answers, a significant portion of these questions do not culminate in an accepted solution. This lack of a clearly identified best answer often results in confusion for both the original poster and future visitors, as well as increased time spent navigating through numerous responses. To address this challenge, we present a method for automatically identifying the most promising answer among unaccepted ones. Our approach involves the application of text mining techniques to extract 13 informative features from a large dataset comprising 15,464 questions, 37,275 answers, and 72,025 comments. These features capture various textual, structural, and user-related aspects of the posts. The extracted data are then used to train machine learning models aimed at predicting the answer most likely to be accepted. The study focuses solely on English-language content available on Stack Overflow. The proposed method demonstrates promising performance, achieving an overall accuracy of 71% and an F1 score of 70%. These results suggest that automated answer recommendation can significantly enhance the user experience by reducing ambiguity and improving the efficiency of information retrieval on Q&A platforms.https://www.tmachineintelligence.ir/article_218721_f0b67ac12ed68f09c05c618d71dd992d.pdf