مجله رایانش نرم و فناوری اطلاعات
سال یازدهم شماره 4 (زمستان 1401)

On the one hand, the development of modern medicine has made it possible to store medical images, and on the other hand, due to the daily increase in the storage of such data, it has also made their management and recovery difficult. Considering that medical images are used as a powerful tool in the early diagnosis of most diseases, providing a powerful system that can retrieve images with similar content from the growing volume of medical images is very effective in control and treatment. In this article, a medical image retrieval system based on Siamese neural network consisting of two convolutional sub-networks with 13 layers is presented. To reach the optimal subset of deep features extracted by Siamese, the Minimum Redundancy-Maximum Relevant (mRMR) technique has been used, and after binary hashing of the features, similar images are retrieved using Hamming distance. Although the proposed model is capable of retrieving a variety of gray scale medical images, two types of lung images have been used to evaluate it, including CT scan images of Covid-19 patients in the CT-COV database and X-ray images of pneumonia patients in the Pneumonia database. The results indicate that the proposed method in the Covid database has been able to achieve an average precision of 93.83% and 92.73% in 5 and 10 retrieved images respectively, and an average precision of 100% in the pneumonia database, which is compared to previous methods have been able to improve the retrieval of lung images.

NoSQL systems are suitable solutions for big data projects and offer a high level of flexibility in design. A good and efficient schema design for NoSQL wide column stores is not only based on the application’s conceptual data model but also on the queries defined in an application’s workload. In these databases, a manual schema design relies on rules of thumb to choose a good schema. Utility these rules without practices is a big challenge in this area. Because these rules are vague and generic, and must be adapted to each application. One of the ways that researchers use to overcome this challenge is presenting automated schema design. The main contribution of this paper is automated schema design for NoSQL wide column stores. This research proposes a workload-driven approach for the mapping from the application conceptual data model to a database schema with a goal of optimizing query performance. This approach uses workload information to achieve a good workload performance, which result in an optimized schema by minimizing the number of requests to the wide column stores. The experimental results show that automated schema generated by the proposed approach leads to a good workload performance.

City Councils Evolution algorithm (CCE) is a metaheuristic algorithm inspired by the formation process of the supreme council of a city due to the formation of councils from the smallest neighborhoods to the largest regions. In this paper, we want to improve the performance of CCE by applying two important changes.The first change is about the continuous reduction of the population size using Linear Population Size Reduction (LPSR) technique. In this technique, the population size in the initial iterations is considered large enough such that it can explore wide areas of search space. As the algorithm progresses, the population size is gradually reduced to increase the convergence speed. The second change is related to the domain of variables, which is constantly reduced to limit the search space, and so the possibility of finding optimal solutions is increased. To evaluate and compare the performance of ICCE with CCE, Chimp Optimization, Black Widow Optimization, Political Optimizer, Barnacles Mating Optimizer, Snake Optimizer, and Aquila Optimizer, we implement them on 29 test functions from 2017 IEEE Congress on Evolutionary Computation (CEC 2017). The results of Friedman mean rank and Wilcoxon signed-rank tests confirm the higher performance of ICCE compared to other algorithms

Today, as the new generation of communication networks is implemented, we are witnessing a considerable change in IoT development and new programs in this context. Despite recent advancements in mobile networks and devices, the limitations of devices connected to this platform in terms of computational power and energy have resulted in sever challenges for running resource-intensive programs with exigent latency requirements. To address these challenges, the concept of computation offloading in Multi-access Edge Computing (MEC) has been recently developed, in which storage and computation resources are provided close to the user. However, due to the user mobility and changes in the profile of offloaded applications over time, the problem of assignment of edge servers to users with the aim of minimizing the overall offloading latency is a complicated task. In this regard, existing mobility-aware offloading approaches are not based on fine-grain offloading and use random and unrealistic mobility models. In this article, to address the aforementioned challenges, we propose a mobility-aware fine-grain computation offloading method to minimize the overall offloading delay. In the proposed approach, the user application is divided into several components and the offloading decision is made for each component according to the mobility and specifications of user components during the time slots defined in the system. In oner hand, this latter results in more efficient offloading decision. In the other hand, it reduces the overhead of migration since the migration of a subset of program’s components imposes lower cost compared to the migration of the entire program. Moreover, we use user profile and location prediction to optimize the offloading decisions considering the underlying context over time. According to the evaluation results, it is observed that the proposed method achieves significantly better performance compared to other alternatives while the complexity of offloading decision is kept very low.

Due to the increasing use of Closed-Loop Supply Chains (CLSCs) in industry scopes in order to optimize raw material usage and reuse of end-of-life or used products, CLSC management is very necessary. Used products return to production cycles as production resources in CLSC processes. Thus, CLSC helps countries to use fewer raw materials for manufacturing and brings an opportunity for manufacturers to reuse end-of-life products, so that they can increase their overcome and decrease the use of natural resources simultaneously. Due to the impact of CLSC on industries, CLSC control is an interesting field of research. In this paper, a new game theory-based architecture for controlling CLSC considering uncertainty will be introduced. For this purpose, a smart Proportional-Integral-Derivative (PID) controller will be used which automatically tunes itself using the game theory method. Thus, creating a game with controller gains and searching for the equilibria point of the game, makes the best state for the controller. So that the controller could act rationally in CLSC and tune itself to optimize its performance. The optimized reaction of the controller used in this model would help the CLSC management system.

Breast cancer is one of the most common causes of death among women around the world, but early and accurate diagnosis of this type of cancer can dramatically improve treatment. Thermal imaging is one of the primary methods of diagnosing breast cancer. The computer diagnosis system can also be used to help physicians to increase the accuracy of interpretation of results. This paper presents an intelligent computer diagnostic system for the detection of breast cancer using thermal imaging. The proposed intelligent computer diagnosis system includes SFTA method for feature extraction and SVM, kNN and D-Tree algorithms for classification of results. The performance of the proposed intelligent computer diagnosis system is evaluated using the DMR-IR and Fluminense Federal University databases and MATLAB2018, when using the cuckoo feature selection algorithm and without using the feature selection algorithm. The results show that the average accuracy, sensitivity and specificity are 99%, 99.5% and 98.03%, respectively, using the cuckoo feature selection algorithm and SVM classification algorithm. Also, the presented computer diagnostic system has advantages compared to other computer diagnosis systems. These results indicate that the use of SFTA feature extraction method, cuckoo feature selection algorithm, SVM classification algorithm and DMR-IR database in the proposed computer diagnosis system can improve the evaluation results.

Today, internet of things (IoT) is widely used in various fields, such as health control, smart city, smart buildings, and so on. One of the severe concerns in IoT-based systems is the issue of energy consumption and its management. The IoT-based systems have limited resources. It is also difficult to access them in many situations; so, their energy sources must be appropriately consumed and managed. In order to design and build IoT systems, many factors, such as the usable chips, the type of communication protocols, the timing of data transfer, and so on, directly affect the system's energy consumption. Therefore, in the design phase of these systems, it is necessary to model and evaluate their energy consumption and their compliance with the requirements. In this paper, a stochastic reward nets (SRN) model is introduced for modeling and quantitative evaluation of energy consumption in IoT-based systems. In this method, first the modeling of the energy source and the influencing factors are explained, then the entire system is modeled using these basic models, Then, by changing the configuration of the system, the influencing factors on energy consumption are studied and measured, and the most suitable configuration is selected using the results. This method can lead to a significant reduction in the costs of a system before its operation. Using this model, a case study has been investigated and the effects of various factors, such as the number of nodes connected to the cluster head, the existence of sleep/wake mode, and different transmission rates on system performance and energy consumption are investigated. In this paper the effect of the system configuration on the amount of energy consumption has been evaluated and it has been shown that the packet delay increases with the increase in the number of nodes connected to the cluster head, and in order to reduce the number of packets in the buffer by 75% and reduce the end-to-end delay of the packets, a transmitter with a transmission rate of 0.037 should be used.