In this paper, we introduce and provide insight into the two innovative applications designed to enhance the lives of persons with Down syndrome, focusing on seamless integration between the two. The first is a mobile application that helps users manage their daily routines by monitoring and predicting activity durations, considering their unique challenges. The second is a web application for parents/teachers/other adults to streamline activity scheduling, progress tracking, and reminders.

In this paper, the multilevel thresholding method based on Kapur’s entropy and the recently proposed multi-swarm particle swarm optimization with a dynamic learning strategy is considered. The multilevel thresholding method is extensively evaluated on ten benchmark images. The experimental results, which include the mean and standard deviation of Kapur’s entropy, obtained from forty independent executions of the thresholding method for each test image and the considered total number of thresholds, demonstrate that multi-swarm particle swarm optimization with dynamic learning strategy can be successfully applied to solve the multilevel thresholding problem.

This paper presents two color image quantization methods, namely RKI-CIQ and RK2-CIQ. These are population-based methods that use the k-means algorithm. Both color quantization methods require only a few control parameters. In this paper, a comparative analysis of the two color image quantization methods is presented. The experimental analysis is based on four test images and different color palette sizes. The obtained results demonstrate the successful application of these color quantization methods.

In this paper, a comparative analysis of different methods for magnetic induction estimation in the vicinity of overhead power lines is presented. The methods for determining magnetic induction, considered in the paper, include the recently proposed artificial neural network based method and the Biot-Savart law based method. In addition, the paper considers a method that employs the genetic algorithm to fit a considered mathematical model to the field measurements. The performance of various methods is evaluated on an actual 400 kV overhead power line. The method based on the artificial neural networks is able to accurately evaluate magnetic induction values along the lateral profile without relying on field measurements using only the description of power line conductor configuration and the current intensity value.

The identification of bacterial colonies is deemed to be crucial in microbiology as it helps in identifying specific categories of bacteria. The careful examination of colony morphology plays a crucial role in microbiology laboratories for the identification of microorganisms. Quantifying bacterial colonies on culture plates is a necessary task in Clinical Microbiology Laboratories, but it can be time‐consuming and susceptible to inaccuracies. Therefore, there is a need to develop an automated system that is both dependable and cost‐effective. Advancements in Deep Learning have played a crucial role in improving processes by providing maximum accuracy with a negligible amount of error. This research proposes an automated technique to extract the bacterial colonies using SegNet, a semantic segmentation network. The segmented colonies are then counted with the assistance of blob counter to accomplish the activity of colony counting. Furthermore, to ameliorate the proficiency of the segmentation network, the network weights are optimized using a swarm optimizer. The proposed methodology is both cost‐effective and time‐efficient, while also providing better accuracy and precise colony counts, ensuring the elimination of human errors involved in traditional colony counting techniques. The investigative assessments were carried out on three distinct sets of data: Microorganism, DIBaS, and tailored datasets. The results obtained from these assessments revealed that the suggested framework attained an accuracy rate of 88.32%, surpassing other conventional methodologies with the utilization of an optimizer.

The visual layout has an enormous influence on human perception and is a subject of many studies, including research on web page similarity comparison. Structure-based approaches use the possibility of direct access to HTML content, whereas visual methods have widespread usage due to the ability to analyze image screenshots of entire web pages. A solution described within this paper will focus on extracting web page layout in forms needed by both above-mentioned approaches.

This article presents a simple software-developed model for calculating the relative frequency of individual symbols and the entropy of the Latin alphabet of a standardised language used by four South-Slavic origin ethnic groups in the Western Balkans in four countries. In addition, a method of applying the Shannon-Fano and Huffman source coding algorithms is presented, which takes into consideration the specificity of the observed alphabet in relation to the English one. The presented model is developed in the MATLAB programming language. The model is tested using an arbitrarily selected text.

This paper presents a model that enables the application of smart waste collection management using artificial intelligence to detect QR-codes on the video stream of surveillance cameras attached to waste collection trucks. A framework model proposal together with a detailed explanation of the key components of the system is shown. It also demonstrates the use of QR-code detection for identification of waste bins and its specific application in smart waste management system.

Objective. This paper presents a novel domain adaptation (DA) framework to enhance the accuracy of electroencephalography (EEG)-based auditory attention classification, specifically for classifying the direction (left or right) of attended speech. The framework aims to improve the performances for subjects with initially low classification accuracy, overcoming challenges posed by instrumental and human factors. Limited dataset size, variations in EEG data quality due to factors such as noise, electrode misplacement or subjects, and the need for generalization across different trials, conditions and subjects necessitate the use of DA methods. By leveraging DA methods, the framework can learn from one EEG dataset and adapt to another, potentially resulting in more reliable and robust classification models. Approach. This paper focuses on investigating a DA method, based on parallel transport, for addressing the auditory attention classification problem. The EEG data utilized in this study originates from an experiment where subjects were instructed to selectively attend to one of the two spatially separated voices presented simultaneously. Main results. Significant improvement in classification accuracy was observed when poor data from one subject was transported to the domain of good data from different subjects, as compared to the baseline. The mean classification accuracy for subjects with poor data increased from 45.84% to 67.92%. Specifically, the highest achieved classification accuracy from one subject reached 83.33%, a substantial increase from the baseline accuracy of 43.33%. Significance. The findings of our study demonstrate the improved classification performances achieved through the implementation of DA methods. This brings us a step closer to leveraging EEG in neuro-steered hearing devices.

The use of augmented reality (AR) allows for the integration of digital information onto our perception of the physical world. In this article, we present a comprehensive review of previously published literature on the implementation of AR in physics education, at the school and the university level. Our review includes an analysis of 96 papers from the Scopus and Eric databases, all of which were published between 1st January 2012 and 1st January 2023. We evaluated how AR has been used for facilitating learning about physics. Potential AR-based learning activities for different physics topics have been summarized and opportunities, as well as challenges associated with AR-based learning of physics have been reported. It has been shown that AR technologies may facilitate physics learning by providing complementary visualizations, optimizing cognitive load, allowing for haptic learning, reducing task completion time and promoting collaborative inquiry. The potential disadvantages of using AR in physics teaching are mainly related to the shortcomings of software and hardware technologies (e.g. camera freeze, visualization delay) and extraneous cognitive load (e.g. paying more attention to secondary details than to constructing target knowledge).

Wireless sensor networks play a key role in various applications such as environmental monitoring, smart cities and medical monitoring. Latency and reliability are two basic aspects that affect the efficiency and quality of service of these networks. In this research, delay and reliability analyzes of WSN were conducted through the application of mathematical models. Different parameters were analyzed such as distance, number of devices, and bandwidth to gain a deeper understanding of their impact on network performance. Mathematical models have been developed that take into account random variables and changing factors in order to conduct a more precise analysis. Through simulations and numerical experiments, Matlab codes will be used to simulate and analyze the actual process, all with the aim of achieving minimum delay and maximum reliability. This research provides a deeper insight into the characteristics of WSN and provides guidelines for the design and optimization of these networks.

The deep operator network (DeepONet) structure has shown great potential in approximating complex solution operators with low generalization errors. Recently, a sequential DeepONet (S-DeepONet) was proposed to use sequential learning models in the branch of DeepONet to predict final solutions given time-dependent inputs. In the current work, the S-DeepONet architecture is extended by modifying the information combination mechanism between the branch and trunk networks to simultaneously predict vector solutions with multiple components at multiple time steps of the evolution history, which is the first in the literature using DeepONets. Two example problems, one on transient fluid flow and the other on path-dependent plastic loading, were shown to demonstrate the capabilities of the model to handle different physics problems. The use of a trained S-DeepONet model in inverse parameter identification via the genetic algorithm is shown to demonstrate the application of the model. In almost all cases, the trained model achieved an R2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R^2$$\end{document} value of above 0.99 and a relative L2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_2$$\end{document} error of less than 10% with only 3200 training data points, indicating superior accuracy. The vector S-DeepONet model, having only 0.4% more parameters than a scalar model, can predict two output components simultaneously at an accuracy similar to the two independently trained scalar models with a 20.8% faster training time. The S-DeepONet inference is at least three orders of magnitude faster than direct numerical simulations, and inverse parameter identifications using the trained model are highly efficient and accurate.

Objectives This study aims to present an overview of the formal recognition of COVID-19 as occupational disease (OD) or injury (OI) across Europe. Methods A COVID-19 questionnaire was designed by a task group within COST-funded OMEGA-NET and sent to occupational health experts of 37 countries in WHO European region, with a last update in April 2022. Results The questionnaire was filled out by experts from 35 countries. There are large differences between national systems regarding the recognition of OD and OI: 40% of countries have a list system, 57% a mixed system and one country an open system. In most countries, COVID-19 can be recognised as an OD (57%). In four countries, COVID-19 can be recognised as OI (11%) and in seven countries as either OD or OI (20%). In two countries, there is no recognition possible to date. Thirty-two countries (91%) recognise COVID-19 as OD/OI among healthcare workers. Working in certain jobs is considered proof of occupational exposure in 25 countries, contact with a colleague with confirmed infection in 19 countries, and contact with clients with confirmed infection in 21 countries. In most countries (57%), a positive PCR test is considered proof of disease. The three most common compensation benefits for COVID-19 as OI/OD are disability pension, treatment and rehabilitation. Long COVID is included in 26 countries. Conclusions COVID-19 can be recognised as OD or OI in 94% of the European countries completing this survey, across different social security and embedded occupational health systems.

Renal cell carcinoma (RCC) represents around 3% of all cancers, with the most frequent histological types being clear-cell RCC (ccRCC), followed by papillary (pRCC) and chromophobe (chRCC). Hypoxia-inducible factors (HIFs), which promote the expression of various target genes, including vascular endothelial growth factor (VEGF) and the high- affinity glucose transporter 1, have an important role in the pathogenesis of RCC. This study investigated the immunohistochemical expression of HIF-1α and VEGF-A, showing significantly higher HIF-1α nuclear expression in pRCC compared to ccRCC, while there was no significant difference in VEGF-A protein expression between the analyzed histological RCC subtypes. The quantitative reverse transcription polymerase chain reaction for HIF1A showed no statistical difference between histological types. Data from publicly available RNA sequencing databases were analyzed and showed that, compared to healthy kidney tissue, VEGFA was significantly up-regulated in ccRCC and significantly down-regulated in pRCC. The comparison between histological subtypes of RCC revealed that VEGFA was significantly up-regulated in ccRCC compared to both pRCC and chRCC. There was no statistically significant difference in survival time between HIF1A high- and low-expression groups of patients. As for VEGFA expression, pRCC patients with low expression had a significantly higher survival rate compared to patients with high VEGFA expression.

This paper explores the importance and role of soft skills in the context of education and career development. Education, as a key stage in the process of accumulation of knowledge and skills, constitutes a fundamental experience in the formation of individual competencies. In most of their lives, individuals devote considerable time and intellectual energy to the process of education, within which they acquire various forms of knowledge and skills. The acquired knowledge and skills then become key elements that shape their future professional career. But what separates highly successful individuals from the rest is often not the technical competencies acquired through formal education, but the so-called “soft skills”. Research on this topic becomes imperative in order to better understand the impact of soft skills on the performance of individuals in order to develop strategies to improve these skills. This paper recognizes the key role of soft skills in education and encourages further research to better understand their importance and to ensure that these skills are comprehensively incorporated into educational and professional programs. The research was conducted on a sample of students from the technical faculties of “Dzemal Bijedic” University of Mostar.