ITIL stands for Information Technology Infrastructure Library. ITIL is the most famous framework for managing IT services. ITIL provides guidelines for the implementation of Service Desk solutions in the business environment of any company. The original scientific and professional contribution of this work is proof that the ITIL framework, through the implementation of a software solution that automates its business processes, can help every company in daily work and can improve its basic and additional business processes. Measurements will be made in a real public company in Bosnia and Herzegovina. Also, the software solution itself will be independently developed with its own original programming code only for the purposes of this research.

Fully automated chatbots are increasingly being applied in the real-estate industry. Although, they are not completely able replace interaction between the real-estate agents, they can automate customer support, save human resources for qualitative tasks, accelerate operations, and improve business branding. In this paper, a chatbot for real-estate is developed. The chatbot is able to engage clients in meaningful conversations in real-time. It provides a 24/7 service and effectively reduces administrative costs. The architecture and infrastructure overview are presented. The rule matching algorithm is presented and discussed in detail.

In the field of telecommunications and cloud communications, accurately and in real-time detecting whether a human or an answering machine has answered an outbound call is of paramount importance. This problem is of particular significance during campaigns as it enhances service quality, efficiency and cost reduction through precise caller identification. Despite the significance of the field, it remains inadequately explored in the existing literature. This paper presents an innovative approach to answering machine detection that leverages transfer learning through the YAMNet model for feature extraction. The YAMNet architecture facilitates the training of a recurrent-based classifier, enabling real-time processing of audio streams, as opposed to fixed-length recordings. The results demonstrate an accuracy of over 96% on the test set. Furthermore, we conduct an in-depth analysis of misclassified samples and reveal that an accuracy exceeding 98% can be achieved with the integration of a silence detection algorithm, such as the one provided by FFmpeg.

Many deep-learning computer vision systems analyse objects not previously observed by the system. However, such tasks can be simplified if the objects are marked beforehand. A straightforward method for marking is printing 2D symbols and attaching them to the objects. Selecting these symbols can affect the performance of the CV system, as similar symbols may require extended training time and a larger training dataset. It is possible to find good symbols differentiated by the given neural network easily. Still, there were no efforts to generalise such findings in the literature, and it is not known if the symbols optimal for one network would work just as well in the other. We explored how transferable symbol selection is between the networks. To this end, 30 sets of randomly selected and augmented symbols were classified by-five neural networks. Each network was given the same training dataset and the same training time. Results were ranked and compared, which allowed the identification of networks which performed similarly so that the symbol selection could be generalised between them.

Deep learning techniques in computer vision (CV) tasks such as object detection, classification, and tracking can be facilitated using predefined markers on those objects. Selecting markers is an objective that can potentially affect the performance of the algorithms used for tracking as the algorithm might swap similar markers more frequently and, therefore, require more training data and training time. Still, the issue of marker selection has not been explored in the literature and seems to be glossed over throughout the process of designing CV solutions. This research considered the effects of symbol selection for 2D-printed markers on the neural network’s performance. The study assessed over 250 ALT code symbols readily available on most consumer PCs and provided a go-to selection for effectively tracking n-objects. To this end, a neural network was trained to classify all the symbols and their augmentations, after which the confusion matrix was analysed to extract the symbols that the network distinguished the most. The results showed that selecting symbols in this way performed better than the random selection and the selection of common symbols. Furthermore, the methodology presented in this paper can easily be applied to a different set of symbols and different neural network architectures.

In this paper, the multilevel image thresholding methods based on the particle swarm optimization algorithm and different chaotic inertia weight strategies are considered. The performance of each chaotic inertia weight strategy is evaluated using a set of standard test images. Different numbers of image classes are considered. In addition, the paper also considers the multilevel thresholding performance based on commonly employed linear decreasing inertia weight and random inertia weight. All considered multilevel thresholding methods are based on Kapur’s entropy. The experimental results demonstrate that the particle swarm optimization with chaotic inertia weight can be successfully used for multilevel image thresholding.

In this paper, a multilevel thresholding method for image segmentation based on Otsu’s between-class variance and multi-swarm particle swarm optimization algorithm with dynamic learning strategy is presented. The considered multilevel image thresholding method is assessed on various standard test images and for different numbers of thresholds. For each test image and a considered number of thresholds, the mean and the standard deviation of Otsu’s objective function over a number of independent runs are evaluated. The experimental results showcased that this method can be successfully employed in multilevel image thresholding.

This study scrutinizes five years of Sarajevo’s Air Quality Index (AQI) data using diverse machine learning models — Fourier autoregressive integrated moving average (Fourier ARIMA), Prophet, and Long short-term memory (LSTM)—to forecast AQI levels. Focusing on various prediction frames, we evaluate model performances and identify optimal strategies for different temporal granularities. Our research unveils subtle insights into each model’s efficacy, shedding light on their strengths and limitations in predicting AQI across varied timeframes. This research presents a robust framework for automatic optimization of AQI predictions, emphasizing the influence of temporal granularity on prediction accuracy, automatically selecting the most efficient models and parameters. These insights hold significant implications for data-driven decision-making in urban air quality control, paving the way for proactive and targeted interventions to improve air quality in Sarajevo and similar urban environments.

Audio fingerprinting techniques have seen great advances in recent years, enabling accurate and fast audio retrieval even in conditions when the queried audio sample has been highly deteriorated or recorded in noisy conditions. Expectedly, most of the existing work is centered around music, with popular music identification services such as Apple’s Shazam or Google’s Now Playing designed for individual audio recognition on mobile devices. However, the spectral content of speech differs from that of music, necessitating modifications to current audio fingerprinting approaches. This paper offers fresh insights into adapting existing techniques to address the specialized challenge of speech retrieval in telecommunications and cloud communications platforms. The focus is on achieving rapid and accurate audio retrieval in batch processing instead of facilitating single requests, typically on a centralized server. Moreover, the paper demonstrates how this approach can be utilized to support audio clustering based on speech transcripts without undergoing actual speech-to-text conversion. This optimization enables significantly faster processing without the need for GPU computing, a requirement for real-time operation that is typically associated with state-of-the-art speech-to-text tools.

Epilepsy represents a neurological disorder of the brain characterized by repeated seizures. These are sudden abnormality in the brain’s electrical activities that temporarily affect normal brain function. Electroencephalogram (EEG) is one of the main diagnostic tools for monitoring the brain activity of patients with epilepsy. Typically, the detection of epileptic activity is carried out by an expert by analyzing the EEG recordings, but this is a difficult, error prone and time-consuming task. In order to get timely and accurate automatic detection of seizure, various approaches based on both conventional and deep learning techniques were proposed in the literature. The aim of this paper is to present a framework for the automatic detection of epileptic seizure based on the functional connectivity matrix obtained from EEG signals and deep learning. Convolutional neural networks (CNN) were employed because of their capability to learn patterns of neural activities based on brain connectivity represented by connectivity matrix. Obtained results are very promising indicating a potential of this approach as an efficient tool for automated seizure detection based on EEG data.

Corticosteroids regulate a number of physiological processes and are synthetic analogs of the natural steroid hormones produced by the adrenal cortex. As drugs, corticosteroids are non-inflammatory and are used for the treatment of plethora of conditions which include arthritis, kidney, skin, lungs or thyroid disorders, for the treatment and relief of symptoms of allergies and symptoms of some gastrointestinal disorders. In addition, glucocorticoids can regulate the effects of inflammatory disorders, including sepsis, autoimmune diseases, and allergies. These conditions are potentially fatal. Consequently, this drug class is among the most commonly prescribed globally. One representative of corticosteroid class of drugs is dexamethasone which is used to treat allergies, adrenal problems, arthritis, asthma, diseases of blood or bone marrow, inflammation, kidney diseases, different types of skin conditions, and episodes of multiple sclerosis. Virulence factors help bacteria colonize the host at the level of the cell. In their nature, these factors are secretory, associated with the membrane or present in the cytosol. Secretory factors allow bacterium to circumvent the host immune response, while membrane factors aid bacterium in adhesion to the host cell. Finally, cytosol factors help bacteria adapt metabolically, physiologically, and morphologically to their changing environment. One such factor is aspartyl proteinase, a protein that degrades other proteins and is a virulence factor in many pathogens playing a role in the host invasion process. Another important virulence factor is the ability to form biofilms, which can render bacteria resistant to antimicrobials. Despite the widespread use of corticosteroids, including dexamethasone, little is known about their possible influence on the expression of virulence factors such as aspartyl proteinase. If such a connection is to exist the use of corticosteroids could elicit pathogenesis in certain microbes. In the here-presented study we wanted to investigate the effects of dexamethasone on the growth, expression of aspartyl proteinase and biofilm formation in three E. coli strains that were previously isolated from patients suffering from urinary tract infection. To this aim, we amended the growth media with 0.5 mg/mL dexamethasone. Bacterial growth was measured over the period of 24 hours and the effect of dexamethasone was established at different time points. Administration of 0.5 mg/mL glucocorticoid drug dexamethasone did not significantly affect bacterial growth. However, it resulted in an increase in concentration of secreted E. coli virulence factor aspartyl proteinase, which increased up to 2.6-fold for some E. coli strains. In addition, we noted the increased biofilm formation in to three out of four studied strains. This study indicates dexamethasone as a possible trigger molecule for the expression of virulence factor aspartyl proteinase in E. coli.

The segmentation of pediatric brain MRI into distinct tissues is important for the evaluation of pediatric brain development and the diagnosis of neurological and neurodevelopmental disorders. However, when the used dataset diverges due to various acquisition protocols or biases among patient cohorts, existing deep learning algorithms cannot guarantee correct predictions. Unsupervised domain adaptation approaches have lately shown enormous potential for addressing this problem by limiting the divergence between the distributions of the used datasets. In this paper, we firstly developed a model called 3DUDRSeg, a 3D encoder-decoder for precise autonomous segmentation of pediatric brain tissues. Our proposed 3DUDRSeg model achieved a 98.88% DSC accuracy rate because of the employment of denes blocks and residual units that help relieve the degradation problem during training the network, allowing the performance advantages to be fully utilized. With this approach, our 3DUDRSeg can create more strong features to deal with the wide range of brain tissue variations. Then, we present 3DAdGanSeg, an entropy-based unsupervised domain adaptation framework for segmenting pediatric brain tissues in unannotated datasets via adversarial learning. The suggested model significantly influences the capability to distinguish the borders between tissue classes, with DSC of 85% and HD95 of 1.479 in the case of the dHCP dataset as the source domain and DSC of 81 % and HD95 of 2.061 when using the Schizophrenia Bulletin 2008 dataset as a source domain.

Over the last few years, remarkable progress has been made in the field of congenital heart diseases. Improvements considering diagnostic modalities, especially imaging, in surgical and interventional techniques, as well as in postoperative therapy and care, have contributed to a significant reduction in mortality and morbidity. One of the most important applications of medical imaging techniques in children is the detection and treatment of congenital heart anomalies. Objective of this article is to show the importance of ultrasonography in the detection of congenital heart diseases in children. The study was conducted on children with simple and complex congenital heart diseases and was conducted on the Pediatric Clinic, UKCS. The research is descriptive on a representative sample. In our study, 166 children were observed, of which 148 children (77 boys, 71 girls) with simple congenital heart diseases, and 18 children (8 boys, 10 girls) with complex congenital heart diseases. Out of the total number of observed children, 115 had a surgical correction, 97 children with simple congenital heart diseases (45 boys, 52 girls) and 18 children with complex congenital heart diseases (8 boys, 10 girls). The number of children monitored through the Cardiac Counseling Center who didn’t undergo surgical correction was 51, all with simple congenital heart diseases. Out of the total number of observed children who were frequently coming for follow-ups, 28 children had changes on the ECG, and 138 of them had no changes on the ECG, 93 were surgically treated, and 73 of them were on conservative therapy. Based on the results of the research, we conclude that ultrasonography is an important method in the detection and treatment of congenital heart diseases.

A technical issue with fluid flow heating is the relatively small temperature increase as the fluid passes through the heating surface. The fluid does not spend enough time inside the heating source to significantly raise its temperature, despite the heating source itself experiencing a substantial increase. To address this challenge, the concept of the multiple circular heating of air was developed, forming the basis of this work. Two PTC heaters with longitudinal fins are located within a closed channel inside housing composed of a thermal insulation material. Air flows circularly from one finned surface to another. Analytical modeling and experimental testing were used in the analysis, with established restrictions and boundary conditions. An important outcome of the analysis was the methodology established for the optimization of the geometric and process parameters based on minimizing the transient thermal entropy. In conducting the analytical modeling, the temperature of the PTC heater was assumed to be constant at 150 °C and 200 °C. By removing the restrictions and adjusting the boundary conditions, the established methodology for the analysis and optimization of various thermally transient industrial processes can be applied more widely. The experimental determination of the transient thermal entropy was performed at a much higher air flow rate of 0.005 m3s−1 inside the closed channel. The minimum transient entropy also indicates the optimal time for the opening of the channel, allowing the heated air to exit. The novelty of this work lies in the controlled circular heating of the fluid and the establishment of the minimum transient thermal entropy as an optimization criterion.