Abstract This research includes all banks in Bosnia and Herzegovina (B&H) and testing internal and external variables on bank profitability indicators. The primary goal of this paper is to determine, through correlation and regression analysis, the strength and significance of the external and internal variables on bank profitability in Bosnia and Herzegovina. Likewise, data were collected from quarterly reports of the Banking Agency of the Federation of B&H and the Banking Agency of the Republika Srpska for the period 2008 Q1 to 2019 Q4. The following dependent variables were used: ROA, ROE and independent variables: GRNGL, GRNPL, GRGDP, concentration ratio of loans of the largest banks in the system (CR Loans), concentration ratio of deposits of the largest banks in the system (CR Deposits), CAR and loan-to-deposit ratio. The study found that there is a significant statistical impact of the variables on ROA and ROE. In addition, this study points out the need for banks to properly select debtors, and control costs, toxic loans and provisions in order to increase profits and reduce costs.

Introduction: Multiple sclerosis (MS) is a chronic, inflammatory, (auto) immune disease of the central nervous system (CNS). Cognitive disorders are found in over 50% of patients. Aim: The aim of the study was to determine the distribution of cognitive disorders in people with MS. Methods: The prospective study included 135 respondents with MS and 50 healthy respondents. The respondents were divided into three groups: the first group consisted of 85 respondents where the disease lasted longer than one year, the second group consisted of 50 respondents with newly diagnosed MS, the third group consisted of 50 healthy respondents. Clinical assessment instruments were: Extended Disability Score in Multiple Sclerosis Patients, Mini Mental Status, Battery of Tests to Assess Cognitive Functions: Wechsler Intelligence Scale, Revised Beta Test, Raven Colored Progressive Matrices, Wechsler Memory Scale, Rey Audio Verbal Learning Test -Osterriecht’s complex character test, verbal fluency test. Results: Cognitive disorders were present in 40-60% of respondents with MS. Visuospatial, visuoconstructive and visuoperceptive functions are worse in the first group. Mnestic functions (learning process, short-term and long-term memory, recollection, verbal-logical memory) were most affected in both groups of respondents, ranging from 30-60%. Poorer cognitive domains are in the first groups of respondents. Immediate working process memory (current learning), memory, attention, short-term and logical memory is worse in the examinees of the first group. At the beginning of the disease, 16% had verbal fluency difficulties, and as the disease progresses, the difficulties become more pronounced. Conclusion: Cognitive disorders are heterogeneous, they can be noticed in the early stages of the disease. They refer to impairments of working memory, executive functions and attention, while global intellectual efficiency is later reduced.

Contributing to the literature on brand stereotyping, we draw on the Stereotype Content Model to investigate whether the content of the brand stereotype (in terms of warmth and competence) impacts consumers' perceptions of functional, emotional and social value. In doing so, we explicitly account for the brand's level of perceived globalness (PBG) and localness (PBL) as known influences on both stereotype content and value perceptions. Across two studies, we find that brand warmth consistently and positively impacts functional and emotional value, whereas brand competence enhances functional value. The impact of the stereotyping dimensions on value is subsequently reflected in increased purchase intentions and higher brand ownership. Surprisingly, none of the latter outcomes is affected by social value. Our findings corroborate previous research showing that PBG and PBL are important drivers of brand stereotype content, but also reveal that brand warmth has a stronger impact on behavioral outcomes than brand competence.

To tackle a specific class of engineering problems, in this paper, we propose an effectively integrated bat algorithm with simulated annealing for solving constrained optimization problems. Our proposed method (I-BASA) involves simulated annealing, Gaussian distribution, and a new mutation operator into the simple Bat algorithm to accelerate the search performance as well as to additionally improve the diversification of the whole space. The proposed method performs balancing between the grave exploitation of the Bat algorithm and global exploration of the Simulated annealing. The standard engineering benchmark problems from the literature were considered in the competition between our integrated method and the latest swarm intelligence algorithms in the area of design optimization. The simulations results show that I-BASA produces high-quality solutions as well as a low number of function evaluations.

Abstract Epilepsy is a neurological disorder characterised by unusual brain activity widely known as seizure affecting 4-7% of the world's population. The diagnosis of this disorder is currently based on analysis of the electroencephalography (EEG) signals in the time-frequency domain. The analysis is performed applying various algorithms that yield high performance, however the challenge of effective real-time epilepsy diagnosis persists. To address this, we have developed a Field Programmable Gate Array (FPGA) based solution for the classification of generalized and focal epileptic seizure types using a feed-forward multi-layer neural network architecture (MLP ANN). The neural network algorithm is trained, validated and tested on 822 captured signals from Temple University Hospital Seizure Detection Corpus (TUH EEG Corpus) database. Inputs into the system were five main features obtained from EEG signals by time-frequency analysis followed by Continuous Wavelet Transform (CWT) and subsequent statistical analysis. Out of the total number of samples, 583 (70 %) of them were utilised during the system development in MATLAB and TensorFlow and 239 (30 %) samples were further used for subsequent testing of the model performance on the FPGA. Subsequently, the adequate parameters of the ANN model were determined by using k-Fold Cross-Validation. Finally, the best performing ANN model in terms of average validation data accuracy achieved during cross-validation was implemented on the FPGA for real-time seizure classification. The digital ANN solution was coded in Very High-Speed Integrated Circuit Hardware Description Language (VHDL) and tested on the FPGA using 30 % reaming data. The results of this research demonstrate that epilepsy diagnosis with quite high accuracy (95.14 %) can be achieved with (5-12-3) MLP ANN implemented on FPGA. Also, the results show the steps towards appropriate implementation of ANN on the FPGA. These results can be utilised as the basis for the design of an application-specific integrated circuit (ASIC) allowing large serial production.

Abstract The paper presents results of the measurements of the tropospheric ozone (O3) concentration and meteorological parameters: temperature, air pressure, relative humidity, speed and wind direction. The data were collected from January 2016 to December 2016 at station located in locality Centre (Banja Luka), Republic of Srpska, Bosnia and Herzegovina. Ozone is one of the most harmful pollutants to plants and health and highly reactive secondary pollutant. The present study covers investigation of the relationship between the concentration of ozone and meteorological parameters as well as time variations of ozone concentration (by hours, months, seasons). This topic has not been studied up to now in this region, although the recent research data indicates that there is a correlation between them and previously obtained from the world’s relevant scientific centres, as already cited above. Statistical analysis confirms string of rolls, which shows directional connection between tropospheric ozone and meteorological parameters, specially temperature (r = 0.148), air pressure (r = –0.292) and relative humidity (r = –0.292). These parameters are the most important meteorological factors influencing the variation in ozone levels during the research. The correlation ozone concentrations with speed and direction of wind is not significant, like other parameters.

Abstract The paper proposes a discrete-time sliding mode controller for single input linear dynamical systems, under requirements of the fast response without overshoot and strong robustness to matched disturbances. The system input saturation is imposed during the design due to inevitable limitations of most actuators. The system disturbances are compensated by employing nonlinear estimation by integrating the signum of the sliding variable. Hence, the proposed control structure may be regarded as a super-twisting-like algorithm. The designed system stability is analyzed as well as the sliding manifold convergence conditions are derived using a discrete-time model of the system in the δ-domain. The results obtained theoretically have been verified by computer simulations.

In the modern telecommunication systems, mobility is one of the key advantage of wireless communications, given that it is possible to transmit/receive data, without caring of having a static position into the network. Of course, mobility poses special issues such as degradations, channel quality fluctuations, fast topology changes, and so on. Modern researches focus their attention on predicting mobile future node positions, in order to a-priori know, for example, what the evolution of the network topology will be or which level of stability each node will reach. Each prediction scheme is based on the storage and analysis of several historical mobility trajectories, in order to train the proper prediction algorithm. In this paper, we focus our attention on the optimization of the space needed to store historical mobility samples, encoding their values and evaluating the conversion error, comparing different encoding functions. Several simulation campaigns have been carried out in order to evaluate the goodness and feasibility of our proposal.

This paper analyzes the vollatility spillover between the conventional index in Malaysia FTSE Malaysia KLCI (KLSE) and the Islamic index in Malaysia FTSE Bursa Malaysia Shariah Index (FTFBMHS). Monthly observations spanning in a period from 2002 to 2018 are obtained from investing.com database. GARCH model and Johansen cointegration test are used to investigate volatility spillover and the relationship between two indices. The results of the analysis indicate that in the short-run there is volatility spillover between FTSE Malaysia KLCI and FTSE Bursa Malaysia Shariah Index, while in the long-run there is no relationship between the two indices. The methodology of compiling Islamic indeces is based on Shariah law. Keywords: Conventional

Introduction: Mathematical modeling of coronavirus disease spread and computer simulations are currently one of the main tools in public health that can give important indicators for prevention planning. Based on mathematical projections and daily updates of information, the measures are either tightened or reduced, in order to protect the health of the population. Aim: The aim of this paper is to present a computer system based on an adequate mathematical model that allows frequent execution of various scenarios of spread severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in any period in the future. Also, the aim of this article is to point out the importance of measures for the prevention of coronavirus disease 2019 (COVID-19) in Bosnia through examples of computer simulation models. Methods: Software solution based on the USLIRD model (Unpopulated - Susceptible - Latent - Infectious - Recovered - Deceased) was developed, with a number of variable parameters ‘reproduction number, delay period, infectivity period, hospital capacity, characteristics of population). By setting these parameters in accordance with the existing and available data, the model is brought to an optimized state with the possibility of a realistic assessment of the course of the infection curve in any future period. Data from the beginning of the pandemic are collected at the Faculty of Mechanical Engineering, University of Sarajevo and updated several times a day. The set of measures is divided into two types. 'Intervention 1' is a measure to close institutions that are at high risk for pandemics, working from home, wearing face mask, enhanced hygiene when entering facilities with a larger number of people. 'Intervention 2' presents restrictive measures that has been introduced as mandatory in Bosnia. The period 01.03.2020 to 01.09.2020 was observed. Results: Without epidemiological measures, Bosnia's health system would quickly collapse. Restrictive measures reduce the intensity of the spread of the infection, save human lives and keep the health system functional, but with consequences on other aspects of society - reduction of economic activities, collapse of the service industry and companies and disorders in mental health status of the population. Four different scenarios of the situation were analyzed. Scenario number three is current condition with measures that are currently in Bosnia. The reintroduction of restrictive measures leads to a decrease in the number of infected population and suppression of the spread of the pandemic, which is shown in scenario 4. Conclusion: Self-discipline, adherence to measures, while trying to avoid restrictive measures should be the way to fight the COVID-19 pandemic. Whatever the consequences, the initiation of restrictive measures to preserve the health of the population should be imperative.

Internet of Things (IoT) becomes an emerging network technology that expedites billions of devices to be connected via the Internet to provide real-time intelligent application services. The benefits of Software-Defined Networking (SDN) can be used to fulfill IoT requirements. Quality of Service provisioning is an on-going demand in software-defined IoT (SD-IoT), particularly for large scale environments. In this paper, we address this issue by proposing a seamless model of AI-aided Traffic Differentiated QoS Routing and Dynamic Offloading in distributed fragmentation optimized SDN-IoT. Firstly, we propose a Multi-Criterion based Deep Packet Inspection method for classifying the network traffic, which is held in Edge Routers (access points). Secondly, we construct a Partially Connected Network Topology using the ISOMAP algorithm for an effective rule placement and routing. We propose a Traffic Differentiated QoS Routing for forwarding data packets via the most suitable switches. We select the optimum route by Deep Alternative Neural Network (DANN). Based on the relationships among switches, the path is selected and flow rules are deployed. The poor QoS is often caused by load imbalance in controllers and switches. To overwhelm this issue, we propose a Dynamic Offloading scheme in SD-IoT. We offload the data packets from the overloaded controller to the underloaded controller using Hassanat Distance-based K-nearest neighbors (HDK-NN) algorithm. Similarly, we propose a Ranking-based Entropy function (R-Ef) to allow dynamic offloading among switches. Simulation is performed using the NS3.26 simulator and the results proved that our proposed AI-aided SD-IoT model provides superior QoS performance compared to previous approaches.

While the genetic evolutionary features of solid tumour growth are becoming increasingly described, the spatial and physical nature of subclonal growth remains unclear. Here we utilise 102 macroscopic whole tumour images from clear cell renal cell carcinoma (ccRCC) patients, with matched genetic and phenotypic data from 756 biopsies. Utilising a digital image processing pipeline the boundaries between tumour and normal tissue were marked by a renal pathologist, and positions of boundary line and biopsy regions were extracted to X- and Y-coordinates. The coordinates were then integrated with genomic data to map exact spatial subclone locations, revealing how genetically distinct subclones grow and evolve spatially. A phenotype of advanced and more aggressive subclonal growth was present in the tumour centre, characterised by an elevated burden of somatic copy number alterations, higher necrosis, proliferation rate and Fuhrman grade. Moreover, metastasising subclones were found to preferentially originate from the tumour centre. Collectively these observations suggest a model of accelerated evolution in the tumour interior, with harsh hypoxic environmental conditions leading to heightened cellular turnover and greater opportunity for driver SCNAs to arise and expand due to selective advantage. Tumour subclone growth was found to be predominantly spatially contiguous in nature, with subclone dispersal a rare event found in two cases, which notably was associated with metastasis. In terms of genetic events, the largest subclones spatially were dominated by driver somatic copy number alterations, suggesting a large selective advantage can be conferred to subclones upon acquisition of these alterations. In conclusion, spatial dynamics is strongly associated with genomic alterations and plays an important role in tumour evolution.

The study of crystalluria is of great importance for the detection of substances of endogenous or exogenous origin that are present inthe urine, to a greater or lesser extent. Urinary sediment crystals can provide valuable answers for the assessment of therapeutic efficacy, as well as congenital and/or acquired pathophysiological conditions. The nature of the observed crystals informs the clinician of the biochemical irregularity of the urine. Crystalluria is of clinical significance only if it has been studied under good test conditions (sample selection, time and storage conditions). Crystalluria interpretations are performed on the basis of the urinary pH determined with statistically significant reliability. When studying crystalluria by light microscopy, it is necessary to provide light polarization or bidirectional illuminationin order to reduce the risk of diagnostic error.

U radu su korišteni podaci izmjere 377 modelnih stabala smreke koja su mjerena u oborenom stanju na širem području unutar državnih raznodobnih sastojina u Kantonu 10 (Hercegbosanski Kanton). Za određivanje volumena krupnog drveta stabala primijenjena je metoda sekcioniranja sa sekcijama nejednakih apsolutnih dužina (najčešće od 1 – 2 m). Za izravnanje veličina volumena krupnog drveta od prsnog promjera i visine stabala primijenjena je metoda višestruke regresijske analize. Za procjenu parametara korištenih funkcija, testiranje značajnosti njihovih razlika te provođenje raznih transformacija, kao softversko rješenje korišteni su StatGraphics Centurion XVII. i Statistica 8.0. U cilju izbora „najboljeg“ modela za procjenu volumena krupnog drveta testiran je veći broj poznatih dendrometrijskih dvoparametarskih volumnih funkcija. Kvaliteta izjednačenja i prikladnost testiranih modela ocjenjivani su na bazi utvrđenih veličina osnovnih statističkih pokazatelja za karakteriziranje jačine korelacijskih veza. Najbolje ocjene parametara pokazao je model V7=a0+a1d1,3+a2h+a3d1,3h+a4d1,32+a5 d1,32h uz utvrđeni koeficijent determinacije: R2 = 0,99 i veličinu standardne greške regresije Sey=0,24 m3. Testirajući značajnost razlika između stvarnih volumena stabala iz uzorka i volumena tih istih stabala utvrđenih primjenom odabranog regresijskog modela, utvrđen je prosječni postotak odstupanja od 0,44%. To znači da su u prosjeku za 0,44% niži volumeni u odnosu na stvarne volumene na uzorku od 377 stabala smreke, što ukazuje da je ovaj regresijski model upotrebljiv za primjenu u praktičnom radu, jer je taj prosječni postotak manji od 1%.