In this work, we present an approach for training parametrized physics informed neural networks to solve PDEs in a self supervised fashion, which means that no labeled input-output data is needed to train the neural network. The main contribution of this work is the achievement of a model with parameterizable boundary condition functions. This means that no retraining is needed to produce correct results for changing boundary conditions.

This paper is effectively a scenario analysis of the energy system of Bosnia and Herzegovina (BiH) from the perspective of the possible future reduction of greenhouse gas (GHG) emissions in the power generation sector, with the aim to become climate neutral by 2050, in compliance with the Green Agenda for the Western Balkan. According to the data from 2016, the share of power generation in the total GHG emissions in BiH was approximately 50%. By using the LEAP (Long-range Energy Alternatives Planning) energy model, two scenarios—the “gradual transition scenario” and the “climate neutral” scenario—have been analyzed for the period 2018–2050, and each scenario included decarbonization measures such as the extensive use of Renewable Energy Sources (RES). Unlike the climate neutral scenario, the gradual transition scenario includes the replacement of certain parts of the old, currently-in-operation Coal-fired Power Plants (CFPPs) with the new CFPP, which is more efficient. In the climate-neutral scenario, that part of the existing CFPPs is replaced by a mix of RESs. The results from the first scenario suggest that the share of CFPPs in electricity generation has gradually decreased from 69.3% to 16.3% in 2050, and CO2 emissions from the power generation sector in 2050 will be 2.2 million tons—roughly 83.5% less than in 2014. According to the second scenario, the emphasis is strongly on the growth and promotion of RESs, which have significantly taken over the roles of major producers of electricity, encouraging the low-carbon development of BiH. Analysis results show that, in 2050, there will be no CO2 emissions from power generation. It can be concluded that specifically designed energy models for the optimization of capacities and CO2 emissions through convergence towards RESs could be an optimistic and promising option for BiH to become climate neutral while meeting increasing energy demands. The results show the required RES capacities needed for achieving climate-neutral power generation by 2050, with the current rate level of power generation. Based on the results, RES investment needs can be estimated. Overall, the results of the scenarios can be used for the strategic planning of the power generation sector in BiH until 2050.

Introduction: Pandemics have affected and will continue to affect humankind. Historically the Human Immunodeficiency Virus changed the way dental clinics operate and the COVID-19 pandemic led to an unprecedented closure of dental clinics leading to short- and long-term impact on oral health.  Aim: To assess the impact on oral health and related behaviours related to modern pandemics.  Method: A literature search across eighteen electronic databases was conducted. Three reviewers screened 2029 articles against inclusion criteria and assessed quality. Included articles underwent thematic analysis, followed by narrative synthesis to describe the results.  Results: Forty-eight articles were included that identified themes: (i) oral health related quality of life, (ii) stress and pandemics, (iii) oral health behaviours, (iv) social capital, (v) access to oral healthcare, (vi) fear as a barrier to accessing oral healthcare and (vii) teledentistry.  Conclusion: Pandemics present multiple challenges to both individuals and oral health professionals that impact on oral health and these challenges disproportionately affect the most vulnerable communities. However, with the right support, these impacts can be mitigated through social capital and support to establish healthy routines. The use of digital technologies should be promoted to reach all communities before the next pandemic arrives.

Conversation can play an essential role in forging bonds between humans and social robots, but participants need to feel like they are being listened to, remembered, and cared about in order to effectively build rapport. In this paper, we propose a novel strategy for conducting small talk with a social robot. Our approach is known as the Tiers of Friendship. It is centered around three core design elements: 1) Persuasive content and character is provided through topic modules created by professional creative writers to ensure engaging conversational content and a compelling personality for the social robot. 2) Conversational memory is achieved by allowing topic modules to specify required information that can be learned through conversation or recalled from previous interactions and organizing topic modules into a hierarchy that enforces information requirements between topics. 3) Dynamicity in conversation is promoted through topic navigation that supports fluid transitions to topics of human interest and employs elements of random ordering to create fresh conversation experiences. In this paper, we show how the Tiers of Friendship can be used to generate conversation content for a social robot that encourages the development of rapport. We describe a working implementation of a small talk system for a social robot based on the Tiers of Friendship that combines off-the-shelf ASR and NLU components and custom robot behavior components implemented via behavior trees on ROS. Finally, in order to evaluate our approach's effectiveness, we conduct an elicitation survey that evaluates conversations in terms of perceived engagement, personality traits, and rapport expectation and discuss the implications for social robotics.

The Covid-19 pandemic is undoubtedly a global crisis, which has had a negative impact on all economic activities. Tourism, as an extremely sensitive branch, is unquestionably the most affected sector. The Republic of Croatia, as well as Bosnia and Herzegovina, has continuously had an increase in tourist traffic in the number of tourist arrivals and overnight stays until the moment of the pandemic crisis. The aim of this paper is to consider the consequences on tourist flows on the example of Bosnia and Herzegovina – Tuzla County and the Republic of Croatia – SisakMoslavina County. The results of the research, conducted through a questionnaire, will detect the problems in the pandemic crisis faced by economic operators in tourism, and we will try to find answers on how to overcome this health crisis with the least possible consequences for tourism, improve the efficiency of tourist boards, as initial affirmers and promoters of tourism.

Dysphagia or swallowing dysfunction is any impairment in the swallowing function that may cause difficulty or discomfort in initiating or transferring the bolus from the oral cavity into the stomach. Dysphagia can cause the bolus to reroute into the airway, known as aspiration, which can lead to more adverse outcomes such as pneumonia or even death. Videofluoroscopic swallowing study (VFSS) is the gold standard procedure for dysphagia diagnosis. In VFSS, trained clinicians calculate swallowing kinematics and inspect pathophysiological processes in a frame-by-frame manner. Though effective, VFSS evaluation is time-consuming, prone to subjectivity in judgment, and human error. In this study, we present a cascaded pipeline that employs various deep learning algorithms to automate VFSS analysis to identify swallowing abnormalities. The pipeline initially segments the VFSS video into static and dynamic frames which include all the relevant features of swallowing for the subsequent VFSS analysis tasks. These tasks include pharyngeal swallow segmentation, hyoid bone tracking, bolus segmentation, and aspiration detection. The pipeline starts with a shallow neural network (NN) that differentiates between static and dynamic VFSS frames with a 98% accuracy using spatio-temporal features from TV-L1 optical flow. Then, a Single Shot Multi-box Detector (SSD) model localizes the hyoid bone body with a mean average precision (mAP) of 40% at an intersection over union (IOU) of 0.5 in a fast and beyond average performance even when the hyoid bone is occluded by the mandible. So far, the developed automated pipeline has shown comparable performance to the manual analysis performed by trained clinicians.

Difference approaches to the study of excited states have undergone a renaissance in recent years, with the development of a plethora of algorithms for locating self-consistent field approximations to excited states. Density functional theory is likely to offer the best balance of cost and accuracy for difference approaches, and yet there has been little investigation of how the parametrization of density functional approximations affects performance. In this work, we aim to explore the role of the global Hartree-Fock exchange parameter in tuning accuracy of different excitation types within the framework of the recently introduced difference projected double-hybrid density functional theory approach and contrast the performance with conventional time-dependent double-hybrid density functional theory. Difference projected double-hybrid density functional theory was demonstrated to give vertical excitation energies with average error and standard deviation with respect to multireference perturbation theory comparable to more expensive linear-response coupled cluster approaches ( J. Chem. Phys.2020, 153, 074103). However, despite benchmarking of local excitations, there has been no investigation of the methods performance for charge transfer or Rydberg excitations. In this work we report a new benchmark of charge transfer, Rydberg, and local excited state vertical excitation energies and examine how the exact Hartree-Fock exchange affects the benchmark performance to provide a deeper understanding of how projection and nonlocal correlation balance differing sources of error in the ground and excited states.

This research aims to examine the traffic noise levels and to improve the performances of the Calculation of Road Traffic Noise model (C.R.T.N.) by applying the statistical multiple linear regression approach. Research methods included traffic noise level measurements with a noise measuring device in an urban area, using a sampling method in different periods. An evaluation of the measured data and prediction results was performed. Based on the predicted values of the C.R.T.N. model and coefficient of determination (R2), multi-linear regression was carried out to determine statistically significant parameters. The obtained multi-linear regression equation defined a new form of C.R.T.N. model. When applying the new improved version based on the C.R.T.N. model, higher accuracy of prediction is achievable. It can be seen that by applying multi-linear regression, the obtained prediction values are acceptably equated with field measurements in the chosen research environment. So, in this way, the differences between the predicted values of the noise level and the values measured in the field were minimized. Finally it can be concluded that when applying the new improved version based on the C.R.T.N. model, higher accuracy of prediction is achievable.

Traditional fuel-powered vehicle emissions have long been recognized as a major barrier to a sustainable environment, and their minimization could ensure both economic support for the sustainable societal fundament and pollution prevention. Electrifying light-duty vehicle fleets, such as taxis, could provide a financial return as well as bring significant economic and environmental improvements. This paper offers a ranked selection of electric vehicles that are presently available on the market, as reviewed by taxi service representatives, as well as their own evaluation of the criteria that influence this selection. This paper provides stability and support when making decisions by deploying stepwise weight assessment ratio analysis and a modified standard deviation method for calculating the subjective and objective weights of the criteria, as well as performing sensitivity analysis to determine how a particular criterion affects the multi-attributive border approximation area. A comparison ranking of the alternatives discovered how a change in the weight value of one of the criteria affected the ranking of the electric vehicle alternatives. According to the research, led by the battery capacity criterion and its values, the Volkswagen ID.3 Pro has the best results and is the taxi of choice in the Brčko District of Bosnia and Herzegovina. Furthermore, the research has demonstrated that the development of electric vehicles for taxi service purposes should strive to extend the range of these vehicles while reducing the battery charging time.

Today’s economic systems are, on the one hand, exposed to various risks and uncertainties with their trends changing almost daily, while on the other hand, they represent an extremely complex system with a large number of sustainable influential parameters. The challenge is to model macroeconomic parameters and achieve sustainability, yet also satisfy conflict situations with an increased level of uncertainty. The aim of this paper is to create an appropriate functional model by examining the mutual influence of various macroeconomic factors. It assesses a total of four scenarios considering mutual influences of: FDI (foreign direct investments), GDP (gross domestic product), imports, exports, inflation rate, RER (real exchange rate) and employment rate as defined parameters. First, the DEA (Data envelopment analysis) model was applied to determine the initial efficiency of two countries, Bosnia and Herzegovina (BIH) and Serbia, for the period 2005–2020. Then, PCA (Principal Component Analysis) was applied in combination with DEA to obtain more precise values. In addition, IMF SWARA (Improved Fuzzy Stepwise Weight Assessment Ratio Analysis) was applied to define weight coefficients of macro-economic parameters. Finally, the CRADIS (compromise ranking of alternatives from distance to ideal solution) model was applied for the final ranking of part of decision-making units. This developed, integrated model can be classified as a mathematical method for economic analysis and gives extended opportunities for solving different problems. The results show that 2009, 2013 and 2016 were the most favorable years in terms of the conditions set when it comes to Bosnia and Herzegovina, and 2012, 2014 and 2016 when it comes to Serbia. These years have been singled out and can be a benchmark for further handling and modeling of macroeconomic elements. In addition, correlation was tested using statistical coefficients.

The use of digital teaching resources became widespread and very helpful during the COVID‐19 pandemic as an alternative to a traditional course with cadavers. Technologies such as augmented reality (AR), virtual reality (VR), 3D models, video lectures and other online resources enable three‐dimensional visualization of the anatomical structures and allow students to learn more interactively. The aim of this study was to compare students' performance in the traditional anatomical courses in teaching neuroanatomy and technology‐based learning methods such as video lectures, 3D models and 3D printed specimens. Four groups of first‐year students of Veterinary Faculty established for the practical classes during the academic year 2021/2022 took part in this research. The total number of students participating in this research was 72. Each group attended separately the theoretical lecture with a demonstration based on a different technique; the control group used formalized specimens, while the three experimental groups used video lectures, 3D models and 3D printed specimens, respectively. Subsequently, all groups completed the same questionnaire testing their short‐term memory of the neuroanatomical structures. After four weeks students were tested for their long‐term memory of the neuroanatomy lecture with the follow‐up test containing an identical list of questions. The test scores using video lectures and 3D printed models were significantly higher compared with the group that learned in the traditional way. This study suggests that alternative approaches such as technology‐based digital methods can facilitate memorization of anatomical terms and structures in a more interactive and sensory engaging way of learning.

Dripping rainfall simulators are important instruments in soil research. However, a large number of non-standardized simulators have been developed, making it difficult to combine and compare the results of different studies in which they were used. To overcome this problem, it is necessary to become familiar with the design and performances of the current rainfall simulators. A search has been conducted for scientific papers describing dripping rainfall simulators (DRS) and papers that are thematically related to the soil research using DRS. Simulator design analysis was performed integrally, for simulators with more than one dripper (DRS>1) and with one dripper (DRS=1). Descriptive and numerical data were extracted from the papers and sorted by proposed categories, according to which the types and subtypes of used simulators are determined. The six groups of elements that simulators could consist of have been determined, as well their characteristics, representation and statistical analyses of the available numerical parameters. The characteristics of simulators are analyzed and presented, facilitating the selection of simulators for future research. Description of future simulators in accordance to the basic groups of simulator elements should provide all data necessary for their easier replication and provide a step closer to the reduction of design diversification and standardization of rainfall simulators intended for soil research.

Systemic sclerosis-associated interstitial lung disease (SSc-ILD) is rare, poorly understood, with heterogeneous characteristics resulting in difficult diagnosis. We aimed to systematically review evidence of soluble markers in peripheral blood or bronchoalveolar lavage fluid (BALF) as biomarkers in SSc-ILD. Method Five databases were screened for observational or interventional, peer-reviewed studies in adults published between January 2000 and September 2021 that assessed levels of biomarkers in peripheral blood or BALF of SSc-ILD patients compared with healthy controls. Qualitative assessment was performed using Critical Appraisal Skills Programme (CASP) checklists. Standardised mean difference (SMD) in biomarkers were combined in random-effects meta-analyses where multiple independent studies reported quantitative data. Results 768 published studies were identified; 38 articles were included in the qualitative synthesis. Thirteen studies were included in the meta-analyses representing three biomarkers: KL6, SP-D and IL-8. Greater IL-8 levels were associated with SSc-ILD in both peripheral blood and BALF, overall SMD 0.88 (95% CI 0.61 to 1.15; I2=1%). Greater levels of SP-D and KL-6 were both estimated in SSc-ILD peripheral blood compared with healthy controls, at an SMD of 1.78 (95% CI 1.50 to 2.17; I2=8%) and 1.66 (95% CI 1.17 to 2.14; I2=76%), respectively. Conclusion We provide robust evidence that KL-6, SP-D and IL-8 have the potential to serve as reliable biomarkers in blood/BALF for supporting the diagnosis of SSc-ILD. However, while several other biomarkers have been proposed, the evidence of their independent value in diagnosis and prognosis is currently lacking and needs further investigation. PROSPERO registration number CRD42021282452.