Artificial intelligence has been used in the medical field for many years finding its way in assisting medical professionals to solve complex tasks in almost every application domain. Patient care and health care costs are in the forefront of any medical system. With the emergence of electronic games (EG), originally entertaining in nature, potential was seen for their application in the medical field in terms of exploring the capabilities of educating patients in the areas of need. Design of serious games emerged with a focus on helping with the patient’s therapeutic modality. Some of the medical field serious games are part of the daily work of a medical professional. With the development of mobile devices, this type of gaming has also become available to the wider population. The aim of this research was to demonstrate the possibilities and benefits of using EG in the daily clinical work of medical professionals. This work has a descriptive character where a section of the literature is given outlining the increased research interest in using serious games as therapy enhancements. We limited ourselves to research involving patients who need medical treatment or intervention.

Background Coronary chronic total occlusion (CTO) is characterized by the presence of collateral blood vessels which can provide additional blood supply to CTO-artery dependent myocardium. Successful CTO recanalization is followed by significant decrease in collateral donor artery blood flow and collateral derecruitment, but data on coronary hemodynamic changes in relation to myocardial function are limited. We assessed changes in coronary flow velocity reserve (CFVR) by echocardiography in collateral donor and recanalized artery following successful opening of coronary CTO. Methods Our study enrolled 31 patients (60 ± 9 years; 22 male) with CTO and viable myocardium by SPECT scheduled for percutaneous coronary intervention (PCI). Non-invasive CFVR was measured in collateral donor artery before PCI, 24 h and 6 months post-PCI, and 24 h and 6 months in recanalized artery following successful PCI of CTO. Results Collateral donor artery showed significant increase in CFVR 24 h after CTO recanalization compared to pre-PCI values (2.30 ± 0.49 vs. 2.71 ± 0.45, p  = 0.005), which remained unchanged after 6-months (2.68 ± 0.24). Baseline blood flow velocity of the collateral donor artery significantly decreased 24 h post-PCI compared to pre-PCI (0.28 ± 0.06 vs. 0.24 ± 0.04 m/s), and remained similar after 6 months, with no significant difference in maximum hyperemic blood flow velocity pre-PCI, 24 h and 6 months post-PCI. CFVR of the recanalized coronary artery 24 h post-PCI was 2.55 ± 0.35, and remained similar 6 months later (2.62 ± 0.26, p = NS). Conclusions In patients with viable myocardium, prompt and significant CFVR increase in both recanalized and collateral donor artery, was observed within 24 h after successful recanalization of CTO artery, which maintained constant during the 6 months. Trial registration ClinicalTrials.gov (Number NCT04060615 ).

In this paper, we consider the problem of learning a generalized Nash equilibrium (GNE) in strongly monotone games. First, we propose a novel continuous-time solution algorithm that uses regular projections and first-order information. As second main contribution, we design a data-driven variant of the former algorithm where each agent estimates their individual pseudo-gradient via zero-order information, namely, measurements of their individual cost function values, as typical of extremum seeking control. Third, we generalize our setup and results for multi-agent systems with nonlinear dynamics. Finally, we apply our algorithms to connectivity control in robotic sensor networks and distributed wind farm optimization.

Air pollution is a complex mixture of toxic components that has the direct impact on human health, life quality, and the environment. In this study, meteorological variables and concentration of air pollutants are used to predict the common air quality index (CAQI) in Bjelave neighborhood, Sarajevo, BiH. CAQI prediction models were built using five popular machine learning techniques in the air pollution domain: Support Vector Regression, Random Forest, Extreme Gradient Boosting, Multiple Linear Regression and Multilayer Perceptron, using three-year period data (2016–2018). Prediction performance was measured using regression metrics: R-squared and RMSE. Ensemble technique, Random Forest method achieved the best performance results from the five evaluated machine learning methods: R2 = 0.99 and RMSE = 2.30, using the dataset when missing values were removed, and R2 = 0.99 and RMSE = 2.58 using the dataset when missing values were imputed using linear regression method.

In this paper, analogy of heat- or mass-transfer problem and progression of species in natural environments is used. Time-dependent spatial distributions of species in a given population (such as microbes, bacteria, viruses or herewith infected individuals etc.) are described by time-dependent convection-diffusion equation, which is widely used in analysis of a number of physical, engineering or environmental problems. In the presented example, a finite-volume method is used for its numerical solution. The source term is calibrated using the available recorded data on the species concentration development in the beginning stage. The integral results of the proposed model are compared with another, integral model (SIR) based on ordinary differential equations. Calculated spatial distributions for three different modeling cases are given.

Using tissue microarrays it was shown that membranous C-terminal MET immunoreactivity and ectodomain (ECD) shedding are associated with poor prognosis in oral cancer. Seen the potential diagnostic value, extrapolation of these results to whole tissue sections was investigated. Since MET orchestrates epithelial to mesenchymal transition (EMT), results were benchmarked to loss of E-cadherin; a readout for EMT known to be associated with poor prognosis. C-terminal MET, N-terminal MET, and E-cadherin immunoreactivities were examined on formalin-fixed paraffin-embedded parallel sections of 203 oral cancers using antibody clones D1C2, A2H2-3 and NCH-38. Inter- and intra-antibody relations were examined using a novel scoring system, nonparametric distribution and median tests. Survival analyses were used to examine the prognostic value of the observed immunoreactivities. Assessment of the three clones revealed MET protein status (no, decoy, transmembranous C-terminal positive), ECD shedding and EMT. For C-terminal MET positive cancers, D1C2 immunoreactivity is independently associated with poor overall survival (OS; HR=2.40; 95% CI, 1.25 to 4.61 and P=0.008) and disease free survival (DFS; HR=1.83; 95% CI 1.07 - 3.14; P=0.027). For both survival measures, this is also the case for ECD shedding (43.4% with HR=2.30; 95% CI, 1.38 to 3.83 and P=0.001 versus HR=1.87; 95% CI 1.19 - 2.92; P=0.006) and loss of E-cadherin (55.3% with HR=2.21; 95% CI, 1.30 to 3.77 and P=0.004 versus HR=1.90; 95% CI 1.20 - 3.01; P=0.007). The developed scoring system accounts for MET protein status, ECD shedding, EMT, and is prognostically informative. These findings may contribute to development of companion diagnostics for MET-based targeted therapy.

Recovering the T2 distribution from multi-echo T2 magnetic resonance (MR) signals is challenging but has high potential as it provides biomarkers characterizing the tissue micro-structure, such as the myelin water fraction (MWF). In this work, we propose to combine machine learning and aspects of parametric (fitting from the MRI signal using biophysical models) and non-parametric (model-free fitting of the T2 distribution from the signal) approaches to T2 relaxometry in brain tissue by using a multi-layer perceptron (MLP) for the distribution reconstruction. For training our network, we construct an extensive synthetic dataset derived from biophysical models in order to constrain the outputs with a priori knowledge of in vivo distributions. The proposed approach, called Model-Informed Machine Learning (MIML), takes as input the MR signal and directly outputs the associated T2 distribution. We evaluate MIML in comparison to a Gaussian Mixture Fitting (parametric) and Regularized Non-Negative Least Squares algorithms (non-parametric) on synthetic data, an ex vivo scan, and high-resolution scans of healthy subjects and a subject with Multiple Sclerosis. In synthetic data, MIML provides more accurate and noise-robust distributions. In real data, MWF maps derived from MIML exhibit the greatest conformity to anatomical scans, have the highest correlation to a histological map of myelin volume, and the best unambiguous lesion visualization and localization, with superior contrast between lesions and normal appearing tissue. In whole-brain analysis, MIML is 22 to 4980 times faster than the non-parametric and parametric methods, respectively.

Cilj rada je ukazati na to utječu li kontinuirane promjene u Međunarodnim standardima financijskoga izvještavanja na usporedivost i dosljednost financijskih izvještaja u kratkom i dugom roku. Za postizanje tog cilja identificirani su čimbenici koji utječu na promjene u računovodstvenim standardima kako bi se ukazalo na to zašto do takvih promjena uopće dolazi. Kako bi se došlo do znanstveno utemeljenih zaključaka, kronološkim redoslijedom analizirane su promjene u 15 Međunarodnih računovodstvenih standarda/Međunarodnih standarda financijskoga izvještavanja koje su zabilježene u vremenskom periodu od 1975. do 2018. godine. U istraživačkom procesu analizirana je dostupna izvorna literatura sa sadržajem standarda važećih u određenoj točki vremena. Različite edicije standarda istražene su primjenom odgovarajućih znanstvenih metoda; metode analize, sinteze, komparativne metoda, metode indukcije i dedukcije te ostalih relevantnih znanstvenih metoda. U radu je znanstveno utemeljenim metodama pokazano kako promjene u računovodstvenim standardima mogu imati utjecaja na kvalitetu financijskoga izvještavanja u određenim slučajevima jer onemogućavaju primjenu temeljnih računovodstvenih načela usporedivosti i dosljednosti. Znanstveni doprinos rada očituje se u identificiranju i sistematizaciji ključnih promjena u standardima te analizi njihova učinka na financijsko izvještavanje poduzeća koja primjenjuju MSFI-e.

Abstract The paper explores the possibilities of creating an econometric model for making short-term forecasts of the Gross Domestic Product of Bosnia and Herzegovina (GDP of B&H). Its aim is to determine the most representative and most efficient model for forecasting the quarterly GDP of B&H. This is the first paper that simultaneously compares ARIMA models, bridge models and factor models in three different time periods. All variables are available for the period of 2006q1-2016q4. The final choice of the model for forecasting the quarterly GDP of B&H was selected on the basis of a comparative analysis of the predictive efficiency of the analysed models. Based on the obtained results, the most efficient model for forecasting quarterly GDP of B&H is the bridge model, which includes four variables as regressor: Retail sale of other goods, Total loans, Manufacturing and Manufacture of food products.

ABSTRACT Introduction The discovery of the anticonvulsant properties of valproic acid and the development of valproic acid/valproate to market authorization for specific epilepsy types and syndromes, as well as their repurposing for other indications, are illustrative examples of both the strengths and weaknesses of drug development strategies. Areas covered This review summarizes and interprets the development and repurposing history of valproic acid/valproate. The article is based on articles, including original studies and systematic reviews obtained from PubMed, Scopus, EBSCO, SCIndeks and Google Scholar databases. Expert opinion Random screening and careful observation of the experimental effects of tested substances were crucial for discovering the anticonvulsant effects of valproic acid, while rational drug design and clinical observation strategies led to repurposing valproic acid and valproate for bipolar disorder maintenance treatmentand prevention of migraine attacks. Early planning and feasibility studies of future clinical trials are essential for obtaining marketing authorization of new substances or new indications of old anticonvulsants. Significant progress has been made recently toward understanding, treatment and prevention of hepatotoxicity caused by valproic acid/valproate, making its long-term administration safer. There are ongoing efforts to repurpose valproic acid/valproate for augmentation with antipsychotic drugs for the treatment of schizophrenia.