Recent work has shown how predictive modeling can endow agents with rich knowledge of their surroundings, improving their ability to act in complex environments. We propose question-answering as a general paradigm to decode and understand the representations that such agents develop, applying our method to two recent approaches to predictive modeling -action-conditional CPC (Guo et al., 2018) and SimCore (Gregor et al., 2019). After training agents with these predictive objectives in a visually-rich, 3D environment with an assortment of objects, colors, shapes, and spatial configurations, we probe their internal state representations with synthetic (English) questions, without backpropagating gradients from the question-answering decoder into the agent. The performance of different agents when probed this way reveals that they learn to encode factual, and seemingly compositional, information about objects, properties and spatial relations from their physical environment. Our approach is intuitive, i.e. humans can easily interpret responses of the model as opposed to inspecting continuous vectors, and model-agnostic, i.e. applicable to any modeling approach. By revealing the implicit knowledge of objects, quantities, properties and relations acquired by agents as they learn, question-conditional agent probing can stimulate the design and development of stronger predictive learning objectives.

Over the course of the last decade, the subfield of artificial intelligence, called deep learning, becomes the main technology that provides breakthroughs in the computer vision area. Likewise, deep learning algorithms made a major impact in the automated driving domain. This research aims to apply and evaluate the performance of two pre-trained deep learning algorithms in order to recognize different street objects. Both RCNN, as well as YOLO algorithms, are used to recognize bikes, cars and pedestrians using the public GRAZ-02 dataset composed of 1476 raw images of street objects. Accuracy greater than 90% is achieved in recognizing all considered objects. The fine-tuning and training of both algorithms is established using databases named ImageNet and COCO, and afterwards, trained models are tried on the test data.

Introduction: Health education is a process of acquiring knowledge and skills in order to improve the health of the individual and the community. It is considered the most effective, most economical and most rational aspect of health care and health culture. Aim To provide data on the effectiveness of printed health-educational materials. Methods This is a quantitative, applied, descriptive-analytical study. According to the type of research, it presents a public health evaluation manipulative study with triple testing. The research was conducted in elementary schools in the Zenica-Doboj Canton. The total number of students participating in the research is divided into groups: examined, control group. The research consisted of four phases. The research tool is a modified questionnaire The Health Behavior in School-aged Children (HBSC) with 38 questions, 8 modules. Results The total number of respondents was 120. The method of distribution of health-educational posters shows a lower but still present statistical significance (p<0.05) in relation to the acquired knowledge and a change in attitudes between the conducted surveys at different time points. There is no statistically significant change (p>0.05) in the level of knowledge and attitudes using leaflets between conducted surveys at three different times. In the control group without education, there was a low statistical significance (p<0.05) in terms of changing the level of knowledge and attitudes. Conclusion The distribution of health-educational posters is recommended in situations where it is necessary to reach a wide audience for a long period of time, if the site of the poster is protected. According to this study, there is no evidence that the leaflet distribution method should be used when it comes to the promotion of healthy lifestyles among healthy children. Alternative methods and ways of health education need to be identified.

Non‐invasive axon diameter distribution (ADD) mapping using diffusion MRI is an ill‐posed problem. Current ADD mapping methods require knowledge of axon orientation before performing the acquisition. Instead, ActiveAx uses a 3D sampling scheme to estimate the orientation from the signal, providing orientationally invariant estimates. The mean diameter is estimated instead of the distribution for the solution to be tractable. Here, we propose an extension (ActiveAxADD) that provides non‐parametric and orientationally invariant estimates of the whole distribution.

Cocoa shell is a by-product of the chocolate industry that is rich in dietary fiber and bioactive components. In this research, the influence of high voltage electric discharge (HVED) treatment on chemical and physical characteristics of the cocoa shell, i.e., the effects of applied time and frequencies on grinding ability, water binding capacity (WBC), dietary fibers and tannin content was investigated. HVED had a significant influence on the chemical and physical properties of cocoa shell, all of which could be linked to changes in fiber properties. Along with the fiber content, grinding ability and water binding capacity were increased. These properties have already been linked to fiber content and soluble/insoluble fiber ratio. However, this research implies that change in fiber properties could be linked to tannin formation via complexation of other polyphenolic components. Additional research is needed to verify this effect and to establish mechanisms of tannin formation induced by HVED and its influence on fiber quantification.

Introduction: Colorectal cancer (CRC) is the third most common cancer worldwide with the incidence of about 1,8 million newly diagnosed cases in 2018. According to the World Cancer Report 2014, in Bosnia and Herzegovina 6700 people died of cancer in 2014, and CRC was the cause of mortality in 724 patients (10%). Prevention programs including screening, state-of-the-art diagnostic modalities and therapeutic approaches to CRC are being constantly improved. Aim: Our study was designed to address the diagnostic accuracy of 18F-FDG PET/CT in the follow-up of CRC in patients with normal or elevated CEA. Methods: We retrospectively analyzed 50 patients previously diagnosed with CRC who were initially surgically treated. All patients were suspicious of recurrence and were referred to as 18F-FDG PET/CT for restaging between February 2014 and February 2019. Possible recurrence was indicated by rising CEA, equivocal radiological findings or clinical findings. Results: Out of a total of 50 patients for whom the follow-up of at least six months was available, 27 had CRC confirmed with the gold standard, and all 27 patients had 18F-FDG PET/CT positive for recurrence, giving a sensitivity of 18F-FDG PET/CT in detecting the recurrence of CRC of 100.0% (0.0% of false-negative – FN results). Out of 23 patients with no signs of CRC recurrence on the gold standard, 19 were also 18F-FDG PET/CT negative, giving a specificity of 18F-FDG PET/CT in detecting the recurrence of CRC of 82.6%, and 17.4% of false-positive – FP results. Out of 31 patients who were 18F-FDG PET/CT positive, 27 had it confirmed pathophysiologically or clinically, giving positive predictive value (PPV) of 18F-FDG PET/CT in detecting CRC recurrence of 87.1%; negative predictive value (NPV) was 100.0%, meaning all 19 patients showing no signs of CRC recurrence when imaged with 18F-FDG PET/CT were gold standard negative as well. Conclusion: 18F-FDG PET/CT proves to be a valid diagnostic tool in detecting recurrence in patients with CRC.

Introduction: Peritoneal dialysis (PD) is an established treatment for patients with end-srage kidney disease. The method was developed as an alternative to hemodialysis (HD) presenting a patient survival rate equivalent to HD and better preservation of residual renal function. Peritoneal dialysis (PD) patients have different peritoneal membrane permeability (transport) characteristics. High peritoneal membrane permeability is associated with increased mortality risk in the patient population. Aim: The aim of this study was to analyze the importance of the peritoneal membrane transport status in patients treated with continuous ambulatory peritoneal dialysis (CAPD). Methods: The study included 60 adult continuous ambulatory peritoneal dialysis (CAPD) patients, 29 male and 31 female, mean age 56.63±15.06 years. All patients treated with conventional glucose- based PD fluids. For the short term (within 3 month after start of PD) and long term (more than 12 months) peritoneal dialysis analysis of peritoneal transport characteristics has been used peritoneal functional test (PFT). With the test, categorisation of patients was possible into high (H), high-average (HA), low average (LA), and low (L) transporters. Results: Dialysate–to plasma ratio (D/P) of creatinine showed significantly increased over time (0.654±0.141 vs... 0.705±0.13, p<0.001). In multivariate analysis age, gender, time on dialysis, comorbid diseases, diabetes mellitus (DM), serum albumin, were considered as independent factors influencing the PFT. The high transporter group had higher D/Pcreat (H 0.84±0.03 vs... LA 0.57±0.05, p<0.001), higher proportion of man (H 100% vs... LA 39.5%, p<0.05), higher proportion of patients with comorbid diseases (H 60% vs... LA 20.9%, p<0.05), lower serum albumin concentration (H 29±6.0 vs... LA 37±5.2, p<0.001), lower D4/D0 glucose ( H 0.23±0.07 vs... LA 0.42±0.14, p<0.001), and lower drained volume (H 600±173 vs... LA 1016±355, p<0.001). Conclusion: The PFT was en easy, inexpensive, reliable test to assess peritoneal transport type and it also provided information about peritoneal clearance of solutes and ultrafiltration. Peritoneal transport type classification was recognized not only as aid for prescription, but also as a prognostic index.

Introduction: Idiopathic Idiopathic membranous nephropathy (iMN) is an immune-complex mediated renal disease which is usually associated with the nephrotic syndrome (NS). The course of the disease is variable. Some patients maintain normal kidney function with or without a spontaneous remission of proteinuria, while others progress to end-stage renal failure or die from complications related to the nephrotic syndrome. Whether or not to treat a patient with idiopathic membranous nephropathy is still controversial. The controversy is mainly related to the toxicity of the therapy and the variable natural course of the disease-spontaneous remission occurs in 40–50% of patients. Aim: The aim of this study was to describe our experience of treatment of an idiopathic membranous nephropathy (iMN), efficacy and complications rate. Case report: Our patient was older, mail gender, in high-risk group with persistent proteinuria 10,68 g/day and stable renal function. We have taken these factors into consideration, along with age and other comorbidities, that may significantly elevate the risk of treatment. We chose to start with early treatment, following the Ponticelli’s group protocol based on high dose corticosteroids (odd months) alternating with clorambucil (even months) for six months. This treatment was accompanied by the steroid side effects, including hyperglycaemia dependance on insulin therapy and pulmonary thromboembolism despite administered prophylactically low molecular weight heparin. The six-month treatment was successfully completed with the reduction of proteinuria to nephritic values 2,86 g/day, despite many complications. Complete remission of the disease with non-significant proteinuria and with stable renal function was achieved in 14 months which has been maintained for 2 years. Conclusion: We suggest that decisions on the timing of start of therapy, whom to treat, best sequence of the use of the various immunosuppressive drugs must be based on an individualized assessment of risks and benefits.

Introduction: Psoriasis is a common chronic skin disorder characterized by inflammation and abnormal epidermal proliferation. Its severity ranges from a chronic plaque psoriasis (CPP) to generalized psoriatic erythroderma (PE). The cause of psoriasis is unknown although most evidence supports the hypothesis that psoriasis is an immunologically mediated disease. The T-helper (Th) 1 and Th17 cells are responsible for the inflammation of psoriasis. Immunoglobulin E (IgE) is a class of immunoglobulin essential for the allergic response. There is some evidence that IgE may take a part in the pathogenesis of psoriasis. Aim: The aim of the study was to compare serum levels of total IgE between patients with psoriasis and healthy subjects, and to assess the difference between localized form (CPP) and extensive form of disease (PE). Methods: Seventy-five patients with psoriasis and 30 healthy subjects were enrolled in this study. Data on age, gender, personal and family history, clinical type and duration of disease were collected and analyzed. Serum levels of IgE were measured using nephelometric method. Results: Serum levels of total IgE were significantly higher in patients than in controls (46.7% vs.. 10%; p<0.05). Statistical difference of IgE concentration was also observed between CPP and PE. Comparison between patients and controls with regard to the median of the serum level of total IgE levels showed a statistically highly significant elevation in patients (425 IU/ml) compared with controls (54.5 IU/ml) (p<0,05). A higher total IgE concentration was observed in the group of patients with a longer period of skin changes. No relation was found between the serum level of IgE and family history of psoriasis, age or sex (p>0.05). Conclusions: This study supports the evidence that elevation of total serum IgE is associated with psoriasis. The exact role of serum IgE in psoriasis should be additionally investigated in future studies.

Objectives The development of third molars can be helpful in dental age estimation of adolescents and in early adult period. We tested the repeatability and accuracy of the three dental age radiographic methods (Olze, Demirjian and Solari and Abramovitch) and evaluated which method is more useful. We also aimed at testing to find the correlation of estimated dental and chronological age by these three methods. Material and methods The orthopantomographs (OPGs) of 1007 individuals (8 - 25 years) were divided into two groups (cca 500 OPGs) - one group of OPGs has been presented with all four third molars, while another one was registered with third molar/s hypodontia. And all of OPGs were assessed, to verify the three methods (Olze, Demirjian and Solari and Abramovitch) for age estimation based on third molar development. Results There was a high Spearman's correlation coefficient between stages of development of wisdom tooth and chronological age of subjects by all these three methods. Conclusion We may recommend using third molars for assessing the dental age by Olze, Demirjian and Solari and Abramovitch dental method as well, on Bosnian and Herzegovinian population.

A questionnaire on biomarker testing previously used in central European countries was extended and distributed in Western and Central European countries to the pathologists participating at the Pulmonary Pathology Society meeting 26–28 June 2019 in Dubrovnik, Croatia. Each country was represented by one responder. For recent biomarkers the availability and reimbursement of diagnoses of molecular alterations in non-small cell lung carcinoma varies widely between different, also western European, countries. Reimbursement of such assessments varies widely between unavailability and payments by the health care system or even pharmaceutical companies. The support for testing from alternative sources, such as the pharmaceutical industry, is no doubt partly compensating for the lack of public health system support, but it is not a viable or long‐term solution. Ideally, a structured access to testing and reimbursement should be the aim in order to provide patients with appropriate therapeutic options. As biomarker enabled therapies deliver a 50% better probability of outcome success, improved and unbiased reimbursement remains a major challenge for the future.

This paper investigates the influence of user dynamics on small-scale fading characteristics in off-body channels. A statistical channel model employing a mobility model for wearable antennas on dynamic users is presented. The model is used to analyze the effects of both user’s linear forward motion and periodic wearable antenna displacement due to changes in posture, for walking and running. The time-variant Doppler shift exhibited by signals arriving from different directions is considered for wearable antennas on the chest, wrist and lower leg. A strong influence of the on-body placement is observed, with antennas on the arms and legs yielding up to a 2.2 times higher maximum Doppler frequency than with the antenna on the torso. Furthermore, the small-scale fading dynamics for these antennas vary considerably during the motion cycle, where one can distinguish between slow and fast phases. The latter is found to yield up to 4 times as high root-mean-square level-crossing rates than the former, in the case of the antenna on the lower leg. These findings imply an important effect of the periodic antenna displacement on fading dynamics in off-body channels, which is typically neglected in literature.

This paper presents an off-body indoor propagation channel model. The model is based on a Friis transmission equation to which it adds the orientation-dependent body shadowing loss by adopting a basic model derived from a cosine function. Furthermore, the developed propagation model accounts for human body dynamics and different material reflections coefficients, both implemented using a random variables with uniform distribution. Created model was evaluated against a set of measurements conducted in the indoor environment at mmWave frequency (60 GHz) gathered from the literature, showing a good fit. The difference between the slopes of the proposed model and measurements is less than 6%.