In this paper, we present an implementation and analysis of the mean shift algorithm. The mean shift is a general non-parametric mode finding/clustering procedure widely used in image processing and analysis and computer vision techniques such as image denoising, image segmentation, motion tracking, etc.

Objectives: Anaerobic bacteria may cause numerous infections in different locations through human body. Those infections can be life-threatening with significant mortality. Wounds represent a suitable habitat for colonization of anaerobic bacteria. Their proliferation contributes to moist and warm environment, hypoxic and necrotic tissue.Methods:A retrospective study was conducted at the Clinical Centre University of Sarajevo from 2015-2017. The study involved wound swab samples, sampled from hospitalized patients. The anaerobic bacteria were isolated using standard procedures.Results: During the period from 01.01.2015. to 31.12.2017, 8386 samples were analyzed on anaerobic bacteria and 872 (10.4%) of specimen were positive. In 2015, 332 (15%) specimens were positive, while during 2016 and 2017, 244 (7,8%) and 296 (9.9%) respectively. Bacteroides spp. was the most common isolate during three year period: 2015-227 (55.5%); 2016-139 (48%); 2017-161 (42,5%). It was followed by Peptococcus spp.: 2015-70 (17.1%); 2016-40 (13.9%); 2017-66 (17.4%), Clostridium spp.: 2015 – 32 (7.8%); 2016-21 (7.3%); 2017- 35 (9.2%), Fusobacterium spp.: 2015 – 49 (11.9%); 2016-32 (11.1%); 2017- 45 (11.9%).VITEK 2 Compact has identified to the level of species 48 isolates which were in pure culture.The largest number of anaerobic bacteria were isolated from the samples received from the Abdominal surgery. The overview of antimicrobial sensitivity showed highest sensitivity to metronidazole (99,9%) and carbapenems (99,9%), respectively.Conclusions The most commonly isolated anaerobic bacteria was Bacteroides spp.Highest number of positive isolates was from abdominal surgery since intra-abdominal infections reflect the microflora of the resected organ. Metronidazole remains the antibiotic of choice in the treatment of anaerobic infections.

Introduction: Intensive unit microflora mainly consists of organism capable of surviving in moist media, such as gram-negative bacteria, skin-colonizing microorganisms, those with the ability to adhere to medical devices, and microorganisms resistant to conventional antibiotics. Therefore, cleansing and disinfection of intensive care units is of great importance in the prevention and control of hospital infections.Material and Methods: The use of the device was demonstrated in the isolation room of the intensive care unit after a patient colonized with hospital bacterial strains was discharged. The first sampling was carried out immediately after the patient was discharged, the second after the standard medical cleansing of the equipment and space, and the third after the disinfection with the “SterisafePro”. The analysis of the smears was performed at the OU Clinical Microbiology. Quantitative method analyses was performed according to standard operative procedure (SOP). The results of the analysis are calculated according to the formula and expressed in the values of CFU / cm2.Results: After the patient was discharged and the area cleaned mechanically, the Acinetobacter baumanii was isolated in three samples, coagulase negative staphylococci in one, while two smears remained sterile. Acinetobacter baumanii and coagulase negative staphylococci were isolated in three samples. After using the disinfecting device, all swabs were sterile.Conclusion: Disinfection of the hospital with the “Sterisafe”Pro” device has proved to be very successful. The advantages of using the “Sterisafe”Pro” device are that is uses no chemicals, has a low labor and usage costs, is harmless to the patients and staff, and is very easy to use. 

Background: Healthcare professionals, including medical and dental students, are at high risk of acquiring hepatitis B infection.Aim: The aim of this study was to examine and compare the knowledge and attitudes of the students of medicine and dental medicine of Faculty of Medicine, University of Mostar, about hepatitis B. Subjects and Methods: The examinees in this study were students of the first and second grade of medical studies and studies of dental medicine. The sample included 105 (71.4%) students of medicine and 42 (28.6%) students of dental medicine. The study was conducted at the Faculty of Medicine, University of Mostar, through the attached questionnaire. Results: Overall, the medical students and dental medicine students showed good knowledge about hepatitis B. Medical students showed much more positive attitudes towards hepatitis B positive patients than dental medicine students. Dental medicine students are more concerned about possible infections and general interactions with infected patients, and would statistically significantly reject to provide healthcare services to hepatitis B positive patients compared to the medical students.Conclusion: It was established that dental medicine students showed a little bit more understanding of the transmission patterns and symptoms of hepatitis B than the medical students. By contrast, medical students showed much more positive attitudes towards patients with hepatitis B than dental medicine students. 

Introduction: Infectious mononucleosis is one of the most common syndromes in our clinical practice. It is characterized by elevated temperature, pharyngitis and generalized lymphadenopathy.Objectives: To describe a clinical charachteristics of infectious mononucleosis (IM) caused by the Epstein Barr virus (EBV) in splenectomized patients since in the literature we found insufficient data.Subjects and Methods: Retrospective analysis of medical documentation of the patient treated in the Clinic for Infectious Diseases of the University Clinical Center of Republic of Srpska.Results: We have described the case of infectious mononucleosis, caused by Epstein Barr virus in a splenectomized patient. In support of acute EBV viral infection were the presence of lymphadenopathy, pharyngitis, hepatomegaly, as well as the occurrence of a typical generalized maculopapulous rash, and positive ELISA EBV VCA IgM and anti EBV VCA / EA IgG, were reported. According to the aforementioned patient, it fulfilled most criteria for setting the diagnosis of acute EBV infectious mononucleosis. Our patient showed some atypical signs such as absence of fever during hospitalization, marked leukocytosis with lymphocytosis (with a maximum increase in leukocyte count at 37.3 x 109, in the differential blood sample dominated lymphocytes with 29.96 x 109 (reference values 1.1-3.35), i.e. 80.3% (Ref. 20.0-46.0). Due to the maintenance of leukocytosis with lymphocytosis, the range of clinical has been extended trials (US abdomen, US neck, CT neck, chest, abdomen and pelvis, sternal puncture, hematologists' consultation). Based on the aforementioned hematologists' consultation, and post-release and recovery monitoring it is concluded that there has been no sign of acute hematologic disease but it has been just EBV-IM.Conclusion: Because of insufficient data on clinical presentation of Epstein Barr viral infections in splenectomized this we believe that this is one of the clinical variants although the possibility of individual variation cannot be excluded. Briefly, we can conclude that the immune system in the splenectomized patient can greatly modify the clinical presentation of Epstein barr viral infection, with the pathogenic mechanism that are still unclear.. However, due to the variety of clinical syndromes and the oncogenic potency of the Epstein-Barr virus, we should be extremely cautious and sometimes expand the diagnostic range beyond conventional examinations. 

Construction materials have a direct impact on the environment, on people, and their health. In addition, building insulation plays a decisive role in terms of energy consumption of buildings and regarding CO2-emissions over their whole life cycle. In order to achieve a holistic concept for green building worldwide, it is necessary to develop ecological insulating materials and to scientifically examine them in terms of their technical properties, as done with particleboards from agricultural waste presented in this article. This study aims to characterize the properties’ tensile and compressive strength, modulus of rupture (MOR), and elasticity (MOE) and thermal conductivity of particleboards affected by parameters, such as waste type (rice straw or flax shives), particleboard density, resin type, and content, as well as the use of treated rice straw. Particleboards made from flax shives had superior properties compared to the rice straw particles. The mechanical properties of the boards increase with an increasing resin content, except for the MOR and MOE, which decrease with an increasing resin content, and reach their peak value at a resin content of 10%.

Abstract Background: Drug-drug interactions are defined as modifications of the drug action that result from the simultaneous administration of another individual drug or several drugs. Nowadays, potential drug-drug interactions (DDIs) are most frequently detected and analyzed using personal digital assistant software programs (online interaction checker tools). Objective: To determine the risk factors for the emergence of all drug-drug interactions in surgical patients with particular emphasis on clinically significant interactions. Patients and methods: This was a retrospective cohort analysis of patients treated at the Surgical Clinic of the Clinical Center Kragujevac. Three interaction checkers were used to reveal drug-drug interactions: Medscape, Epocrates and Micromedex. Results: The study included total of 200 patients, aged 58.54±17.08 years. Average number of drug-drug interactions per patient was between 10.50±9.10 (Micromedex) and 18.75±17.14 (Epocrates). Number of prescribed drugs, antidepressive therapy, antiarrhythmic therapy, number of pharmacological/therapeutic subgroups (2nd level of ATC classification) prescribed, delirium or dementia, diabetes, heart failure, and number of physicians who prescribed drugs to single patient were identified as risk factors for drug-drug interactions while length of hospitalization in days and age of patient in years emerged as protective factors. Conclusion: Drug-drug interactions are relatively common in surgical patients and predisposed by factors such as number of prescribed drugs or drug group per patient, number of physicians who prescribed drugs, antidepressive therapy, antiarrhythmic therapy, presence of delirium or dementia, diabetes and heart failure. On the other hand, prolonged hospitalization and higher age are factors that reduce the risk of interactions in surgical patients.

This paper presents performance analysis of an adaptive peak cancellation (PC) method to reduce the high peak-to-average power ratio (PAPR) for OFDM systems, while keeping the out-of-band (OoB) power leakage as well as an in-band distortion power below the pre-determined level. In this work, the increase of adjacent leakage power ratio (ACLR) and error vector magnitude (EVM) are estimated recursively using the detected peak amplitude. We present analytical framework for OFDM-based systems with theoretical bit error rate (BER) representations and detection of optimum peak threshold based on predefined EVM and ACLR requirements. Moreover, the optimum peak detection threshold is selected based on theoretical design to maintain the pre-defined distortion level. Thus, their degradations are restricted below the pre-defined levels which correspond to target OoB radiation. We also discuss the practical design of peak-cancellation signal with target OoB radiation and in-band distortion through optimizing the windowing size of the PC signal. Numerical results show the improvements with respect to both achievable BER and PAPR with the PC method in eigen-beam space division multiplexing (E-SDM) systems under restriction of OoB power radiation. It can also be seen that the theoretical BER shows good agreements with simulation results.

Introduction: Although many predictive tools have already been developed, efforts are still proceeding to identify a reliable biomarker to predict the prognosis of the patients with acute heart disorders. Objectives: The aim was to evaluate the role of renal injury biomarkers (serum cystatin C, serum and urine interleukin-18, IL-18) and heart failure biomarkers (plasma B-type natriuretic peptide, BNP) in the prediction of the postdischarge requirement of renal replacement therapy (RRT) and/or 6-month mortality in patients with acute heart disorders. Patients and Methods: In patients diagnosed with acute heart disorders (acute heart failure [AHF] and/or acute coronary syndrome [ACS]) and admitted to the intensive care units, baseline clinical parameters, renal and cardiac biomarkers were determined. Patients were followed up for 6 months. The composite outcome was the postdischarge requirement of RRT and/or 6-month mortality. Results: Of 120 patients, 5.8% continued RRT after discharge. The 6-month mortality was 20%. Cox logistic regression analysis showed that urine IL-18 (P=0.021), plasma BNP (P=0.046), Acute Physiology and Chronic Health Evaluation (APACHE) II score (P=0.002), and left ventricular diastolic dysfunction (P=0.045) were independent predictors of the postdischarge requirement of RRT and/or 6-month mortality. For predicting RRT and/or 6-month mortality, using urine IL-18 cutoff value of 29.1 pg/mL showed 66.7% sensitivity and 67.7% specificity (area under the curve, AUC 0.70, P=0.003), while using plasma BNP cutoff value of 881.6 pg/mL showed 66.7% sensitivity and 70.8% specificity (AUC 0.76, P<0.001). Conclusion: Urine IL-18 and plasma BNP are independently predictive for the postdischarge requirement of RRT and/or 6-month mortality in patients with acute heart disorders.

Artificial neural networks (ANNs) are notoriously power- and time-consuming when implemented on conventional von Neumann computing systems. Recent years have seen an emergence of research in hardware that strives to break the bottleneck of von Neumann architecture and optimise the data flow; namely to bring memory and computing closer together. One of the most often suggested solutions is the physical implementation of ANNs in which their synaptic weights are realised with analogue resistive devices, such as resistive random-access memory (RRAM). However, various device- and system-level non-idealities usually prevent these physical implementations from achieving high inference accuracy. We suggest applying a well-known concept in computer science -- committee machine (CM) -- in the context of RRAM-based neural networks. Using simulations and experimental data from three different types of RRAM devices, we show that CMs employing ensemble averaging can successfully increase inference accuracy in physically implemented neural networks that suffer from faulty devices, programming non-linearities, random telegraph noise, cycle-to-cycle variability and line resistance. Importantly, we show that the accuracy can be improved even without increasing the number of devices.