A5023924351- Gene expression and cancer classification
- Neonatal Respiratory Health Research
- Single-cell and spatial transcriptomics
- Bioinformatics and Genomic Networks
- Lung Cancer Diagnosis and Treatment
- Cloud Computing and Resource Management
- Cystic Fibrosis Research Advances
- Forensic and Genetic Research
- Cancer-related molecular mechanisms research
- Gastric Cancer Management and Outcomes
- COVID-19 and healthcare impacts
- Chronic Obstructive Pulmonary Disease (COPD) Research
- Erythropoietin and Anemia Treatment
- Pulmonary Hypertension Research and Treatments
- Respiratory Support and Mechanisms
- Ultrasound in Clinical Applications
- Infection Control and Ventilation
- Advanced Data Storage Technologies
- Transplantation: Methods and Outcomes
- Neurogenetic and Muscular Disorders Research
- Data-Driven Disease Surveillance
- Parallel Computing and Optimization Techniques
- Molecular Biology Techniques and Applications
- Census and Population Estimation
- Genetic Associations and Epidemiology
University of Belgrade
2016-2025
Ospedale San Luigi Gonzaga
2019-2021
Ospedale Regina Margherita
2008
When applying data mining or machine learning techniques to large and diverse datasets, it is often necessary construct descriptive predictive models. Descriptive models are used discover relationships between the attributes of while identify characteristics that will be collected in future. Bioinformatics high-dimensional, making practically impossible apply majority "classical" algorithms for classification clustering. Even if useful, training with multidimensional significantly increases...
Surgical resection is the treatment of choice to cure patients with non-small-cell lung cancer (NSCLC); nevertheless, assessment lower limit surgical tolerance remains difficult. Ventilatory inefficiency (measured as ventilation CO(2) production ratio (V'(E)/V'(CO2) slope) a survival predictor in pulmonary hypertension (PH) and chronic heart failure (CHF) considered marker PH obstructive disease (COPD). The aim this study was investigate role V'(E)/V'(CO2) slope preoperative mortality...
Studying the association of gene function, diseases, and regulatory network reconstruction demands data compatibility. Data from different databases follow distinct schemas are accessible in heterogenic ways. Although experiments differ, may still be related to same biological entities. Some entities not strictly biological, such as geolocations habitats or paper references, but they provide a broader context for other The datasets can share similar properties, which found within datasets....
Accurate clinical assessment of liver fibrosis is essential and the aim our study was to compare combine hemodynamic Doppler ultrasonography, stiffness by transient elastography, non-invasive serum biomarkers with degree confirmed biopsy, thereby determine value combining method in prediction significant fibrosis.We included 102 patients chronic disease various etiology. Each patient evaluated using ultrasonography measurements velocity flow pattern at portal trunk, hepatic splenic artery,...
Background/Aim. In recent years, many studies have demonstrated a proximal shift in the distribution of adenomas and colorectal cancers. The aim this study was to investigate whether there are differences incidence anatomical cancers spanning 20 year time gap. Methods. We performed retrospective observational diagnosed during total colonoscopy high volume tertiary care facility two 1-year periods ? 1990 2010. Results. During analyzed period, 4,048 colonoscopies were performed, 1,148 2,900...
Background & Aims: Acute-on-chronic liver failure (ACLF) is characterized by a rapid progression to multiple organ and associated with very high mortality rate of 50-90%. Novel therapies are being investigated such as Erythropoietin (EPO). The aim this prospective cohort study was analyse the value EPO in predicting prognosis determine which patients may benefit most from therapy.
 Methods: According EASL-CLIF criteria, 104 consecutive were diagnosed ACLF, separated into two groups...
<title>Abstract</title> When using data mining or machine learning techniques on large and diverse datasets, it is often necessary to construct descriptive predictive models. Descriptive models are used for discovering relationships among the attributes of while identify characteristics that will be collected in future. Bioinformatics high-dimensional, making practically impossible apply majority "classic" algorithms classification clustering. Even when useful, training with multidimensional...
Single cell transcriptomics measures gene expression data of large number genes, concurrently, from tens thousands cells present in a studied biological sample. It is difficult to obtain good classification results due high dimensionality and variability states. We performed preliminary study assess the feasibility using supervised machine learning methods classify peripheral blood mononuclear (PBMC) types single data. analyzed PBMC set $(\sim 120,000$ cells), selected 47 genes (from 30698...
Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is widely used for nodal staging in lung cancer patients. Ultrasound elastography (UE) a new sonographical analysis that translates tissue stiffness into colour signal. <b>Aim:</b> to assess the value of UE discriminating malignant from non-malignant mediastinal lymph nodes (LNs). <b>Methods:</b> We evaluated 32 LNs 28 patients who underwent EBUS-TBNA Oct 2018 Feb 2019. EBUS indication, location and measurements,...