Jernej Završnik

Decision trees based on automatic learning and their use in cardiology

Computerized information systems, especially decision support systems, have become an increasingly important role in medical applications, particularly in those where important decision must be made effectively and reliably. But the possibility of using computers in medical decision making is limited by many difficulties, including the complexity of conventional computer languages, methodologies and tools. Thus a conceptual simple decision making model with the possibility of automating learning should be used. In this paper we introduce a cardiological knowledge-based system based on the decision tree approach supporting the mitral valve prolapse determination. Prolapse is defined as the displacement of a bodily part from its normal position. The term mitral valve prolaps (PMV), therefore, implies that the mitral leaflets are displaced relative to some structure, generally taken to be the mitral annulus. The implications of the PMV are the following: disturbed normal laminar blood flow, turbulence of the blood flow, injury of the chordae tendinae, the possibility of thrombuss composition, bacterial endocarditis, and finally hemodynamic changes defined as mitral insufficiency and mitral regurgitation. Uncertainty persists about how it should be diagnosed and about its clinical importance. It is our deep belief that the echocardiography enables properly trained experts armed with proper criteria to evaluate PMV almost 100%. But unfortunately, there are some problems concerned with the use of echocardiography. In that manner we have decided to start a research project aimed at finding new criteria and enabling the general practitioner to evaluate PMV using conventional methods and to select potential patients from the general population. To empower one to perform needed activities we have developed a computer tool called ROSE (computeRised prOlaps Syndrom dEtermination) based on algorithms of automatic learning. This tool supports the definition of new criteria and the selection of potential PMV-patients.

Impact of Modified Treatment in Echocardiographically Confirmed Pseudo-Pulseless Electrical Activity in Out-of-Hospital Cardiac Arrest Patients with Constant End-Tidal Carbon Dioxide Pressure during Compression Pauses

This study evaluated the ability of focused echocardiography (FE) and capnography to differentiate between pulseless electrical activity (PEA) and pseudo-PEA in out-of-hospital cardiac arrest, and the potential survival benefits with modified treatment. In PEA patients with stable end-tidal carbon dioxide pressure (P etCO2) during the compression pause and concomitant FE showing cardiac kinetic activity, the compression pause was prolonged for 15 s and an additional 20 IU vasopressin was administered. If pulselessness persisted, compressions were continued. Fifteen of the 16 patients studied (94%) achieved restoration of spontaneous circulation (ROSC); eight patients (50%) attained a good neurological outcome (Cerebral Performance Category 1–2). In an historical PEA group with stable P etCO2 values (n = 48), ROSC was achieved in 26 patients (54%); four patients (8%) attained Cerebral Performance Category 1–2. Echocardiographical verification of the pseudo-PEA state enabled additional vasopressor treatment and cessation of chest compressions, and was associated with significantly higher rates of ROSC, survival to discharge and good neurological outcome.

Medical diagnosis prediction using genetic programming

Medical decision making is one of the most demanding tasks in modern medicine. Based on a huge amount of solved cases and rapidly growing number of new findings, medical decision making is supposed to become more and more reliable and effective. However, physicians and medical experts are facing a new arising problem - information overload which they are not able to overcome. A solution might be found in decision support systems with intelligent information processing abilities that help to process a huge amount of data and suggest a possible decision for each new case. Regarding the simplicity of decision trees and effectiveness of evolutionary programming techniques we developed a decision support system based on automatic programming and used it to solve the mitral valve prolapse classification problem.

Citation context and impact of 'sleeping beauties' in paediatric research.

Objectives ‘Sleeping beauties’, i.e. publications that are not cited for a long while, present interesting findings in science. This study analysed the citation trends of sleeping beauties in paediatric research. Methods The study used bibliometric software to analyse the papers citing sleeping beauties in paediatric research, to understand the context in which paediatric sleeping beauties were finally cited and the impact of these sleeping beauties on paediatric research. Results Two paediatric sleeping beauties, addressing medical homes and the transition from paediatric to adult health care, respectively, awakened in response to organizational needs. Both presented novel concepts of paediatric service organization that became important because of an increased need for optimization of services. Conclusion All sleeping beauties bring new knowledge that becomes important only after several years. Paediatric sleeping beauties exhibited unique characteristics; however, their presence in paediatric research shows that knowledge acquisition in paediatrics resembles that in other disciplines.

The application of structured spreadsheet modeling in medicine

Abstract The effective use of computers in medical praxis is limited first by difficulties of medical staff in learning and using conventional design paradigms and second by the variety of medical applications. Only when medical staff can use computers quickly, easily, and in a natural, intuitive fashion will medical computer-supported systems be used enthusiastically and successfully. In the present paper we show that structured spreadsheet modeling, which is a combination of structured modeling and spreadsheet software, can be used effectively and efficiently in the process of designing, testing, and evaluating various classes of medical models.

Biomechanical characteristics of skeletal muscles and associations between running speed and contraction time in 8- to 13-year-old children.

Objective To investigate associations between running speeds and contraction times in 8- to 13-year-old children. Method This longitudinal study analyzed tensiomyographic measurements of vastus lateralis and biceps femoris muscles’ contraction times and maximum running speeds in 107 children (53 boys, 54 girls). Data were evaluated using multiple correspondence analysis. Results A gender difference existed between the vastus lateralis contraction times and running speeds. The running speed was less dependent on vastus lateralis contraction times in boys than in girls. Analysis of biceps femoris contraction times and running speeds revealed that running speeds of boys were much more structurally associated with contraction times than those of girls, for whom the association seemed chaotic. Conclusion Joint category plots showed that contraction times of biceps femoris were associated much more closely with running speed than those of the vastus lateralis muscle. These results provide insight into a new dimension of children’s development.

Tensiomyographic Assessment of Muscle Contractile Properties in 9- to 14-Year Old Children

While there are numerous data on the skeletal muscle fiber type composition in adults, little is known about the changes in fiber type composition and contractile properties during maturational growth in children. Using noninvasive tensiomyography, we measured contraction time (Tc), an indirect estimate of the myosin heavy chain I (MHC-I) proportion, to assess the longitudinal changes of the biceps brachii (BB), biceps femois (BF), vastus lateralis (VL), and erector spinae (ES) muscles in 53 boys and 54 girls. The children were 9 years at the start of the study and returned for 5 follow-up measurements until the age of 14 years. The ES has the shortest and the BF has the longest Tc. The VL and ES of boys have shorter Tc than those of girls. When applying the relationship between proportion of MHC-I and Tc established in adults to childrens TMG data, we found a slow-to-fast transition in the VL between, at least, the ages of 6 to 10 years, when it stabilized to adult proportions. Regular participation in sports was associated with a faster BF, but not in the VL. Our data represents a first non-invasive indication of the developmental changes in muscle fiber type composition in children.

Diagnostic process optimisation: a two levelled approach [paediatric cardiology]

Every-day routine responsibilities of medical staff can be enormous. The appearance of new computer-based information technology has eased these activities enormously and enabled medical staff to perform their work more efficiently and effectively. One possible application of computer technology in paediatric cardiology, where examinations can be very expensive, is the optimisation of the diagnostic process in a manner to minimise the number of examinations, and to reduce costs and the risk to the patients. In this paper, we present an information system which supports diagnostic process optimisation (DIAPRO) and the approach used to design it.<<ETX>>

Tracing and controlling child's development using COCIDIN

Two activities are essential for the successful care of children, namely adequate trace (measurement, recording, trend and pattern calculations) and control (measurement, comparison and possible mediation) of their development. To enable efficient trace and control we should follow as more as possible various development parameters. The intention of this paper is to present the COCIDIN (Computerized cOntinued ChIlds Development INdex) and its design. COCIDIN is a numeric index founded on observations of ten important parameters and can be used as a suitable base for above activities.<<ETX>>

Bibliometric characteristics of predatory journals in pediatrics

To the Editor: In the past decade, scientific publishing was characterized with the open-access model, unfortunately with the side effect of the appearance of the so-called predatory journals. These predatory journals caused a serious problem to the integrity of medical research, due to rising retraction rates, irreproducible results, and a flood of low quality publications(1). Features of predatory journals are an extremely quick, non existing or unclear peer-review process (sometimes less than one week), false credibility of the editorial board, less than five members of the editorial board, lack of or falsification of institutional affiliations, homepages containing spelling errors, bogus impact factors, e-mail spam, small number of published papers and smaller publication and processing fees(2). To help researchers identify predatory journals, some webpages maintain the lists of possible predatory journals(3,4). The field of pediatrics is also a victim of this latest trend. While the archived Beale list(3) contains just two, the BHI Predatory Journal list(4) includes 26 suspicious pediatric journals. Interestingly, the two journals from the Beale list are not included in the BHI list; however, all the journals in the BHI list are published by suspicious publishers contained in the Beale list. None of the suspicious journals are indexed in Medline, but two of them are indexed in Scopus. Most of those journals could be relatively easy to identify using the features enumerated above. Nevertheless, we were interested how less obviously identifiable predatory journals could be distinguished from their non-suspicious counterpart. In that manner we compared two journals included in Beale and BHI lists indexed in Scopus (PJ1 and PJ2) with a pediatric journal (JComp) having a comparable SJR (Scopus Journal Rank), a randomly selected pediatric journal with an average SJR (JAvg) and the first ranked Journal in the Pediatrics, Perinatology and Child Health subject area in Scopus (JFirst) using bibliometric indicators. We analyzed papers published during the period 2013–2017. The search was done on 18 November 2017. We compared the number of published papers, type of papers, the production of ten most prolific countries and five most prolific institutions, the number of countries needed to produce 80% of papers, number of countries and authors, and the size of county co-author networks. The results of the comparison are presented in Table 1. We see that the JFirst and JAvg differs quite clearly from PJ1 and PJ2 in terms of bibliometrics indicators. On the contrary, the JComp falls in somewhere between two above groups, and could belong to each one. Thereafter, we may claim that the main, not yet reported differences between non-suspicious journals (NSJs) and possible predatory journals (PPJs) are: Number of types of papers published. PPJs don’t publish editorials, erratum, and articles in press; The distribution of papers in NSJs is spread to substantially more authors and countries than in PPJs The most prolific NSJs authors and institutions produce much smaller share of total number of papers than most prolific PPJs authors and institutions A considerable number of countries are involved in coauthor networks in NSJs than in PPJs.