error modelling
Dotaz
Zobrazit nápovědu
Nízká pravdivost a nedostatečná srovnatelnost výsledků měření patří mezi chronické problémy laboratorní medicíny a může být jedním z faktorů rizika zdravotní péče o pacienty. Situace by měla být teoreticky řešena požadavkem na návaznost kalibrátorů, obsaženým ve Směrnici 98/79 IVD. Tento požadavek je však podle mého názoru plněn v řadě případů formálně a výrobcům uniká často smysl návaznosti jako nástroje zvyšování srovnatelnosti a pravdivosti měření. Za této situace je nutno počítat se signifi kantními hodnotami bias jako se součástí denní praxe činnosti klinických laboratoří a považovat je za významný zdroj dílčí nejistoty výsledků měření. Ukazujeme na několika příkladech, že hodnoty nejistot se započtenými hodnotami bias se dramaticky neliší od klasicky řadu let používaných hodnot celkových chyb měření.
Low value of trueness and insufficient level of comparability and interchangeability belongs to main problems of laboratory medicine and can be one of many risks of patients care. Theoretically is situation resolved by requirements on the traceability of calibrators, introduced in Directive IVD 98/79 EC. Practically it seems that reason of traceability as tool for elevation of level in comparability is not for many producers clear. Often only formal documentation is taken in account seriously. In these situation is necessary to calculate that signifi cant bias is part of daily work in clinical laboratories and its exclusion from uncertainty estamation is illusion. We show on some causes that uncertainty values with included bias are not signifi cantly different from classically used total error values.
... 24 -- 4.4 What is a model? 25 -- 4.5 What is dose-response modelling? ... ... 27 -- 4.6 Risk versus safety in dose-response modelling 31 -- 4.7 Summary 33 -- 5. ... ... 49 -- 6.2 Models and distributions 51 -- 6.2.1 Dose-response models 51 -- 6.2.1.1 Continuous dose-response ... ... models 51 -- 6.2.1.2 Quantal dose-response models 53 -- 6.2.1.3 Thresholds 53 -- 6.2.1.4 Severity (degree ... ... Sampling error 65 -- 6.5.2 Study error 66 -- 6.5.3 Model error 66 -- 6.6 Benchmark dose and benchmark ...
Environmental health criteria, ISSN 0250-863X 239
xx, 137 s. : il., tab. ; 21 cm
Je známo, že pro analýzu reliability testů, které se užívají pro měření v psychologické diagnostice, je k dispozici velice silná a podrobně rozpracovaná metodologie. V diagnostice se však vyskytují i rozličné klasifikační procedury, kdy observovaná proměnná má pouze nominální charakter. U takových klasifikací se informace o jejich reliabilitě vyskytuje velice vzácně. Jedním z důvodů je samozřejmě fakt, že pro nominální klasifikace není k dispozici tak vypracovaná metodologie, jako u testového měření. Analýza spolehlivosti je v těchto případech většinou založena na Cohen-Fleissově kappa koncepci, tedy analýze konkordance při dvou, nebo více replikacích klasifikace. Cílem této stati je prezentace nové, netradiční metody analýzy reliability klasifikačních procedur, jejímž základem je pravděpodobnostní model chyby v klasifikaci. Tento model je založen na analogickém principu jako „true-error“ model klasické teorie testů.
It is known that there is a very powerful and detailed methodology for a reliability analysis of tests that are used for the measurement in psychological diagnostics. However, there are also various diagnostic classification procedures in which observed variable is measured only on a nominal scale. For such classifications, the information about their reliability is available very rarely. Of course, one reason for this is the fact that there is no sufficiently developed methodology for the nominal classification, unlike for the test measurements. The analysis of reliability in these cases is usually based on the Cohen-Fleiss kappa conception, i.e. the concordance analysis of two or more replications of classification. The aim of this paper is to present new, innovative methods of the analysis of reliability of classification procedures, based on a probabilistic model of error in classification. This model is based on analogous principles as the „true-error“ model of classical test theory.
- MeSH
- lidé MeSH
- matematika MeSH
- psychologie * MeSH
- statistika jako téma * MeSH
- výzkum * MeSH
- Check Tag
- lidé MeSH
Supervised learning of perceptron networks is investigated as an optimization problem. It is shown that both the theoretical and the empirical error functionals achieve minima over sets of functions computable by networks with a given number n of perceptrons. Upper bounds on rates of convergence of these minima with n increasing are derived. The bounds depend on a certain regularity of training data expressed in terms of variational norms of functions interpolating the data (in the case of the empirical error) and the regression function (in the case of the expected error). Dependence of this type of regularity on dimensionality and on magnitudes of partial derivatives is investigated. Conditions on the data, which guarantee that a good approximation of global minima of error functionals can be achieved using networks with a limited complexity, are derived. The conditions are in terms of oscillatory behavior of the data measured by the product of a function of the number of variables d, which is decreasing exponentially fast, and the maximum of the magnitudes of the squares of the L(1)-norms of the iterated partial derivatives of the order d of the regression function or some function, which interpolates the sample of the data. The results are illustrated by examples of data with small and high regularity constructed using Boolean functions and the gaussian function.
Monographs on Statistics and Applied Probability ; 63
305 s.
Path integration is thought to rely on vestibular and proprioceptive cues yet most studies in humans involve primarily visual input, providing limited insight into their respective contributions. We developed a paradigm involving walking in an omnidirectional treadmill in which participants were guided on two sides of a triangle and then found their back way to origin. In Experiment 1, we tested a range of different triangle types while keeping the distance of the unguided side constant to determine the influence of spatial geometry. Participants overshot the angle they needed to turn and undershot the distance they needed to walk, with no consistent effect of triangle type. In Experiment 2, we manipulated distance while keeping angle constant to determine how path integration operated over both shorter and longer distances. Participants underestimated the distance they needed to walk to the origin, with error increasing as a function of the walked distance. To attempt to account for our findings, we developed configural-based computational models involving vector addition, the second of which included terms for the influence of past trials on the current one. We compared against a previously developed configural model of human path integration, the Encoding-Error model. We found that the vector addition models captured the tendency of participants to under-encode guided sides of the triangles and an influence of past trials on current trials. Together, our findings expand our understanding of body-based contributions to human path integration, further suggesting the value of vector addition models in understanding these important components of human navigation.
- MeSH
- chůze fyziologie MeSH
- dospělí MeSH
- lidé MeSH
- orientace fyziologie MeSH
- podněty MeSH
- propriocepce fyziologie MeSH
- prostorová navigace fyziologie MeSH
- teoretické modely MeSH
- vnímání prostoru fyziologie MeSH
- výpočetní biologie metody MeSH
- Check Tag
- dospělí MeSH
- lidé MeSH
- mužské pohlaví MeSH
- ženské pohlaví MeSH
- Publikační typ
- časopisecké články MeSH
- Research Support, U.S. Gov't, Non-P.H.S. MeSH
This paper focuses on wrapper-based feature selection for a 1-nearest neighbor classifier. We consider in particular the case of a small sample size with a few hundred instances, which is common in biomedical applications. We propose a technique for calculating the complete bootstrap for a 1-nearest-neighbor classifier (i.e., averaging over all desired test/train partitions of the data). The complete bootstrap and the complete cross-validation error estimate with lower variance are applied as novel selection criteria and are compared with the standard bootstrap and cross-validation in combination with three optimization techniques - sequential forward selection (SFS), binary particle swarm optimization (BPSO) and simplified social impact theory based optimization (SSITO). The experimental comparison based on ten datasets draws the following conclusions: for all three search methods examined here, the complete criteria are a significantly better choice than standard 2-fold cross-validation, 10-fold cross-validation and bootstrap with 50 trials irrespective of the selected output number of iterations. All the complete criterion-based 1NN wrappers with SFS search performed better than the widely-used FILTER and SIMBA methods. We also demonstrate the benefits and properties of our approaches on an important and novel real-world application of automatic detection of the subthalamic nucleus.
- MeSH
- teoretické modely MeSH
- velikost vzorku * MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH
- validační studie MeSH
Terapeutické monitorování léků umožňuje úpravu dávkování u pacienta na základě stanovení koncentrace léčiva v krvi. Software pro farmakokinetické modelování umožňuje predikovat průběh koncentrací při pravidelném i nepravidelném dávkování, při nestandardních odběrových časech, před dosažením ustáleného stavu nebo při měnící se funkci eliminačních orgánů, zejména ledvin. Příčinou neočekávaných výsledků může být chyba odběru (při odběru ze stejného místa, jako je léčivo aplikováno, nebo odběru z centrálního žilního katétru), neúplná léková anamnéza při změnách dávkování před odběrem, chybějící informace o použití nasycovací dávky, nebo léková interakce. Specifickým případem je podávání léčiva dialyzovaným pacientům, kdy při intermitentní dialýze je nezbytná výrazná redukce dávek, zatímco u kontinuální dialýzy je dávkování úměrné hodnotám sérového kreatininu. Ke správné interpretaci výsledků jsou nezbytné správné a úplné vstupní údaje pacienta a zejména správná dávková anamnéza.
Therapeutic drug monitoring allows the adjustment of the patient's dosage based on the drug concentration. Pharmacokinetic modelling software enables to predict the concentrations with regular and irregular dosing, with non-standard sampling times, before reaching a steady state or with changing function of elimination organs, in particular kidneys. Unexpected results can be caused by a sampling error (esp. when sample was taken from the same site as the drug was administered, or sampling from a central venous catheter), an incomplete drug history during dosage changes before sampling, missing information on the use of a loading dose, or a drug interaction. A specific case is the administration of the drug to dialysis patients, where a significant dose reduction is necessary in intermittent dialysis, while in continuous dialysis the dosage is proportional to serum creatinine values. To correctly interpret the results, correct and complete input data of the patient and, in particular, a full dose history are necessary.
- Klíčová slova
- farmakokinetická analýza,
- MeSH
- biomarkery farmakologické analýza krev MeSH
- farmakokinetika MeSH
- léčivé přípravky analýza aplikace a dávkování krev MeSH
- lidé MeSH
- monitorování léčiv * metody přístrojové vybavení statistika a číselné údaje MeSH
- odběr biologického vzorku MeSH
- Check Tag
- lidé MeSH
- Publikační typ
- kazuistiky MeSH
Correlated ab initio calculations on large systems, such as the popular MP2 (or RI-MP2) method, suffer from the intramolecular basis set superposition error (BSSE). This error is typically manifested in molecules with folded structures, characterized by intramolecular dispersion interactions. It can dramatically affect the energy differences between various conformers as well as intramolecular stabilities, and it can even impair the accuracy of the predictions of the equilibrium molecular structures. In this study, we will present two extreme cases of intramolecular BSSE, the internal stability of [n]helicene molecules and the relative energies of various conformers of phenylalanyl-glycyl-phenylalanine tripeptide (Phe-Gly-Phe), and compare the calculated data with benchmark values (experimental or high-level theoretical data). As a practical and cheap solution to the accurate treatment of the systems with large anticipated value of intramolecular BSSE, the recently developed density functional method augmented with an empirical dispersion term (DFT-D) is proposed and shown to provide very good results in both of the above described representative cases. (c) 2007 Wiley Periodicals, Inc.
OBJECTIVES: This article summarized errors obtained by diverse techniques used for the derivation of cross-sectional contours in nonadult humeri and tibiae. MATERIALS AND METHODS: We analyzed cross-sectional contours in a total sample of 62 humeral and 75 tibial diaphyses in the age between birth and 12 years divided into three age groups. Long bone 35% (humeri) and midshaft (tibiae) cross-sections were taken on micro-CT images and analyzed by EPJMacro in FIJI. Properties were extracted from contours derived by manual, automatic, spline, and ellipse techniques. Agreement between techniques was assessed using manually extracted properties such as the true value using percent prediction error (%PE), reduced major axis regression, and ±95% limits of agreement. RESULTS: The lowest measurement errors were obtained for total areas, moderate errors for cortical areas and section moduli, and the highest errors for medullary areas for both bones. Derivation of humeral nonadult cross-sectional properties is less sensitive to the technique used for derivation of periosteal and endosteal contours, reaching mean %PEs below 5%. In contrast, tibial nonadult cross-sectional properties are more sensitive to the technique used and exceed 5% for some combinations. DISCUSSION: Automatic techniques provide reasonably high agreement with manually extracted contours for nonadult humeri but low agreement for tibiae. Semiautomatic approaches-spline and ellipse techniques-may reduce the error for all studied properties in tibiae, especially when combined with manually traced periosteal contours. The positive effect of the semiautomatic technique on measurement error is low for humeri.
- MeSH
- anatomie průřezová MeSH
- antropologie fyzická MeSH
- antropometrie MeSH
- dítě MeSH
- humerus * anatomie a histologie diagnostické zobrazování MeSH
- kojenec MeSH
- lidé MeSH
- novorozenec MeSH
- okostice * anatomie a histologie diagnostické zobrazování MeSH
- předškolní dítě MeSH
- rentgenová mikrotomografie MeSH
- reprodukovatelnost výsledků MeSH
- statistické modely MeSH
- tibie * anatomie a histologie diagnostické zobrazování MeSH
- Check Tag
- dítě MeSH
- kojenec MeSH
- lidé MeSH
- novorozenec MeSH
- předškolní dítě MeSH
- Publikační typ
- časopisecké články MeSH
- práce podpořená grantem MeSH