SEARCH · PubMed Health

Results for “Model interpretability”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,639 records · Page 91Linked to original sources

PubMed2003

Internal dose assessment in radiation accidents.

Although numerous models have been developed for occupational and medical internal dosimetry, they may not be applicable to an accident situation. Published dose coefficients relate effective dose to intake, but if acute deterministic effects are possible, effective dose is not a useful parameter. Consequently, dose rates to the organs of interest need to be computed from first principles. Standard bioassay methods may be used to assess body contents, but, again, the standard models for bioassay interpretation may not be applicable because of the circumstances of the accident and the prompt initiation of decorporation therapy. Examples of modifications to the standard methodologies include adjustment of biological half-times under therapy, such as in the Goiania accident, and the same effect, complicated by continued input from contaminated wounds, in the Hanford 241Am accident.

Americium↗

PubMed1998

Empirical comparisons of proportional hazards, poisson, and logistic regression modeling of occupational cohort data.

This research was conducted to examine the effect of model choice on the epidemiologic interpretation of occupational cohort data. Three multiplicative models commonly employed in the analysis of occupational cohort studies--proportional hazards. Poisson, and logistic regression--were used to analyze data from an historical cohort study of workers exposed to formaldehyde. Samples were taken from this dataset to create a number of predetermined scenarios for comparing the models, varying study size, outcome frequency, strength of risk factors, and follow-up length. The Poisson and proportional hazards models yielded nearly identical relative risk estimates and confidence intervals in all situations except when confounding by age could not be closely controlled in the Poisson analysis. Logistic regression findings were more variable, with risk estimates differing most from the proportional hazards results when there was a common outcome or strong relative risk. The logistic model also provided less precise estimates than the other two. Thus, although logistic was the easiest model to implement, it should be used only in occupational cohort studies when the outcome is rare (5% or less), and the relative risk is less than approximately 2. Even then, the proportional hazards and Poisson models are better choices. Selecting between these two can be based on convenience in most circumstances.

Cohort Studies↗

PubMed2004

A novel approach to predict a toxicological property of aromatic compounds in the Tetrahymena pyriformis.

The TOPological Substructural MOlecular DEsign (TOPS-MODE) has been successfully used in order to explain the toxicity in the Tetrahymena pyriformis on a large data set. The obtained models for the training set had good statistical parameters (R(2)=0.72-0.81, p<0.05) an also the prediction power of the models found was adequate (Q(2)=0.70-0.80). A detailed study of the influence of variable numbers in the equation and the statistical outliers was carried out; leading to a good final model with a better physicochemical interpretation than the rest of the published models. Only two molecular descriptors codifying dipolar and hydrophobic features were introduced. Finally, the fragment contributions to the toxicity prediction evidenced the powerful of this topological approach.

Animals↗

PubMed2005

Parallel hybrid modeling methods for a full-scale cokes wastewater treatment plant.

Parallel hybrid modeling methods are applied to a full-scale cokes wastewater treatment plant. Within the hybrid model structure, a mechanistic model specifies the basic dynamics of the relevant process and a non-parametric model compensates for the inaccuracy of the mechanistic model. First, a simplified mechanistic model is developed based on Activated Sludge Model No. 1 and the specific process knowledge of the cokes wastewater treatment process. Then, the mechanistic model is combined with five different non-parametric models--feedforward back-propagation neural network, radial basis function network, linear partial least squares (PLS), quadratic PLS and neural network PLS (NNPLS)--in parallel configuration. These models are identified with the same data obtained from the plant operation to predict dynamic behavior of the process. The performance of each parallel hybrid model is compared based on their ease of model building, prediction accuracy and interpretability. For this application, the parallel hybrid model with NNPLS as non-parametric model gives better performance than other parallel hybrid models. In addition, the NNPLS model is used to analyze the behavior of the operation data in the reduced space and allows for fault detection and isolation.

Bioreactors↗

PubMed2004

Empirical Bayes estimates generated in a hierarchical summary ROC analysis agreed closely with those of a full Bayesian analysis.

BACKGROUND AND OBJECTIVE: A range of fixed-effect and random-effects meta-analytic methods are available to obtain summary estimates of measures of diagnostic test accuracy. The hierarchical summary receiver operating characteristic (HSROC) model proposed by Rutter and Gatsonis in 2001 represents a general framework for the meta-analysis of diagnostic test studies that allows different parameters to be defined as a fixed effect or random effects within the same model. The Bayesian method used for fitting the model is complex, however, and the model is not widely used. The objective of this report is to show how the model may be fitted using the SAS procedure NLMIXED and to compare the results to the fully Bayesian analysis using an example. METHODS: The HSROC model, its assumptions, and its interpretation are described. The advantages of this model over the usual summary ROC (SROC) regression model are outlined. A complex example is used to compare the estimated SROC curves, expected operating points, and confidence intervals using the alternative approaches to fitting the model. RESULTS: The empirical Bayes estimates obtained using NLMIXED agree closely with those obtained using the fully Bayesian analysis. CONCLUSION: This alternative and more straightforward method for fitting the HSROC model makes the model more accessible to meta-analysts.

Bayes Theorem↗

PubMed2005

Comparison of recent methods for inference of variable influence in neural networks.

Neural networks (NNs) belong to 'black box' models and therefore 'suffer' from interpretation difficulties. Four recent methods inferring variable influence in NNs are compared in this paper. The methods assist the interpretation task during different phases of the modeling procedure. They belong to information theory (ITSS), the Bayesian framework (ARD), the analysis of the network's weights (GIM), and the sequential omission of the variables (SZW). The comparison is based upon artificial and real data sets of differing size, complexity and noise level. The influence of the neural network's size has also been considered. The results provide useful information about the agreement between the methods under different conditions. Generally, SZW and GIM differ from ARD regarding the variable influence, although applied to NNs with similar modeling accuracy, even when larger data sets sizes are used. ITSS produces similar results to SZW and GIM, although suffering more from the 'curse of dimensionality'.

Algorithms↗

PubMed2006

Metabolite projection analysis for fast identification of metabolites in metabonomics. Application in an amiodarone study.

The assignment of significantly changed NMR signals, which were identified with the help of multivariate models, to individual metabolites in biofluids is a manual and tedious task requiring knowledge in chemometrics and NMR spectroscopy. Metabolite projection analysis, introduced in this work, allows automatic linking of multivariate models with metabolites by skipping the level of manual NMR signal identification. The method depends on the projection of sets of metabolite NMR spectra from a database into PCA or PLS models of NMR spectra of biofluid samples. Metabolites that are significantly changed can be identified graphically in metabolite projection plots or numerically as projected virtual concentration. The method is demonstrated together with a newly introduced algorithm for refined nonequidistant binning using a metabonomics study with amiodarone as administered drug. Amiodarone can induce phospholipidosis in the lung and liver, which is accompanied by associated organ toxicity in these organs. It is shown how metabolite projection analysis allows easy and fast tentative assignment of all structures of metabolites whose concentrations in the urine samples significantly changed upon dosage. These metabolites had also been identified previously by manually interpreting the multivariate models and spectra. Among these metabolites, phenylacetylglycine was also identified as being significantly increased. This metabolite has recently been proposed as urinary biomarker for phospholipidosis.

Amiodarone↗

PubMed2003

Smoothed dissipative particle dynamics.

We present a fluid particle model that is both a thermodynamically consistent version of smoothed particle hydrodynamics (SPH) and a version of dissipative particle dynamics (DPD), capturing the best of both methods. The model is a discrete version of Navier-Stokes equations, like SPH, and includes thermal fluctuations, like DPD. This model solves some problems with the physical interpretation of the original DPD model.

Journal Article↗

PubMed1999

Applications in survival analysis.

Survival or failure time traits such as herd life and days open are both important economically and pose a number of challenges to an analysis based on linear mixed models. The main features of a survival trait are that it is the time until some event occurs, and some of the observations are censored. Survival models and the associated estimation procedures provide a flexible means of modeling survival traits. In this paper I will discuss the application of survival analysis based on the Weibull distribution. The components that make up a survival model will be presented along with their interpretation. Issues related to the model construction and estimation will be presented.

Animals↗

PubMed1992

Radiobiological cell survival models. A methodological overview.

A central issue in clinical radiobiological research is the prediction of responses to different radiation qualities. The choice of cell survival and dose-response model greatly influences the results. In this context the relationship between theory and model is emphasized. Generally, the interpretations of experimental data depend on the model. Cell survival models are systematized with respect to their relations to radiobiological theories of cell kill. The growing knowledge of biological, physical, and chemical mechanisms is reflected in the formulation of new models. The present overview shows that recent modelling has been more oriented towards the stochastic fluctuations connected to radiation energy deposition. This implies that the traditional cell survival models ought to be complemented by models of stochastic energy deposition processes and repair processes at the intracellular level.

Cell Survival↗

PubMed1988

Machine learning in quantitative histopathology.

The role of expert systems functioning as process controllers in learning image understanding systems is discussed. Numeric learning systems already have found a number of applications in cytologic and histopathologic diagnosis. Depending on the required capabilities, systems of increasing complexity are needed. Expert systems to guide scene segmentation in histopathologic imagery require model-based reasoning. Diagnostic image interpretation with learning capability demands a full model of the human expert's competence, including a considerable variety of knowledge representation schemes and inference strategies, coordinated by a meta-process controller.

Artificial Intelligence↗

PubMedDate not supplied

Investigating hospital efficiency in the new NHS: the role of the translog cost function.

The reforms to the United Kingdom's National Health Service (NHS) of recent years have greatly increased the role of economic incentives in the hospital sector. Hospitals now have to compete for the business of GP and health authority purchasers and are assumed to have an incentive to minimise costs. This makes the analysis of cost functions much more relevant than has previously been the case. The objective of this paper is to assess the potential usefulness of the translog cost function applied in the NHS internal market. Three main issues are identified that limit the role of this type of cost function in the internal market: the adequacy of the econometric model (including data quality); the assumptions underlying the model, and; the interpretation of economies of scale, marginal costs and economies of scope that can be derived from such a cost function. It is concluded that at present the application of translog cost function analysis in the NHS is of limited usefulness, but that it does indicate areas for further methodological research.

Economic Competition↗

PubMedDate not supplied

Incorporating tracer-tracee differences into models to improve accuracy.

The ideal tracer for metabolic studies is one that behaves exactly like the tracee. Compounds labeled with isotopes come the closest to this ideal because they are chemically identical to the tracee except for the substitution of a stable or radioisotope at one or more positions. Even this substitution, however, can introduce a difference in metabolism that may be quantitatively important with regard to the development of the mathematical model used to interpret the kinetic data. The doubly labeled water method for the measurement of carbon dioxide production and hence energy expenditure in free-living subjects is a good example of how differences between the metabolism of the tracers and the tracee can influence the accuracy of the carbon dioxide production rate determined from the kinetic data.

Body Water↗

PubMed2005

Quantitative analysis of backbone dynamics in a crystalline protein from nitrogen-15 spin-lattice relaxation.

A detailed analysis of nitrogen-15 longitudinal relaxation times in microcrystalline proteins is presented. A theoretical model to quantitatively interpret relaxation times is developed in terms of motional amplitude and characteristic time scale. Different averaging schemes are examined in order to propose an analysis of relaxation curves that takes into account the specificity of MAS experiments. In particular, it is shown that magic angle spinning averages the relaxation rate experienced by a single spin over one rotor period, resulting in individual relaxation curves that are dependent on the orientation of their corresponding carousel with respect to the rotor axis. Powder averaging thus leads to a nonexponential behavior in the observed decay curves. We extract dynamic information from experimental decay curves, using a diffusion in a cone model. We apply this study to the analysis of spin-lattice relaxation rates of the microcrystalline protein Crh at two different fields and determine differential dynamic parameters for several residues in the protein.

Bacterial Proteins↗

PubMed1992

Analysis of fluctuations of "minimal" excitatory postsynaptic potentials during long-term potentiation in guinea pig hippocampal slices.

In previous studies, quantal analysis assuming a simple binomial model has shown that long-term potentiation (LTP) is accompanied by an increase in both mean quantal content (m) and quantal size (v), whereby the increase in m predominates. In the present study, "compound" binomial distributions with variable probabilities were convolved with Gaussian distributions in computer experiments to simulate amplitude histograms of intracellular excitatory postsynaptic potentials (EPSPs). A deconvolution procedure assuming equal "quantal" separation (v) between discrete components, but without assuming binomial statistics, was applied to the simulated distributions to determine v. It was found that with a small ratio of standard deviation of noise to v (Sn/v less than 0.4), a reliable estimate of v can be obtained even for small samples (N = 100). When Sn/v was larger (0.4-0.6), approximate v estimates (within +/- 10-20% of the simulated v) could be obtained by averaging estimates from about 10 small samples (N = 100). "Minimal" EPSPs were recorded in area CA1 of guinea pig hippocampal slices. 37 EPSP amplitude samples of 9 neurones were measured before and up to 55 min after 10 tetanizations of stratum radiatum. In accordance with the previous data, the increase in v accounted for only about 10% of the average post-tetanic increase in EPSP amplitude and was not correlated with the latter. However, for an EPSP subset with small LTP magnitude, the increase in v accounted for an essential part of the LTP magnitude while the increase in m did not correlate with it. The results are in agreement with previous data obtained in the context of the simple binomial model and are interpreted as indicating primarily a presynaptic mechanism of LTP maintenance. The results suggest two types of synaptic mechanism of LTP maintenance related to the increases in m and v, respectively. The latter mechanism is saturated at about 10 to 30% increase in post-tetanic amplitude above the pre-tetanic EPSP amplitude.

Animals↗

PubMed2002

Investigation of the transduction mechanism of infrared detection in Melanophila acuminata: photo-thermal-mechanical hypothesis.

Differential phase optical low coherence reflectometry (OLCR) was used to detect sub-wavelength displacements in the infrared-sensitive thoracic pit organ of Melanophila acuminata (Coleoptera: Buprestidae) upon absorption of infrared radiation at 3.39 microm. The displacement had more complex morphology but similar amplitude ( approximately 100 nm at 1 W cm(-2)) when compared to the displacement measured from the exocuticle in an alternate region on the beetle's body. In addition, a simplified finite difference model was developed to predict the temperature distribution and resultant thermal expansion in the pit organ tissue. The experimental and model results were interpreted to help clarify the mechanism by which the sensilla in the pit organ convert infrared radiation to neural signals. The results of this paper are discussed in relation to the photo-thermal-mechanical transduction hypothesis. This is the first experimental examination of the transduction mechanism in Melanophila acuminata.

Animals↗

PubMed1975

[Pulsecytophotometric analysis of cervical cells of postmenopausal women (author's transl)].

49 suspensions with cells of the cervix uteri exclusively taken from postmenopausal women were analysed with the pulsecytophotometer. 7 of these cases had an histologically verified invasive cancer of the cervix uteri. There were no false negative results, the percentage of false positive measurements was about 30%. This good result may arise from selecting the cell material analized exclusively from postmenopausal patients and the choice of mathematical model for histogram interpretation, which has been constructed analogously to the cycle of mitosis of cells. Comparing the different methods of preparation of the suspensions the standard procedure (RNase, pepsin, ultrasonic) before staining with ethidium bromide seems to be the best one.

Automation↗

PubMed1998

Towards a psycho-social theoretical framework for sexual health promotion.

Numerous interventions have been designed to promote safer sexual behaviour amongst young people. However, relatively few have proved effective, which is, at least partially, due to the lack of development of theoretically based programmes. An understanding of the origins and control of sexual behaviour can be derived from basic social science research. Unless this is applied to the design of behaviour-change programmes they are unlikely to target the most important determinants of young people's sexual behaviour and are, therefore, unlikely to be effective. This paper outlines some of the key theoretical insights which have been drawn upon in the development of a new sex education programme currently being tested in Scottish schools. The theoretical basis is intentionally eclectic, combining social psychological cognitive models with sociological interpretations, since we are not concerned to advance any particular theory but to find which are most useful in promoting sexual health. First, the social influences on sexual behaviour are considered, and then the way in which these translate into individual perceptions and beliefs. Finally, the paper attempts to develop a theoretical understanding of sexual interaction and the social contexts of sexual behaviour.

Adolescent↗