PubMed Health⌕ Search

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 1,711 records · Page 95Linked to original sources
PubMed2005

A model of prefrontal cortical mechanisms for goal-directed behavior.

Many behavioral tasks require goal-directed actions to obtain delayed reward. The prefrontal cortex appears to mediate many aspects of goal-directed decision making. This article presents a model of prefrontal cortex function emphasizing the influence of goal-related activity on the choice of the next motor output. The model can be interpreted in terms of key elements of Reinforcement Learning Theory. Different neocortical minicolumns represent distinct sensory input states and distinct motor output actions. The dynamics of each minicolumn include separate phases of encoding and retrieval. During encoding, strengthening of excitatory connections forms forward and reverse associations between each state, the following action, and a subsequent state, which may include reward. During retrieval, activity spreads from reward states throughout the network. The interaction of this spreading activity with a specific input state directs selection of the next appropriate action. Simulations demonstrate how these mechanisms can guide performance in a range of goal-directed tasks, and provide a functional framework for some of the neuronal responses previously observed in the medial prefrontal cortex during performance of spatial memory tasks in rats.

Animals↗
PubMed2000

Genes and the physics of the DNA double-helix.

The processing of the genetic information stored in the double-helical DNA implies the separation of the two strands, the physics of which is described by the helix-coil transition model. Is there a relationship between genetic maps and DNA physical stability maps that plot the sequence-specific propensity for the thermal disruption of the double-helix? Here, with appropriate methodological formulations, such maps are derived for a large set of sequences, including complete genomes. The superposition of the two maps leads to a contrasted picture with correlations ranging between two extremes: from almost perfect (with the genes precisely delineated as stable regions) to more or less complete unrelatedness. The simplest explanation for the results is that the observed striking correlations correspond to the relics of a primeval organisation of the genetic message, with the physics of DNA playing a role in the delimitation of coding regions. In order to trace the evolutionary fate of this signal further, a detailed study of the yeast complete genome is performed. In this study, the superposition of the genetic and physical stability maps is examined in the light of information concerning gene duplication. On the basis of this analysis it is concluded that the 'signature' associated with the supposed archaic signal is in the process of being erased, most probably because the underlying feature is no longer under selective pressure. There are many evolutionary implications for the results presented and for their proposed interpretations, notably concerning models of mutational dynamics in relation to erasure processes.

Algorithms↗
PubMed2000

Does the site of postgraduate family medicine training predict performance on summative examinations? A comparison of urban and remote programs.

BACKGROUND: The location of postgraduate medical training is shifting from teaching hospitals in urban centres to community practice in rural and remote settings. We were interested in knowing whether learning, as measured by summative examinations, was comparable between graduates who trained in urban centres and those who trained in remote and rural settings. METHODS: Family medicine training programs in Ontario were selected as a model of postgraduate medical training. The results of the 2 summative examinations--the Medical Council of Canada Qualifying Examination (MCCQE) Part II and the College of Family Physicians of Canada (CFPC) certification examination--for graduates of the programs at Ontario's 5 medical schools were compared with the results for graduates of the programs in Sudbury and Thunder Bay from 1994 to 1997. The comparability of these 2 cohorts at entry into training was evaluated using the results of their MCCQE Part I, completed just before the family medicine training. RESULTS: Between 1994 and 1997, 1013 graduates of family medicine programs (922 at the medical schools and 91 at the remote sites) completed the CFPC certification examination; a subset of 663 completed both the MCCQE Part I and the MCCQE Part II. The MCCQE Part I results for graduates in the remote programs did not differ significantly from those for graduates entering the programs in the medical schools (mean score 531.3 [standard deviation (SD) 69.8] and 521.8 [SD 74.4] respectively, p = 0.33). The MCCQE Part II results did not differ significantly between the 2 groups either (mean score 555.1 [SD 71.7] and 545.0 [SD 76.4] respectively, p = 0.32). Similarly, there were no consistent, significant differences in the results of the CFPC certification examination between the 2 groups. INTERPRETATION: In this model of postgraduate medical training, learning was comparable between trainees in urban family medicine programs and those in rural, community-based programs. The reasons why this outcome might be unexpected and the limitations on the generalizability of these results are discussed.

Analysis of Variance↗
PubMed2001

An overview of recent charged-particle radiation biology in Italy.

Radiobiology with charged particles is being carried out in Italy since several decades, starting with the experiments with protons in Milan. Later, also other groups entered the field, such as those in Naples, in Legnaro (LNL) and in Rome. In the last 10-15 years the activities in the field began to grow in a significant way. This happened in concomitance with the involvement of various researchers and Institutions in European and international projects devoted to radiation protection aspects, such as those aimed at elucidating and modelling radiation action mechanisms (EC/EU) and those aimed at radiation protection in space (NASA). A special role has been played since then by the Laboratori Nazionali di Legnaro of the INFN, where a radiobiology facility for low energy light ions was set up and operated in 1985. A formidable stimulus for charged-particle radiobiology was more recently given by the onset of plans for developing hadrontherapy Centres in Italy. The TERA Foundation first, and than the TOP Project at the Istituto Superiore di Sanità, at the same time favoured the spreading in Italy of radiobiology research with charged particles and encouraged co-operation among various groups. The Italian radiobiology community, though relatively small, developed a number of valuable activities with charged particles, mostly at the cellular and molecular levels, that span from mechanisms of radiation action to radiation protection in space and to therapy with charged hadrons. In this article, due to space limitations, we have just been able to list the present activities and to briefly review some research that forms a common background for the various areas. This includes the work on Chinese hamster V79 cells irradiated with light ions at LNL, that provided extensive data on the relationships between radiation quality, molecular damage and cellular effects, and the related work aimed at possible interpretation by mechanistic models. It appears that the multiplicity of objectives does not represent a factor of weakness for the relatively small Italian radiobiology community. Synergistic effects have been found because the basic radiobiology is the same for different areas such as radiation protection and hadrontherapy, and because the expertise and the methods developed for a given purpose may find useful applications in others. Though it is difficult to forecast the future development in Italy of the various areas that will take advantage of charged-particle radiobiology, it is expected that hadrontherapy and space radiation protection together, independently of their relative weight, will be important driving forces for the future development of the field in our Country.

Animals↗
PubMed2001

Pore distribution effect of activated carbon in adsorbing organic micropollutants from natural water.

Adsorption isotherms of organic micropollutants in coexistence with natural organic matter (NOM) were analyzed to evaluate the impacts of pore size distribution of activated carbon (AC) on the competition effects of the NOM. Single solute adsorption experiments and simultaneous adsorption experiments with NOM contained in a coagulation-pretreated surface water were performed for four agricultural chemicals and three coal-based activated carbons (ACs) having different pore distributions. The results showed that, for all the carbons used, the adsorption capacity of the chemicals was reduced distinctly in the presence of NOM. Such a reduction was more apparent for AC with a larger portion of small pores suitable for the adsorption of small organic molecules and for the agricultural chemicals with a more hydrophilic nature. Ideal adsorbed solution theory (IAST) incorporated with the Freundlich isotherm expression (IAST-Freundlich model) could not interpret the impact of NOM on the adsorption capacity of the chemicals unless a pore blockage effect caused by the adsorption of NOM was also considered. By taking into account this effect, the adsorption isotherm of the chemicals in the presence of NOM was well described, and the capacity reduction caused by the NOM was quantitatively assessed from the viewpoints of the site competition and the pore blockage. Analytical results clearly indicated that pore blockage was an important competition mechanism that contributed to 10-99% of the total capacity reductions of the chemicals, the level depended greatly on the ACs, the chemicals and the equilibrium concentrations, and could possibly be alleviated by broadening the pore size distributions of the ACs to provide a large volume percentage for pores with sizes above 30 A.

Adsorption↗
PubMed1979

The contractile basis of ameboid movement. VI. The solation-contraction coupling hypothesis.

The contracted pellets derived from a high-speed supernate of Dictyostelium discoideum (S3) were investigated to determine the functional activity associated with this specific subset of the cellular motile apparatus. A partially purified model system of gelation and contraction (S6) was prepared from the contracted pellets, and the presence of calcium- and pH-sensitive gelation and contraction in this model demonstrated that a functional cytoskeletal-contratile complex remained at least partially associated with the actin and myosin during contraction. Semi-quantitative assays of gelation and solation in the myosin-free preparation S6 included measurements of turbidity, relative viscosity, and strain birefringence. The extent of gelation was optimal at pH 6.8 and a free calcium ion concentration of approximately 3.0 x 10(-8) M. Solation was favored when the free calcium ion concentration was greater than 7.6 x 10(-7) M or when the pH was increased or decreased from pH 6.8. Gelation was reversibly inhibited by increasing the free calcium ion concentration to approxomately 4.6 x 10(-6) M at pH 6.8. The solation-gelation process of this model has been interpreted to involve the reversible cross-linking of actin filaments. The addition of purified D. discoideum myosin to S6 served to reconstitute calcium- and pH-regulated contraction. The results from this study indicate that contraction is coupled functionally to the local breakdown (solation) of the gel. Therefore, solation has been identified as a structural requirement for extensive shortening during contraction. We have called this concept the solation-contraction coupling hypothesis. Fractionation of a preparation derived from the contracted pellets yielded a fraction consisting of actin and a 95,000-dalton polypeptide that exhibited calcium-sensitive gelation at 28 degrees C and a fraction composed of actin and 30,000- and 18,000-dalton polypeptides that demonstrated calcium-sensitive genlation at 0 degrees C.

Calcium↗
PubMed1996

Characterization of the radioactive metabolites of the 5-HT1A receptor radioligand, [O-methyl-11C]WAY-100635, in monkey and human plasma by HPLC: comparison of the behaviour of an identified radioactive metabolite with parent radioligand in monkey using PET.

N-(2-(4-(2-Methoxy-phenyl)-1-piperazin-1-yl)ethyl)-N-(2-pyridyl) cyclohexanecarboxamide (WAY-100635), labelled in the O-methyl group with carbon-11 (t1/2 = 20.4 min), is a promising radioligand for application with positron emission tomography (PET) to the study of 5-HT1A receptors in living human brain. An understanding of the metabolism of this new radioligand is crucial to the development of a biomathematical model for the interpretation of the kinetics of radioactivity uptake in brain in terms of receptor-binding parameters. After intravenous injection of [O-methyl-11C]WAY-100635 into humans, radioactivity was found to clear rapidly from blood and plasma. By using established methods for the analysis of radioactivity in plasma, it was found that intravenously injected [O-methyl-11C]WAY-100635 is rapidly metabolised to more polar radioactive compounds in a cynomolgus monkey and in humans. Thus, at 60 min postinjection, parent radioligand represented 40% and 5% of the radioactivity in monkey and human plasma, respectively. In monkey and human, one of the radioactive metabolites was identified as the descyclohexanecarbonyl analogue of the parent radioligand, namely [O-methyl-11C]WAY-100634. This compound is known to have high affinity for 5-HT1A receptors and alpha 1-adrenoceptors. In a PET experiment it was demonstrated that, after IV injection of [O-methyl-11C]WAY-100634 into a cynomolgus monkey, radioactivity was avidly taken up by brain. Uptake of radioactivity was higher in 5-HT1A receptor-rich frontal cortex than in cerebellum, which is devoid of 5-HT1A receptors. Polar radioactive metabolites appeared in plasma. The results suggest that the use of WAY-100635 labelled with carbon-11 in its cyclohexanecarbonyl moiety may provide enhanced signal contrast in PET studies and a possibility to develop a simple biomathematical model for regional brain radioactivity uptake.

Animals↗
PubMed2006

Theoretical model of infrared spectra of hydrogen bonds in molecular crystals and its application to interpretation of infrared spectra of 1-methylthymine.

A theoretical model for vibrational interactions in the hydrogen bonds in molecular crystals with four molecules forming two centrosymmetric dimers in the unit cell is presented. The model takes into account anharmonic-type couplings between the high-frequency N-H(D) and the low-frequency N...O stretching vibrations in each hydrogen bond, resonance interactions (Davydov coupling) between equivalent hydrogen bonds in each dimer, resonance interdimer interactions within a unit cell, and Fermi resonance between the N-H(D) stretching fundamental and the first overtone of the N-H(D) in-plane bending vibrations. The vibrational Hamiltonian, selection rules, and expressions for the integral properties of an absorption spectrum are derived. The model is used for theoretical simulation of the NH stretching bands of 1-methylthymine and its ND derivative at 300 K. The effect of deuteration is successfully reproduced by our model. Infrared, far-infrared, Raman, and low-frequency Raman spectra of 1-methylthymine and its deuterated derivative have been measured. Experimental geometry and frequencies are compared with the results of density functional theory calculations performed at the B3LYP6-311++G**, B3LYP/cc-pVTZ, B3PW916-311++G**, and B3PW91/cc-pVTZ levels.

Journal Article↗
PubMed2003

Analytic expressions for rate and CV of a type I neuron driven by white gaussian noise.

We study the one-dimensional normal form of a saddle-node system under the influence of additive gaussian white noise and a static "bias current" input parameter, a model that can be looked upon as the simplest version of a type I neuron with stochastic input. This is in contrast with the numerous studies devoted to the noise-driven leaky integrate-and-fire neuron. We focus on the firing rate and coefficient of variation (CV) of the interspike interval density, for which scaling relations with respect to the input parameter and noise intensity are derived. Quadrature formulas for rate and CV are numerically evaluated and compared to numerical simulations of the system and to various approximation formulas obtained in different limiting cases of the model. We also show that caution must be used to extend these results to the Theta neuron model with multiplicative gaussian white noise. The correspondence between the first passage time statistics for the saddle-node model and the Theta neuron model is obtained only in the Stratonovich interpretation of the stochastic Theta neuron model, while previous results have focused only on the Ito interpretation. The correct Stratonovich interpretation yields CVs that are still relatively high, although smaller than in the Ito interpretation; it also produces certain qualitative differences, especially at larger noise intensities. Our analysis provides useful relations for assessing the distance to threshold and the level of synaptic noise in real type I neurons from their firing statistics. We also briefly discuss the effect of finite boundaries (finite values of threshold and reset) on the firing statistics.

Action Potentials↗
PubMed1996

A review of parametric modelling techniques for EEG analysis.

This review provides an introduction to the use of parametric modelling techniques for time series analysis, and in particular the application of autoregressive modelling to the analysis of physiological signals such as the human electroencephalogram. The concept of signal stationarity is considered and, in the light of this, both adaptive models, and non-adaptive models employing fixed or adaptive segmentation, are discussed. For non-adaptive autoregressive models, the Yule-Walker equations are derived and the popular Levinson-Durbin and Burg algorithms are introduced. The interpretation of an autoregressive model as a recursive digital filter and its use in spectral estimation are considered, and the important issues of model stability and model complexity are discussed.

Biomedical Engineering↗
PubMed1994

High-resolution, non-crystallographic structural studies of large integral membrane proteins.

The work described here clearly demonstrates that n.m.r. is a viable method with which to resolve molecular details about membrane proteins and can give information at a resolution comparable with that gained from crystallographic studies, where, in the limited number of cases studied so far, such information is available. Sensitivity is not a major problem although other difficulties may prevent particular kinds of information being resolved. However, because bond orientational details are obtained ab initio, the method is quite model independent and interpretationally unique. The requirements for solid-state n.m.r. methods to be applied to a large membrane protein are that: a sufficient amount of functionally active protein is available; the protein should be located in bilayers; and suitable specific isotopic enrichment is achieved, in ideal cases, either by chemical or biosynthetic means. Achieving these requirements will need to draw on a range of skills, possibly genetic or chemical or both, and a full understanding of the biochemistry is essential to ensure that any structural information gained is functionally relevant. The structural information potentially available includes: (a) conformation details about specific parts of a protein or its binding sites; (b) distance measurements between specific labels introduced into the protein; (c) dynamics of particular parts of the protein; (d) chain folding and residue orientation; and (e) the mechanistic changes that can occur during a functional cycle. Much developmental work still needs to be done both on the sample handling instrumental side and with the theoretical aspect of the method.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗
PubMed2001

Marker concentration patterns of labelled leaf and stem particles in the rumen of cattle grazing bermuda grass (Cynodon dactylon) analysed by reference to a raft model.

Large (>1600 microm), ingestively masticated particles of bermuda grass (Cynodon dactylon L. Pers.) leaf and stem labelled with 169Yb and 144Ce respectively were inserted into the rumen digesta raft of heifers grazing bermuda grass. The concentration of markers in digesta sampled from the raft and ventral rumen were monitored at regular intervals over approximately 144 h. The data from the two sampling sites were simultaneously fitted to two pool (raft and ventral rumen-reticulum) models with either reversible or sequential flow between the two pools. The sequential flow model fitted the data equally as well as the reversible flow model but the reversible flow model was used because of its greater application. The reversible flow model, hereafter called the raft model, had the following features: a relatively slow age-dependent transfer rate from the raft (means for a gamma 2 distributed rate parameter for leaf 0.0740 v. stem 0.0478 h(-1)), a very slow first order reversible flow from the ventral rumen to the raft (mean for leaf and stem 0.010 h(-1)) and a very rapid first order exit from the ventral rumen (mean of leaf and stem 0.44 h(-1)). The raft was calculated to occupy approximately 0.82 total rumen DM of the raft and ventral rumen pools. Fitting a sequential two pool model or a single exponential model individually to values from each of the two sampling sites yielded similar parameter values for both sites and faster rate parameters for leaf as compared with stem, in agreement with the raft model. These results were interpreted as indicating that the raft forms a large relatively inert pool within the rumen. Particles generated within the raft have difficulty escaping but once into the ventral rumen pool they escape quickly with a low probability of return to the raft. It was concluded that the raft model gave a good interpretation of the data and emphasized escape from and movement within the raft as important components of the residence time of leaf and stem particles within the rumen digesta of cattle.

Analysis of Variance↗
PubMed1990

Agonist potency and apparent affinity: interpretation using classical and steady-state ternary-complex models.

The relative potencies of agonists have been used in the past to classify receptors. This could be justified on the basis of ideas and equations developed using the occupancy model of drug action. Although attempts have been made recently to develop methods of analysis based on other models, these have not been entirely satisfactory. In this article Dennis Mackay explores possible interpretations of relative potency, apparent affinity and efficacy on the basis of a steady-state ternary-complex model. He concludes that use of these quantities to classify receptor-transducer systems may not be entirely dependable. However the steady-state model can explain why use of agonists to classify receptors has been successful in the past and why properties of agonists deduced from ligand binding studies may differ from their properties deduced from experiments on intact tissues.

Animals↗
PubMed2003

Pore formation by 6-ketocholestanol in phospholipid monolayers and its interpretation by a general nucleation-and-growth model accounting for the sigmoidal shape of voltage-clamp curves of ion channels.

6-Ketocholestanol (KC), a steroid that differs from cholesterol mainly by the presence of a carbonyl group, forms pores inside a dioleoylphosphatidylcholine monolayer self-assembled on mercury by a mechanism similar to that of channel-forming peptides and proteins. The potential steps responsible for pore formation by KC molecules give rise to potentiostatic charge vs time curves whose sigmoidal shape and potential dependence can be quantitatively interpreted on the basis of a mechanism of nucleation and growth of KC clusters. Pore formation by KC allows the penetration of thallous ions across the otherwise impermeable phosphatidylcholine monolayer, while pore disruption taking place at more negative potentials causes a drop in thallous ion permeation. Pore disruption is also accounted for by a mechanism of nucleation and growth of holes inside the KC clusters. The kinetic model of nucleation and growth is general, and accounts quantitatively for the sigmoidal shape and potential dependence of the classical Hodgkin-Huxley voltage-clamp curves of potassium channels in squid giant axon,(1) using a minimum number of free parameters.

Animals↗
PubMed1998

Moiré patterns in high resolution electron microscopy images of MoS2.

The contrast observed in the (0001) HREM images of MoS2 are interpreted with a model based on rotational moiré patterns. Computer-simulated images show that rotations of two supercells by angles of 3, 5, 8, and 16 degrees around their centers reproduce the contrast observed in the experimental images. The calculations suggest that these rotations occur between the sulphur layers of the MoS2 structure.

Computer Simulation↗
PubMed1997

Radiation and speciation of spider monkeys, genus Ateles, from the cytogenetic viewpoint.

The chromosomes of 22 animals of four subspecies of the genes Ateles (A. paniscus paniscus, A. p. chamek, A. belzebuth hybridus, and A. b. marginatus) were compared using G/C banding and NOR (nucleolar organizer region) staining methods. The cytogenetic data of Ateles in the literature were also used to clarify the phylogenetic relationships of the species and subspecies and to infer the routes of radiation and speciation of these taxa. Chromosomes 6 and 7 that showed more informative geographic variation and the apomorphic form 4/12, exclusively in A. p. paniscus, are the keys for understanding the evolution, radiation, and specification of the Ateles taxa. The ancestral populations of the genus originated in the southwestern Amazon Basin (the occurrence area of A. paniscus chamek) and spread in the Amazon Basin and westward, crossing the Andes and colonizing Central America and northwesternmost regions of South America. The evolutionary history of the northern South American taxa is interpreted using the model of biogeographical evolution postulated by Haffer [Science 185:131-137, 1969]. Ateles paniscus paniscus is the genetically most differentiated form and probably derives from A. belzebuth hybridus. Based on the karyotype differences, the populations of Ateles can be divided into four different groups. These findings indicate the necessity of a more coherent taxonomic arrangement for the taxa of Ateles.

Animals↗
PubMed2001

Comparison of artificial neural networks with other statistical approaches: results from medical data sets.

BACKGROUND: In recent years, considerable attention has been given to the development of sophisticated techniques for exploring data sets. One such class of techniques is artificial neural networks (ANNs). Artificial neural networks have many attractive theoretic properties, specifically, the ability to detect non predefined relations such as nonlinear effects and/or interactions. These theoretic advantages come at the cost of reduced interpretability of the model output. Many authors have analyzed the same data set, based on these factors, with both standard statistical methods (such as logistic or Cox regression) and ANN. METHODS: The goal of this work is to review the literature comparing the performance of ANN with standard statistical techniques when applied to medium to large data sets (sample size > 200 patients). A thorough literature search was performed, with specific criteria for a published comparison to be included in this review. RESULTS: In the 28 studies included in this review, ANN outperformed regression in 10 cases (36%), was outperformed by regression in 4 cases (14%), and the 2 methods had similar performance in the remaining 14 cases (50%). However, in the 8 largest studies (sample size > 5000), regression and ANN tied in 7 cases, with regression winning in the remaining case. In addition, there is some suggestion of publication bias. CONCLUSIONS: Neither method achieves the desired performance. Both methods should continue to be used and explored in a complementary manner. However, based on the available data, ANN should not replace standard statistical approaches as the method of choice for the classification of medical data.

Humans↗