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 667 records · Page 37Linked to original sources

PubMed1994

Young children's ability to understand a model as a spatial representation.

Children's ability to understand that a real environment can be represented in a symbolic form (i.e., by a model) is an important developmental achievement. Researchers have claimed that children who are just 3 years of age appreciate models as representations. This research was based on tasks that involved having young children use a model to locate a hiding place in an actual room. In this article, however, we point out the difficulties in interpreting previous model tasks, and we describe two studies that showed that young 3-year-olds could perform model tasks successfully when the hiding place they were looking for was a unique place in the model (and room). When the hiding place was unique, the children had to note only that place and they needed no further knowledge about the relationship between the model and the room. When the hiding place was one of two identical places, however, the children needed to take spatial relationships into account to distinguish the correct place, and young 3-year-olds were unable to do this. Four-year-olds were able to use spatial relationships to distinguish identical places when the model was aligned with the space it represented, but they had difficulty when the model was not aligned. Five-year-olds could use spatial relationships effectively between one model space and another whether or not the model was aligned.

Child↗

PubMed2003

The structure of full-length LysR-type transcriptional regulators. Modeling of the full-length OxyR transcription factor dimer.

The LysR-type transcriptional regulators (LTTRs) comprise the largest family of prokaryotic transcription factors. These proteins are composed of an N-terminal DNA binding domain (DBD) and a C-terminal cofactor binding domain. To date, no structure of the DBD has been solved. According to the SUPERFAMILY and MODBASE databases, a reliable homology model of LTTR DBDs may be built using the structure of the Escherichia coli ModE transcription factor, containing a winged helix- turn-helix (HTH) motif, as a template. The remote, but statistically significant, sequence similarity between ModE and LTTR DBDs and an alignment generated using SUPERFAMILY and MODBASE methods was independently confirmed by alignment of sequence profiles representing ModE and LTTR family DBDs. Using the crystal structure of the E.coli OxyR C-terminal domain and the DBD alignments we constructed a structural model of the full-length dimer of this LTTR family member and used it to investigate the mode of protein-DNA interaction. We also applied the model to interpret, in a structural context, the results of numerous biochemical studies of mutated LTTRs. A comparison of the LTTR DBD model with the structures of other HTH proteins also provides insights into the interaction of LTTRs with the C-terminal domain of the RNA polymerase alpha subunit.

Amino Acid Sequence↗

PubMed2006

Diagnostic architectural and dynamic features at breast MR imaging: multicenter study.

PURPOSE: To prospectively determine the prevalence and predictive value of three-dimensional (3D) and dynamic breast magnetic resonance (MR) imaging and contrast material kinetic features alone and as part of predictive diagnostic models. MATERIALS AND METHODS: The study protocol was approved by the institutional review board or ethics committees of all participating institutions, and informed consent was obtained from all participants. Although study data collection was performed before HIPAA went into effect, standards that would be compliant with HIPAA were adhered to. Data from the International Breast MR Consortium trial 6883 were used in the analysis. Women underwent 3D (minimum spatial resolution, 0.7 x 1.4 x 3 mm; minimal temporal resolution, 4 minutes) and dynamic two-dimensional (temporal resolution, 15 seconds) MR imaging examinations. Readers rated enhancement shape, enhancement distribution, border architecture, enhancement intensity, presence of rim enhancement or internal septations, and the shape of the contrast material kinetic curve. Regression was performed for each feature individually and after adjustment for associated mammographic findings. Multivariate models were also constructed from multiple architectural and dynamic features. Areas under the receiver operating characteristic curve (Az values) were estimated for all models. RESULTS: There were 995 lesions in 854 women (mean age, 53 years +/- 12 [standard deviation]; range, 18-80 years) for whom pathology data were available. The absence of enhancement was associated with an 88% negative predictive value for cancer. Qualitative characterization of the dynamic enhancement pattern was associated with an Az value of 0.66 across all lesion architectures. Focal mass margins (Az = 0.76) and signal intensity (Az = 0.70) were highly predictive imaging features. Multivariate models were constructed with an Az value of 0.880. CONCLUSION: Architectural and dynamic features are important in breast MR imaging interpretation. Multivariate models involving feature assessment have a diagnostic accuracy superior to that of qualitative characterization of the dynamic enhancement pattern.

Adolescent↗

PubMed1999

Cerebral blood flow regulation in REM sleep: a model for flow-metabolism coupling.

The pattern of metabolic and circulatory changes occurring during REM sleep in the whole brain is also observed at a regional level in different instances of functional activation. This pattern is characterized by an increase in metabolic rate, blood flow, glucose and oxygen uptake, the increase in glucose uptake generally exceeding oxygen uptake. A model of interpretation is presented, based on the assumption that substantial limitation to oxygen diffusion exists in the brain. According to the model, microregions lying at mid-distance between capillaries may become hypoxic, depending on metabolic rate and blood-cell PO2 difference. At increasing metabolic rates, O2 consumption in pericapillary microregions increases and the PO2 drop becomes steeper. As a consequence, in microregions far from capillaries a decrease in O2 availability occurs, in concomitance with the increase in metabolic rate, so that non-oxidative glucose metabolism develops locally. A similar spatial PO2 pattern forms in the case of arterial hypoxia, when capillary PO2, and then blood-cell PO2 difference, is reduced. The hypoxic microregions are the source of vasodilatatory messages, the consequent vasodilatation increasing average capillary PO2 and then favoring O2 diffusion to the tissue. Oxygen thus appears to be a better candidate than glucose as a mediator of blood flow-metabolism coupling. This is supported by its higher extraction fraction and by the fact that, in physiologic conditions, arterial hypoxia (and not hypoglycemia) acts on cerebral blood flow. Moreover, the diffusion capacity of glucose in the brain is higher than that of oxygen, so that diffusion limitation is more likely to occur for oxygen. The present model allows consistent organization of the stereotyped changes in cerebral blood flow and glucose and oxygen uptake occurring both in REM sleep and in other instances of brain activation.

Animals↗

PubMed1994

Application of 3-dimensional homology modeling of cytochrome P450 2B1 for interpretation of site-directed mutagenesis results.

Three-dimensional structures of cytochrome P450 2B1 were modeled based on the crystallographic structure of P450cam. The effect of the alignment, loop choice, and minimization with or without water was assessed. Although final models were similar in overall structure, the identity of active site residues depended upon the alignment. An example is Phe-206, which may or may not form part of the active site. The choice of the loop conformation had a lesser effect, while including water in the final minimization step was essential for preserving the shape and size of the active site. The best model (model 2) was in good agreement with the data from site-directed mutagenesis studies, and correctly predicted the effect of substitutions at 9 out of 10 amino acid positions. Thus, residues important for P450 2B1 activity, such as Ile-114, Phe-206, Ile-290, Thr-302, Val-363, and Gly-478, constitute part of the active site and are able to interact with the substrate androstenedione through hydrophobic interactions. On the other hand, Ser-303, Ser-360 and Lys-473 are far from the active site and/or cannot interact with the substrate, in agreement with experimental data. The model indicates other residues likely to be important for enzyme function, such as Tyr-111, Leu-209, Ile-477, and Ile-480, which can be tested experimentally. The substrate may assume numerous binding orientations consistent with observed patterns of hydroxylation at C15 and C16. The replacement in the model of certain amino acid residues to mimic residue substitutions from site-directed mutagenesis studies and docking of the substrate into the modified active site allowed a plausible explanation for alterations in regio- and stereospecificities of some mutants of P450 2B1, such as Gly-478-->Ala or Val-363-->Ala.

Amino Acid Sequence↗

PubMed2001

Probing structure and dynamics of DNA with 2-aminopurine: effects of local environment on fluorescence.

2-Aminopurine (2AP) is an analogue of adenine that has been utilized widely as a fluorescence probe of protein-induced local conformational changes in DNA. Within a DNA strand, this fluorophore demonstrates characteristic decreases in quantum yield and emission decay lifetime that vary sensitively with base sequence, temperature, and helix conformation but that are accompanied by only small changes in emission wavelength. However, the molecular interactions that give rise to these spectroscopic changes have not been established. To develop a molecular model for interpreting the fluorescence measurements, we have investigated the effects of environmental polarity, hydrogen bonding, and the purine and pyrimidine bases of DNA on the emission energy, quantum yield, and intensity decay kinetics of 2AP in simple model systems. The effects of environmental polarity were examined in a series of solvents of varying dielectric constant, and hydrogen bonding was investigated in binary mixtures of water with 1,4-dioxane or N,N-dimethylformamide (DMF). The effects of the purine and pyrimidine bases were studied by titrating 2AP deoxyriboside (d2AP) with the nucleosides adenosine (rA), cytidine (rC), guanosine (rG), and deoxythymidine (dT), and the nucleoside triphosphates ATP and GTP in neutral aqueous solution. The nucleosides and NTPs each quench the fluorescence of d2AP by a combination of static (affecting only the quantum yield) and dynamic (affecting both the quantum yield and the lifetime, proportionately) mechanisms. The peak wavelength and shape of the emission spectrum are not altered by either of these effects. The static quenching is saturable and has half-maximal effect at approximately 20 mM nucleoside or NTP, consistent with an aromatic stacking interaction. The rate constant for dynamic quenching is near the diffusion limit for collisional interaction (k(q) approximately 2 x 10(9) M(-1) s(-1)). Neither of these effects varies significantly between the various nucleosides and NTPs studied. In contrast, hydrogen bonding with water was observed to have a negligible effect on the emission wavelength, fluorescence quantum yield, or lifetime of 2AP in either dioxane or DMF. In nonpolar solvents, the fluorescence lifetime and quantum yield decrease dramatically, accompanied by significant shifts in the emission spectrum to shorter wavelengths. However, these effects of polarity do not coincide with the observed emission wavelength-independent quenching of 2AP fluorescence in DNA. Therefore, we conclude that the fluorescence quenching of 2AP in DNA arises from base stacking and collisions with neighboring bases only but is insensitive to base-pairing or other hydrogen bonding interactions. These results implicate both structural and dynamic properties of DNA in quenching of 2AP and constitute a simple model within which the fluorescence changes induced by protein-DNA binding or other perturbations may be interpreted.

2-Aminopurine↗

PubMed1991

Combining physiologic models and symbolic methods to interpret time-varying patient data.

This paper describes a methodology for representing and using medical knowledge about temporal relationships to infer the presence of clinical events that evolve over time. The methodology consists of three steps: (1) the incorporation of patient observations into a generic physiologic model, (2) the conversion of model states and predictions into domain-specific temporal abstractions, and (3) the transformation of temporal abstractions into clinically meaningful descriptive text. The first step converts raw observations to underlying model concepts, the second step identifies temporal features of the fitted model that have clinical interest, and the third step replaces features represented by model parameters and predictions into concepts expressed in clinical language. We describe a program, called TOPAZ, that uses this three-step methodology. TOPAZ generates a narrative summary of the temporal events found in the electronic medical record of patients receiving cancer chemotherapy. A unique feature of TOPAZ is its use of numeric and symbolic techniques to perform different temporal reasoning tasks. Time is represented both as a continuous process and as a set of temporal intervals. These two temporal models differ in the temporal ontology they assume and in the temporal concepts they encode. Without multiple temporal models, this diversity of temporal knowledge could not be represented.

Adult↗

PubMed2001

Flow of Multicomponent Electrolyte Solutions through Narrow Pores of Nanofiltration Membranes.

The slow flow of a multicomponent electrolyte solution in a narrow pore of a nanofiltration membrane is considered. The well-known semiempirical method of subdivision of electrical potential into quasi-equilibrium and streaming parts and the definition of streaming concentrations and pressure are discussed. The usefulness of this tool for solving the electrohydrodynamic equations is shown and justified: the use of a small parameter enables a system of electrohydrodynamic partial differential equations to be reduced to a system of ordinary differential equations for streaming functions. Boundary conditions for streaming functions at both the capillary inlet and outlet are derived. The proposed model is developed for the flow of a multicomponent electrolyte solution with an arbitrary number of ions. This is coupled with (i) the introduction of specific interactions between all ions and the pore wall and (ii) the inclusion of the dissociation of water in both conservation and transport equations. Effective distribution coefficients of ions are introduced that are functions of both the specific interaction potentials and the surface potential of the nanofiltration membrane material. The axial dependency of surface potential is expressed by the use of a charge regulation model from which the discontinuity in electric potential and ion pore concentrations at the pore inlet and outlet can be described. A relation between the frequently used capillary and homogeneous models of nanofiltration membranes is developed. An example of application of the homogeneous model for interpretation of experimental data on nanofiltration separation of electrolyte solutions is presented, which shows a reasonable predictive ability for the homogeneous model. Copyright 2001 Academic Press.

Journal Article↗

PubMed1975

Intrauterine pressure wave form characteristics in hypocontractile labor before and after oxytocin administration.

The data demonstrate that the contractions of hypocontractile active labor and normal spontaneous labor are different in several measures in addition to maximal amplitude. Furthermore, when the pathophysiology is corrected by the use of oxytocin, the contractions resemble those of normal spontaneous labor except in the maximal rate of tension development. Our data tend to support the subcellular model of uterine contractility, although the incompleteness of these models limits interpretation.

Adrenocorticotropic Hormone↗

PubMed1998

Structure of the Coolidge Axis II Inventory Personality Disorder Scales from the five-factor model perspective.

Coolidge, et al. in 1994 tested the generality and comprehensiveness of the five-factor model of personality as applied to personality disorder by performing a canonical correlation analysis for the scales from the Coolidge Axis II Inventory and the NEO Personality Inventory testing 178 undergraduates (106 men and 72 women). Their results did not support the generality and comprehensiveness of the five-factor model for interpreting the structure of personality disorders. A major problem with this study was that the data did not show good simple structure and meaningfulness because no rotation was performed for the canonical variates. The present study tested the hypothesis that the results of Coolidge, et al. might be attributed to the failure to rotate canonical variates to obtain good simple structure. For 220 students in introductory psychology (104 men and 116 women), canonical correlation analysis with varimax rotation was performed for scores on the Coolidge Axis II Inventory scales and the NEO Five-Factor Inventory scales. The analysis indicated five canonical variate pairs which were interpreted as Neuroticism, Extraversion, Openness, Disagreeableness, and Conscientiousness, supporting the tested hypothesis as well as the generality and comprehensiveness of this model for describing the structure of personality disorders.

Adolescent↗

PubMed2026

Prediction of antimicrobial minimum inhibitory concentration from bacterial genomes using a scalable and interpretable machine learning approach.

Although machine learning models can predict antimicrobial susceptibility from bacterial whole genome sequencing (WGS), state-of-the-art approaches are computationally demanding or dependent on knowledge of genetic resistance determinants. Here, we describe an efficient data-driven approach to predicting minimum inhibitory concentration (MIC) by progressively extending and refining predictive genome segments, independent of prior knowledge of resistance determinants. Resultant models had high interpretability - known and potentially novel resistance determinants were captured. Using 762 clinical E. coli strains, 71.6% of predictions were within one dilution of the measured MIC. Models trained with this algorithm generalised better onto external data (F1 score = 0.85) compared with alternative models trained on annotated resistance determinants (F1 = 0.82) or k-mer counts (F1 = 0.74). Computational demands were low (RAM usage 23.6GB vs 38.8GB for k-mer model). These advantages represent an important advance in predicting antimicrobial susceptibility from WGS, with potential applications for clinical diagnostics, drug development, and surveillance.

Journal Article↗

PubMed2002

Harmonic analysis of low-frequency bioelectrode behavior.

This paper concerns the modeling and interpretation of harmonics observed as a result of the nonlinear electrical properties of biomedical electrode/electrolyte interfaces. The higher order harmonics have been calculated assuming that the nonlinearity of the interfacial impedance is dominated at low frequencies by the nonlinear faradaic charge transfer resistance. The harmonic distribution in the output signal is compared between 1) the author's theoretical model based on the Butler-Volmer equation and 2) Schwan's empirical model and results. The influence of different parameters such as the number of electrons involved in the faradaic reaction and the transfer coefficient was investigated in order to physically interpret the experimental results. A good agreement was found between the authors' model and some of the experimental data previously reported in the literature. Further, potentially productive areas of research have been identified.

Computer Simulation↗

PubMed2001

Physiological significance of the rate constants in compartmental analysis.

Transfer (rate) constants in compartmental analysis are generally considered solely in operational terms, and little attention has been paid to their physiological interpretation. In this study we have examined the significance of their roles in physiological terms and the implications of this interpretation. When freely diffusible tracers are introduced into the blood stream by bolus injection, the sum of the exit rate constants from the blood compartment gives the number of blood volumes turned over per unit time; when multiplied by the blood volume, this value is the cardiac output. When corrected for body weight, the product yields the cardiac index, a biological constant for large mammals. The ratio of the rate constant supplying an organ to the exit-constant sum gives fractional cardiac output, and when multiplied by the cardiac output the ratio gives organ blood flow, independent of diffusibility. For freely diffusible tracers, organ blood flow can be calculated directly from the product of the organ rate constant and blood volume. At equilibrium and at equal concentration, the ratio of rate constants between adjacent compartments gives their relative volumes; this interpretation is fundamental when partitioning compartments for detailed analysis. These considerations have been validated by testing in both animals and man. Multiple solution sets may occur in mammillary models. Knowledge of the physiological interpretation of intercompartmental rate constants is of importance not only in model interpretation but in model validation, where this information serves as an independent check on model structure and its realization.

Animals↗

PubMed2004

Rites of passage of the engram: reconsolidation and the lingering consolidation hypothesis.

Memory consolidation refers to the progressive stabilization of items in long-term memory as well as to the memory phase(s) during which this stabilization takes place. The textbook account is that, for each item in memory, consolidation starts and ends just once. In recent years, however, the notion that memories reconsolidate upon their reactivation and hence regain sensitivity to amnestic agents has been revitalized. This issue is of marked theoretical and clinical interest. Here we review the recent literature on reconsolidation and infer, on the basis of the majority of the data, that blockade of reconsolidation does not induce permanent amnesia. Further, in several systems, reconsolidation occurs only in relatively fresh memories. We propose a framework model, which interprets reconsolidation as a manifestation of lingering consolidation, rather than recapitulation of a process that had already come to a closure. This model reflects on the nature of consolidation in general and makes predictions that could guide further research.

Animals↗

PubMedDate not supplied

Neurochemical changes following postmortem ischemia in the rat retina.

Glutamate and gamma-aminobutyric acid (GABA) are the dominant amino acids in the retina and brain. The manufacturing and degradation pathways of both of these amino acids are intricately linked with the tricarboxylic acid cycle leading to rapid redistribution of these amino acids after metabolic insult. Postmortem ischemia in mammalian retina predominantly results in a loss of glutamate and GABA from neurons and accumulation of these amino acids within Müller cells. This accumulation of glutamate and GABA in Müller cells may occur as a result of increased release of these neurotransmitters from neurons, and decreased degradation. Quantification of the semisaturation value (half-maximal response) for glutamate and GABA Müller cell loading during postmortem ischemia indicated a shorter semisaturation value for GABA than glutamate. Such changes are consistent with a single aerobically dependent GABA-degradation pathway, and the existence of multiple glutamate-degradation pathways. Comparison with the in vitro ischemic model showed similar qualitative characteristics, but a markedly increased semisaturation time for glutamate and GABA Müller cell loading (a factor of 5-10) in the postmortem ischemia model. We interpret these differences to indicate that the in vitro condition provides a more immediate and/or severe ischemic insult. In the postmortem ischemia model, the delayed glial cell loading implies the availability of internal stores of both glucose and/or oxygen. Increased glial and neuronal immunoreactivity for the amino acids involved in transamination reactions, aspartate, alanine, leucine, and ornithine was observed, indicating a potential shift in the equilibrium of transamination reactions associated with glutamate production. These findings provide evidence that, in the rat retina, there are multiple pathways subserving glutamate production/degradation that include a multitude of transamination reactions. Further evidence is therefore provided to support a role for all four amino acids in glutamate metabolism within a variety of retinal neurons and glia.

Alanine↗

PubMed2004

Identifying and accommodating statistical outliers when setting prospective payment rates for inpatient rehabilitation facilities.

OBJECTIVE: To demonstrate how a Bayesian outlier accommodation model identifies and accommodates statistical outlier hospitals when developing facility payment adjustments for Medicare's prospective payment system for inpatient rehabilitation care. DATA SOURCES/STUDY SETTING: Administrative data on costs and facility characteristics of inpatient rehabilitation facilities (IRFs) for calendar years 1998 and 1999. STUDY DESIGN: Compare standard linear regression and the Bayesian outlier accommodation model for developing facility payment adjustors for a prospective payment system. DATA COLLECTION: Variables describing facility average cost per case and facility characteristics were derived from several administrative data sources. PRINCIPAL FINDINGS: Evidence was found of non-normality of regression errors in the data used to develop facility payment adjustments for the inpatient rehabilitation facilities prospective payment system (IRF PPS). The Bayesian outlier accommodation model is shown to be appropriate for these data, but the model is largely consistent with the standard linear regression used in the development of the IRF PPS payment adjustors. CONCLUSIONS: The Bayesian outlier accommodation model is more robust to statistical outlier IRFs than standard linear regression for developing facility payment adjustments. It also allows for easy interpretation of model parameters, making it a viable policy alternative to standard regression in setting payment rates.

Bayes Theorem↗

PubMed2006

Biological systems modeling and analysis: a biomolecular technique of the twenty-first century.

It is proposed that computational systems biology should be considered a biomolecular technique of the twenty-first century, because it complements experimental biology and bioinformatics in unique ways that will eventually lead to insights and a depth of understanding not achievable without systems approaches. This article begins with a summary of traditional and novel modeling techniques. In the second part, it proposes concept map modeling as a useful link between experimental biology and biological systems modeling and analysis. Concept map modeling requires the collaboration between biologist and modeler. The biologist designs a regulated connectivity diagram of processes comprising a biological system and also provides semi-quantitative information on stimuli and measured or expected responses of the system. The modeler converts this information through methods of forward and inverse modeling into a mathematical construct that can be used for simulations and to generate and test new hypotheses. The biologist and the modeler collaboratively interpret the results and devise improved concept maps. The third part of the article describes software, BST-Box, supporting the various modeling activities.

Computational Biology↗

PubMed2000

A mathematical micturition model to restore simple flow recordings in healthy and symptomatic individuals and enhance uroflow interpretation.

We describe a model and report a new method to extract quantified data from the simple analysis of whole uroflow curves in healthy and symptomatic individuals. Recorded flow curves were compared with the curves theoretically predicted from a mathematical micturition model. This model was developed by relating each physiological event occurring during micturition to a set of mathematical equations. Due to improvements in speed of computer calculations, a very fast and adaptable mathematical micturition model became available. A total of 302 uroflow curves from 142 patients (61 men and 81 women) were studied. The control group consisted of seven men and 25 women; the symptomatic groups comprised 54 men and 56 women. The mathematical model was applied to analyze all the recorded curves. For patients with lower urinary tract symptoms, specific modelization parameters were introduced according to the clinical condition to be tested. Using two compulsory (patient sex and voided volume) and two optional parameters (clinical condition and urethral catheter size), our mathematical model was able to produce uroflow calculated curves similar to observed curves. In the control group, the calculated and recorded uroflow curves were found to superimpose with an impressive accuracy, i.e., a quadratic error <1%. Test-re-test studies gave the same determination of the specific parameters. In benign prostatic hyperplasia patients, the compressive effect on both prostatic urethra and bladder neck was separately identified. The same intra-individual values were found at 2-week intervals in the group with no treatment (P = no significance), whereas after 3 months of treatment with an alpha-blocker, a decrease in values was noted in responder patients (P < 0.001). In women with various degrees of cystocele, a constrictive effect was identified and found to be identical for successive flows during the same urodynamic testing. This large prospective study demonstrated the relevance of a sophisticated, heavily computerized micturition model, taking into account physiological voiding parameters, to the study of flow in healthy individuals and patients with benign prostatic hyperplasia or cystocele. Curve-fitting led to the determination of critical events during flow such as break point and plateau phase. Determination of these events may enhance uroflow interpretation by providing additional information on the detrusor function. Neurourol. Urodynam. 19:153-176, 2000.

Adult↗