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 433 records · Page 24Linked to original sources
PubMed1979

Metabolism of the acutely ischemic dog heart. II. Interpretation of a model.

The glycolytic oscillations occurring in an acutely ischemic dog heart are analyzed with a computer model. The major regulations of the glycolytic pathway flux occur at phosphohexose isomerase, which is inhibited by accumulated pentose shunt intermediates; at phosphorylase, which shapes the first cycle of the oscillation; and at aldolase, which shapes the last two cycles. Aldolase is not under normal substrate control. Its activity, and that of some subsequent glycolytic enzymes, appears to be regulated by known interactions with the muscle proteins. The mitochondria become reduced as a result of anoxia, and their metabolism reorganizes to export rather than import reducing equivalents. It is in general feasible to account for the behavior of this preparation in terms of the known metabolism of less severely perturbed hearts, especially (but not completely) in terms of effects of anoxia. The reasons for the inapplicability of the crossover theorem previously used to analyze this preparation are described.

Animals↗
PubMed1986

Model analysis of steady-state ventilatory response to CO2 into component factors.

A method is described to partition measured values of steady-state ventilatory response into an estimation of the blood flow in the respiratory controller and the sensitivity of the controller to CO2 assuming proportional control. The analysis is derived from the describing equations of a computer model and leads to the definition of a grid of lines emanating from a hypothetical reference point at negative ventilation and zero central nervous system metabolism. Data from the literature reporting differences in CO2 response among normal subjects and changes in resting ventilation and cerebral blood flow with age are reinterpreted from this perspective. Use of a structural model to interpret physiological data is shown to give a different meaning to data reduction in contrast to interpretation using statistical models like regression.

Adult↗
PubMed2026

Cancer of unknown primary: the evolution of tissue of origin identification in the artificial intelligence era.

Cancer of Unknown Primary (CUP) presents substantial diagnostic and therapeutic challenges owing to its heterogeneous nature and the absence of an identifiable primary tumor site. This review provides a structured search of the pathogenesis, epidemiological characteristics, and limitations of traditional diagnostic and therapeutic approaches for CUP, with an emphasis on the evolution of Tissue of Origin (TOO) identification techniques. Recent advances in precision medicine have accelerated the development of machine learning-based TOO identification tools, representing a paradigm shift in CUP diagnostics. Deep learning (DL) algorithms that integrate multi-omics data (such as genomics and transcriptomics) with clinical features have markedly enhanced the accuracy of tracing tumor origin, and artificial intelligence (AI) driven TOO models are increasingly being incorporated into clinical practice, offering new insights for pathological diagnosis, treatment selection, and prognostic evaluation. Nevertheless, several challenges remain, including issues of data standardization, model generalizability, and interpretability. Ethical considerations related to data privacy, algorithmic fairness, and clinical implementation also warrant careful attention. Future research should focus on establishing standardized multi-center databases, developing more interpretable AI models, and fostering multidisciplinary collaborative strategies for CUP management. Through continued refinement of technical solutions and regulatory guidelines, TOO identification is anticipated to progress from research to routine clinical application, ultimately supporting precise and personalized care for patients with CUP.

Artificial intelligence↗
PubMed2001

F X Nottingham and F X Taunton. Two novel mutations in factor X resulting in loss of functional activity and an interpretation using molecular modelling.

We report two novel mutations in the Factor X gene which result in a bleeding tendency in two unrelated Caucasian families. Although the mutations occur at adjacent codons in exon 8 and result in reduced functional activity with normal antigen levels, the patterns of inheritance appear to be quite distinct. Factor X Nottingham (alanine 404 threonine) appears to be associated with an autosomal recessive pattern of inheritance. In contrast, Factor X Taunton (arginine 405 glycine) results in a mode of inheritance consistent with an autosomal dominant pattern, all five of the heterozygotes in this family being clinically affected. Molecular modelling studies suggest that, in the case of Factor X Nottingham, a drastic conformational change causes major unfolding of the protein. For Factor X Taunton, less extreme conformational changes occur causing loss of functional activity such that substrate binding sites might be maintained. It is proposed that competition with wild type for substrate binding could occur leading to a dominant negative effect.

Adolescent↗
PubMed2002

Guiding molecules towards drug-likeness.

This review discusses computational methods for the prediction of drug-likeness. The coverage of published works include the assessment of historical practices of lead generation and optimization, surveys of the properties of known drugs and their constituent fragments and scaffolds, methods for delineating drug space, optimization techniques for simultaneously enhancing multiple properties and drug-like characteristics, similarity metrics and the application of more advanced pattern recognition algorithms for the prediction of drug-likeness. Areas which could be improved in this field are the scope of the datasets used to build models, the chemical interpretability of models, the use of multivariate optimization methods for drug design and the application of underappreciated statistical methods proven to work in other fields.

Animals↗