Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386495836> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4386495836 endingPage "107803" @default.
- W4386495836 startingPage "107803" @default.
- W4386495836 abstract "Reusing Electronic Health Records (EHRs) for Machine Learning (ML) leads on many occasions to extremely incomplete and sparse tabular datasets, which can hinder the model development processes and limit their performance and generalization. In this study, we aimed to characterize the most effective data imputation techniques and ML models for dealing with highly missing numerical data in EHRs, in the case where only a very limited number of data are complete, as opposed to the usual case of having a reduced number of missing values. We used a case study including full blood count laboratory data, demographic and survival data in the context of COVID-19 hospital admissions and evaluated 30 processing pipelines combining imputation methods with ML classifiers. The imputation methods included missing mask, translation and encoding, mean imputation, k-nearest neighbors’ imputation, Bayesian ridge regression imputation and generative adversarial imputation networks. The classifiers included k-nearest neighbors, logistic regression, random forest, gradient boosting and deep multilayer perceptron. Our results suggest that in the presence of highly missing data, combining translation and encoding imputation—which considers informative missingness—with tree ensemble classifiers—random forest and gradient boosting—is a sensible choice when aiming to maximize performance, in terms of area under curve. Based on our findings, we recommend the consideration of this imputer-classifier configuration when constructing models in the presence of extremely incomplete numerical data in EHR." @default.
- W4386495836 created "2023-09-07" @default.
- W4386495836 creator A5009929703 @default.
- W4386495836 creator A5017610355 @default.
- W4386495836 creator A5026025555 @default.
- W4386495836 creator A5043527279 @default.
- W4386495836 creator A5045453819 @default.
- W4386495836 creator A5057240190 @default.
- W4386495836 creator A5069033015 @default.
- W4386495836 creator A5071504728 @default.
- W4386495836 date "2023-12-01" @default.
- W4386495836 modified "2023-10-09" @default.
- W4386495836 title "Extremely missing numerical data in Electronic Health Records for machine learning can be managed through simple imputation methods considering informative missingness: a comparative of solutions in a COVID-19 mortality case study" @default.
- W4386495836 cites W1677182931 @default.
- W4386495836 cites W1678356000 @default.
- W4386495836 cites W2052635053 @default.
- W4386495836 cites W2095095223 @default.
- W4386495836 cites W2122111042 @default.
- W4386495836 cites W2125677766 @default.
- W4386495836 cites W2146292423 @default.
- W4386495836 cites W2152575748 @default.
- W4386495836 cites W2155653793 @default.
- W4386495836 cites W2257979135 @default.
- W4386495836 cites W2341770358 @default.
- W4386495836 cites W2342249984 @default.
- W4386495836 cites W2801490189 @default.
- W4386495836 cites W2919115771 @default.
- W4386495836 cites W3096413426 @default.
- W4386495836 cites W3096831136 @default.
- W4386495836 cites W3109352657 @default.
- W4386495836 cites W3113005448 @default.
- W4386495836 cites W3163032678 @default.
- W4386495836 cites W3170657538 @default.
- W4386495836 cites W3177828909 @default.
- W4386495836 cites W3185195079 @default.
- W4386495836 cites W3186462793 @default.
- W4386495836 cites W3197782790 @default.
- W4386495836 cites W3216683616 @default.
- W4386495836 cites W4200310129 @default.
- W4386495836 cites W4318273625 @default.
- W4386495836 doi "https://doi.org/10.1016/j.cmpb.2023.107803" @default.
- W4386495836 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37703700" @default.
- W4386495836 hasPublicationYear "2023" @default.
- W4386495836 type Work @default.
- W4386495836 citedByCount "0" @default.
- W4386495836 crossrefType "journal-article" @default.
- W4386495836 hasAuthorship W4386495836A5009929703 @default.
- W4386495836 hasAuthorship W4386495836A5017610355 @default.
- W4386495836 hasAuthorship W4386495836A5026025555 @default.
- W4386495836 hasAuthorship W4386495836A5043527279 @default.
- W4386495836 hasAuthorship W4386495836A5045453819 @default.
- W4386495836 hasAuthorship W4386495836A5057240190 @default.
- W4386495836 hasAuthorship W4386495836A5069033015 @default.
- W4386495836 hasAuthorship W4386495836A5071504728 @default.
- W4386495836 hasBestOaLocation W43864958361 @default.
- W4386495836 hasConcept C105795698 @default.
- W4386495836 hasConcept C119857082 @default.
- W4386495836 hasConcept C124101348 @default.
- W4386495836 hasConcept C151956035 @default.
- W4386495836 hasConcept C154945302 @default.
- W4386495836 hasConcept C169258074 @default.
- W4386495836 hasConcept C33923547 @default.
- W4386495836 hasConcept C41008148 @default.
- W4386495836 hasConcept C46686674 @default.
- W4386495836 hasConcept C58041806 @default.
- W4386495836 hasConcept C9357733 @default.
- W4386495836 hasConceptScore W4386495836C105795698 @default.
- W4386495836 hasConceptScore W4386495836C119857082 @default.
- W4386495836 hasConceptScore W4386495836C124101348 @default.
- W4386495836 hasConceptScore W4386495836C151956035 @default.
- W4386495836 hasConceptScore W4386495836C154945302 @default.
- W4386495836 hasConceptScore W4386495836C169258074 @default.
- W4386495836 hasConceptScore W4386495836C33923547 @default.
- W4386495836 hasConceptScore W4386495836C41008148 @default.
- W4386495836 hasConceptScore W4386495836C46686674 @default.
- W4386495836 hasConceptScore W4386495836C58041806 @default.
- W4386495836 hasConceptScore W4386495836C9357733 @default.
- W4386495836 hasLocation W43864958361 @default.
- W4386495836 hasLocation W43864958362 @default.
- W4386495836 hasOpenAccess W4386495836 @default.
- W4386495836 hasPrimaryLocation W43864958361 @default.
- W4386495836 hasRelatedWork W1574575415 @default.
- W4386495836 hasRelatedWork W2024529227 @default.
- W4386495836 hasRelatedWork W2081476516 @default.
- W4386495836 hasRelatedWork W2181530120 @default.
- W4386495836 hasRelatedWork W2581984549 @default.
- W4386495836 hasRelatedWork W3028371478 @default.
- W4386495836 hasRelatedWork W3144172081 @default.
- W4386495836 hasRelatedWork W3179858851 @default.
- W4386495836 hasRelatedWork W4211215373 @default.
- W4386495836 hasRelatedWork W3123177881 @default.
- W4386495836 hasVolume "242" @default.
- W4386495836 isParatext "false" @default.
- W4386495836 isRetracted "false" @default.
- W4386495836 workType "article" @default.