Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV
Introduction<p>Bayesian approaches to patient phenotyping in clinical observational studies have been limited by the computational challenges associated with applying the Markov Chain Monte Carlo (MCMC) approach to real-world data. Approximate Bayesian inference via optimization of the variati...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | , |
| Published: |
2024
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1852026235531558912 |
|---|---|
| author | Brian Buckley (137185) |
| author2 | Adrian O'Hagan (19779114) Marie Galligan (4402525) |
| author2_role | author author |
| author_facet | Brian Buckley (137185) Adrian O'Hagan (19779114) Marie Galligan (4402525) |
| author_role | author |
| dc.creator.none.fl_str_mv | Brian Buckley (137185) Adrian O'Hagan (19779114) Marie Galligan (4402525) |
| dc.date.none.fl_str_mv | 2024-10-02T04:22:10Z |
| dc.identifier.none.fl_str_mv | 10.3389/fams.2024.1302825.s001 |
| dc.relation.none.fl_str_mv | https://figshare.com/articles/dataset/Data_Sheet_1_Variational_Bayes_latent_class_analysis_for_EHR-based_phenotyping_with_large_real-world_data_CSV/27148170 |
| dc.rights.none.fl_str_mv | CC BY 4.0 info:eu-repo/semantics/openAccess |
| dc.subject.none.fl_str_mv | Statistics Computation Theory and Mathematics Ordinary Differential Equations, Difference Equations and Dynamical Systems Financial Mathematics Applied Mathematics not elsewhere classified Optimisation Numerical and Computational Mathematics not elsewhere classified Applied Statistics Mathematical Physics not elsewhere classified Numerical Computation Computation Theory and Mathematics not elsewhere classified variational Bayes latent class analysis patient phenotyping real-world evidence electronic health records |
| dc.title.none.fl_str_mv | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| dc.type.none.fl_str_mv | Dataset info:eu-repo/semantics/publishedVersion dataset |
| description | Introduction<p>Bayesian approaches to patient phenotyping in clinical observational studies have been limited by the computational challenges associated with applying the Markov Chain Monte Carlo (MCMC) approach to real-world data. Approximate Bayesian inference via optimization of the variational evidence lower bound, variational Bayes (VB), has been successfully demonstrated for other applications.</p>Methods<p>We investigate the performance and characteristics of currently available VB and MCMC software to explore the practicability of available approaches and provide guidance for clinical practitioners. Two case studies are used to fully explore the methods covering a variety of real-world data. First, we use the publicly available Pima Indian diabetes data to comprehensively compare VB implementations of logistic regression. Second, a large real-world data set, Optum™ EHR with approximately one million diabetes patients extended the analysis to large, highly unbalanced data containing discrete and continuous variables. A Bayesian patient phenotyping composite model incorporating latent class analysis (LCA) and regression was implemented with the second case study.</p>Results<p>We find that several data characteristics common in clinical data, such as sparsity, significantly affect the posterior accuracy of automatic VB methods compared with conditionally conjugate mean-field methods. We find that for both models, automatic VB approaches require more effort and technical knowledge to set up for accurate posterior estimation and are very sensitive to stopping time compared with closed-form VB methods.</p>Discussion<p>Our results indicate that the patient phenotyping composite Bayes model is more easily usable for real-world studies if Monte Carlo is replaced with VB. It can potentially become a uniquely useful tool for decision support, especially for rare diseases where gold-standard biomarker data are sparse but prior knowledge can be used to assist model diagnosis and may suggest when biomarker tests are warranted.</p> |
| eu_rights_str_mv | openAccess |
| id | Manara_a5ea95df3698de5272fa774cb7e4cede |
| identifier_str_mv | 10.3389/fams.2024.1302825.s001 |
| network_acronym_str | Manara |
| network_name_str | ManaraRepo |
| oai_identifier_str | oai:figshare.com:article/27148170 |
| publishDate | 2024 |
| repository.mail.fl_str_mv | |
| repository.name.fl_str_mv | |
| repository_id_str | |
| rights_invalid_str_mv | CC BY 4.0 |
| spelling | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSVBrian Buckley (137185)Adrian O'Hagan (19779114)Marie Galligan (4402525)StatisticsComputation Theory and MathematicsOrdinary Differential Equations, Difference Equations and Dynamical SystemsFinancial MathematicsApplied Mathematics not elsewhere classifiedOptimisationNumerical and Computational Mathematics not elsewhere classifiedApplied StatisticsMathematical Physics not elsewhere classifiedNumerical ComputationComputation Theory and Mathematics not elsewhere classifiedvariational Bayeslatent class analysispatient phenotypingreal-world evidenceelectronic health recordsIntroduction<p>Bayesian approaches to patient phenotyping in clinical observational studies have been limited by the computational challenges associated with applying the Markov Chain Monte Carlo (MCMC) approach to real-world data. Approximate Bayesian inference via optimization of the variational evidence lower bound, variational Bayes (VB), has been successfully demonstrated for other applications.</p>Methods<p>We investigate the performance and characteristics of currently available VB and MCMC software to explore the practicability of available approaches and provide guidance for clinical practitioners. Two case studies are used to fully explore the methods covering a variety of real-world data. First, we use the publicly available Pima Indian diabetes data to comprehensively compare VB implementations of logistic regression. Second, a large real-world data set, Optum™ EHR with approximately one million diabetes patients extended the analysis to large, highly unbalanced data containing discrete and continuous variables. A Bayesian patient phenotyping composite model incorporating latent class analysis (LCA) and regression was implemented with the second case study.</p>Results<p>We find that several data characteristics common in clinical data, such as sparsity, significantly affect the posterior accuracy of automatic VB methods compared with conditionally conjugate mean-field methods. We find that for both models, automatic VB approaches require more effort and technical knowledge to set up for accurate posterior estimation and are very sensitive to stopping time compared with closed-form VB methods.</p>Discussion<p>Our results indicate that the patient phenotyping composite Bayes model is more easily usable for real-world studies if Monte Carlo is replaced with VB. It can potentially become a uniquely useful tool for decision support, especially for rare diseases where gold-standard biomarker data are sparse but prior knowledge can be used to assist model diagnosis and may suggest when biomarker tests are warranted.</p>2024-10-02T04:22:10ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.3389/fams.2024.1302825.s001https://figshare.com/articles/dataset/Data_Sheet_1_Variational_Bayes_latent_class_analysis_for_EHR-based_phenotyping_with_large_real-world_data_CSV/27148170CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/271481702024-10-02T04:22:10Z |
| spellingShingle | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV Brian Buckley (137185) Statistics Computation Theory and Mathematics Ordinary Differential Equations, Difference Equations and Dynamical Systems Financial Mathematics Applied Mathematics not elsewhere classified Optimisation Numerical and Computational Mathematics not elsewhere classified Applied Statistics Mathematical Physics not elsewhere classified Numerical Computation Computation Theory and Mathematics not elsewhere classified variational Bayes latent class analysis patient phenotyping real-world evidence electronic health records |
| status_str | publishedVersion |
| title | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| title_full | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| title_fullStr | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| title_full_unstemmed | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| title_short | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| title_sort | Data_Sheet_1_Variational Bayes latent class analysis for EHR-based phenotyping with large real-world data.CSV |
| topic | Statistics Computation Theory and Mathematics Ordinary Differential Equations, Difference Equations and Dynamical Systems Financial Mathematics Applied Mathematics not elsewhere classified Optimisation Numerical and Computational Mathematics not elsewhere classified Applied Statistics Mathematical Physics not elsewhere classified Numerical Computation Computation Theory and Mathematics not elsewhere classified variational Bayes latent class analysis patient phenotyping real-world evidence electronic health records |