EUVIMEDScopri EUVIMED
Open access

Clustering‐Informed Shared‐Structure Variational Autoencoder for Missing Data Imputation in Large‐Scale Healthcare Data

Yasin Khadem Charvadeh iD, Kenneth Seier, Katherine S Panageas, Danielle Vaithilingam, Mithat Gönen, Yuan Chen iD

Statistics in Medicine · 2025

DOI: 10.1002/sim.70335 ↗

At a glance

Journal article in Statistics in Medicine, 2025. Written by Yasin Khadem Charvadeh and 5 co-authors. Institution: Department of Epidemiology & Biostatistics, Memorial Sloan Kettering Cancer Center, New York, New York, USA. Major MEDLINE topics: Electronic Health Records, Neural Networks, Computer. Cited 1 times so far (Crossref); 30 references. The full text is freely available in PubMed Central. Funded by: NCI NIH HHS, National Cancer Institute.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Despite advancements in healthcare data management, missing data in electronic health records (EHR) and patient‐reported outcomes remain a persistent challenge, limiting their usability in healthcare analytics. Conventional imputation methods often struggle to capture complex nonlinear relationships, require extensive computation time, and are limited in addressing various types of missing data mechanisms. To overcome these challenges, we propose the clustering‐informed shared‐structure variational autoencoder (CISS‐VAE), which utilizes the strengths of Bayesian neural networks. This model can effectively capture complex associations and accommodate various missing data mechanisms, including missing not at random (MNAR). We also develop iterative learning algorithms that further enhance missing data imputation accuracy while preventing overfitting. Comprehensive simulations demonstrate the superior accuracy of our model compared to traditional and contemporary methods. We apply our method to EHR data from early‐stage breast cancer patients at Memorial Sloan Kettering Cancer Center, aiming to mitigate the impact of missing data and enhance health monitoring and analyses.

Subjects

MeSH terms (MEDLINE)

Electronic Health RecordsNeural Networks, ComputerHumansBayes TheoremCluster AnalysisAlgorithmsBreast NeoplasmsFemaleComputer SimulationModels, StatisticalPatient Reported Outcome MeasuresData Interpretation, StatisticalAutoencoder

Author keywords

electronic health recordsmissing data imputationmissing not at randomvariational autoencoder

Authors and institutions

  • Department of Epidemiology & Biostatistics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.

ORCID profiles: Yasin Khadem Charvadeh, Yuan Chen

Funding

  • NCI NIH HHS
  • National Cancer Institute

Cite

APA

Khadem Charvadeh, Y., Seier, K., Panageas, K. S., Vaithilingam, D., Gönen, M., & Chen, Y. (2025). Clustering‐Informed Shared‐Structure Variational Autoencoder for Missing Data Imputation in Large‐Scale Healthcare Data. Statistics in Medicine. https://doi.org/10.1002/sim.70335

Vancouver

Khadem Charvadeh Y, Seier K, Panageas KS, Vaithilingam D, Gönen M, Chen Y. Clustering‐Informed Shared‐Structure Variational Autoencoder for Missing Data Imputation in Large‐Scale Healthcare Data. Statistics in Medicine. 2025. doi:10.1002/sim.70335

BibTeX
@article{KhademCharvadeh2025,
  author = {Khadem Charvadeh, Yasin and Seier, Kenneth and Panageas, Katherine S and Vaithilingam, Danielle and Gönen, Mithat and Chen, Yuan},
  title = {Clustering‐Informed Shared‐Structure Variational Autoencoder for Missing Data Imputation in Large‐Scale Healthcare Data},
  journal = {Statistics in Medicine},
  year = {2025},
  doi = {10.1002/sim.70335}
}
Start for free →