Evaluation of imputation strategies for multi-centre studies: Application to a large clinical pathology dataset

Lucy Grigoroff; Reika Masuda; John Lindon; Janonna Kadyrov; Jeremy K Nicholson; Elaine Holmes; Julien Wist

doi:10.1371/journal.pone.0335852

Back

Evaluation of imputation strategies for multi-centre studies: Application to a large clinical pathology dataset

Journal article

Open access

Peer reviewed

Evaluation of imputation strategies for multi-centre studies: Application to a large clinical pathology dataset

Lucy Grigoroff, Reika Masuda, John Lindon, Janonna Kadyrov, Jeremy K Nicholson, Elaine Holmes and Julien Wist

PloS one, Vol.20(11), e0335852

2025

DOI: https://doi.org/10.1371/journal.pone.0335852

PMID: 41264609

Files and links (1)

pdf

Published2.35 MBDownload View

Published (Version of Record)CC BY V4.0, Open Access

Abstract

As part of a strategy for accommodating missing data in large heterogeneous datasets, two Random Forest-based (RF) imputation methods, missForest and MICE were evaluated along with several strategies to help navigate the inherently incomplete structure of the dataset. Background: A total of 3817 complete cases of clinical chemistry variables from a large-scale, multi-site preclinical longitudinal pathology study were used as an evaluation dataset. Three types of ‘missingness’ in various proportions were artificially introduced to compare imputation performance for different strategies including variable inclusion and stratification. Results: MissForest was found to outperform MICE, being robust and capable of automatic variable selection. Stratification had minimal effect on missForest but severely deteriorated the performance of MICE. Conclusion: In general, storing and sharing datasets prior to any correction is a good practise, so that imputation can be performed on merged data if necessary.

Details

Title: Evaluation of imputation strategies for multi-centre studies: Application to a large clinical pathology dataset
Authors/Creators: Lucy Grigoroff - Australian National Phenome Centre, and Centre for Computational and Systems Medicine, Health Futures Institute, Murdoch University, Perth, Australia
Reika Masuda - Murdoch University
John Lindon - Imperial College London
Janonna Kadyrov - Murdoch University
Jeremy K Nicholson - Imperial College London
Elaine Holmes - Imperial College London
Julien Wist - Chemistry Department, Universidad del Valle, Cali, Colombia
Publication Details: PloS one, Vol.20(11), e0335852
Publisher: Public Library of Science
Identifiers: 991005830208207891
Murdoch Affiliation: Health Futures Institute; Centre for Computational and Systems Medicine
Language: English
Resource Type: Journal article

UN Sustainable Development Goals (SDGs)

This output has contributed to the advancement of the following goals:

Metrics

1 File views/ downloads

9 Record Views