Clustering with missing and left‐censored data: A simulation study comparing multiple‐imputation‐based procedures

Faucheux, Lilith; Resche‐Rigon, Matthieu; Curis, Emmanuel; Soumelis, Vassili; Chevret, Sylvie

doi:10.1002/bimj.201900366

Cited by 9 publications

(19 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(b) T distribution for ρ = 0.6 As expected, for M ≥ 2 the instability is constant whatever the proportion of missing values, the number of individuals or the correlation between variables. Similar results are observed when data are generated under a MAR mechanism or when data are imputed by FCS-RF (see Figures 9,10,11 in Appendix).…”

Section: Influence Of Msupporting

confidence: 76%

“…Like CSPA, this approach in two steps cannot be considered as based on the median partition problem [20]. Lately, [10] proposed consensus based on the MultiCons algorithm [27]. The algorithm presents many advantages, in particular it allows a visualization of the hidden cluster structure in the data set, but it does not aim at minimizing the median partition problem [p. 16] [27].…”

Section: Partitions Pooling After MImentioning

confidence: 99%

“…how to apply step 3) ? [9,10,11] brought some answers in terms of partitions pooling, but the question of the uncertainty measure has not been discussed.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Clustering with missing data: which equivalent for Rubin's rules?

Audigier¹,

Niang²

2020

Preprint

View full text Add to dashboard Cite

Multiple imputation (MI) is a popular method for dealing with missing values. However, the suitable way for applying clustering after MI remains unclear: how to pool partitions? How to assess the clustering instability when data are incomplete? By answering both questions, this paper proposed a complete view of clustering with missing data using MI. The problem of partitions pooling is here addressed using consensus clustering while, based on the bootstrap theory, we explain how to assess the instability related to observed and missing data. The new rules for pooling partitions and instability assessment are theoretically argued and extensively studied by simulation. Partitions pooling improves accuracy while measuring instability with missing data enlarges the data analysis possibilities: it allows assessment of the dependence of the clustering to the imputation model, as well as a convenient way for choosing the number of clusters when data are incomplete, as illustrated on a real data set.

show abstract

Section: Influence Of Msupporting

confidence: 76%

Section: Partitions Pooling After MImentioning

confidence: 99%

See 1 more Smart Citation

Clustering with missing data: which equivalent for Rubin's rules?

Audigier¹,

Niang²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Associations between circulating antibody levels and each categorical variable were assessed using the Wilcoxon–Mann–Whitney (2 levels variables) or Kruskal–Wallis (>2 levels variables) tests. Clinical patient profiles were determined using a clustering approach that handles missing data through multiple imputation and consensus clustering [ 10 ]. The variables used were circulating antibody levels, age, gender, need for oxygen therapy, pulmonary extension, BNP, D-dimers, lymphocyte count, ferritin, troponin, CRP, and NLR.…”

Section: Methodsmentioning

confidence: 99%

SARS-CoV-2 Antibodies in Hemodialysis Patients Six Months after Infection Compared to Healthcare Workers

Boulanger¹,

Saksi²,

Achiche³

et al. 2021

International Journal of Nephrology

View full text Add to dashboard Cite

Background. The humoral response to SARS-CoV-2 infection in hemodialysis patients needs to be clarified. Methods. In this retrospective study performed in two dialysis facilities, we measured the circulating levels of SARS-CoV-2 antibodies in patients who were on maintenance hemodialysis during the first wave of the epidemic in March and April 2020 and were still alive 6 months later. We also investigated associations between the patients diagnosed as infected during the first wave and several clinical, biological, and radiological parameters of COVID-19. Finally, we compared these circulating levels of SARS-CoV-2 antibodies with those of a control group of healthcare workers infected during the same period. Results. Of the 299 hemodialysis patients who recovered from the first wave of the epidemic 6 months before, 59 had a positive SARS-CoV-2 antibody whereas only 45 patients were diagnosed as infected during the first wave of the epidemic. All infected hemodialysis patients developed circulating antibodies. Using a clustering method, a significant correlation was identified between the cluster with the lowest circulating levels of SARS-CoV-2 antibodies and the severity of COVID-19 based on several parameters including CRP, BNP, lymphocyte count, neutrophil-lymphocyte ratio, and oxygen requirements, as well as pulmonary involvement on chest scan. Moreover, the circulating levels of the SARS-CoV-2 antibodies in surviving hemodialysis patients (n = 59) were similar to those of the control group (n = 17). Conclusion. The main finding of this study is that all of the surviving hemodialysis patients who were diagnosed with SARS-CoV-2 infection from March to April 2020 developed a persistent humoral response with significant circulating levels of SARS-CoV-2 antibodies, 6 months later. Another important finding is that surviving hemodialysis patients who had more severe disease had lower circulating levels of SARS-CoV-2 antibodies. Finally, circulating levels of SARS-CoV-2 antibodies were similar in surviving hemodialysis patients and healthcare workers without kidney disease.

show abstract

“…Approaches often make assumptions including few discrete time steps (Young et al 2018;Huopaniemi et al 2014); a single piecewise linear function (Young et al 2018) or Gaussian mixture model (Huopaniemi et al 2014); significantly more samples per object than number of objects (Mattar, Hanson, and Learned-Miller 2012;Gaffney and Smyth 2005); very small windows of potential misalignment (Liu, Tong, and Wheeler 2009;Listgarten et al 2007); or known lag time (Li et al 2011). Methods that directly measure similarity between time-series, e.g., dynamic time warping (Cuturi 2011) or methods that aggregate multiple imputation methods (Faucheux et al 2021) can also be used for clustering time-series data. Our method aims to cluster intervalcensored multivariate time series without these constraints.…”

Section: Related Workmentioning

confidence: 99%

Clustering Interval-Censored Time-Series for Disease Phenotyping

Chen¹,

Krishnan

Sontag³

2022

AAAI

View full text Add to dashboard Cite

Unsupervised learning is often used to uncover clusters in data. However, different kinds of noise may impede the discovery of useful patterns from real-world time-series data. In this work, we focus on mitigating the interference of interval censoring in the task of clustering for disease phenotyping. We develop a deep generative, continuous-time model of time-series data that clusters time-series while correcting for censorship time. We provide conditions under which clusters and the amount of delayed entry may be identified from data under a noiseless model. On synthetic data, we demonstrate accurate, stable, and interpretable results that outperform several benchmarks. On real-world clinical datasets of heart failure and Parkinson's disease patients, we study how interval censoring can adversely affect the task of disease phenotyping. Our model corrects for this source of error and recovers known clinical subtypes.

show abstract

Clustering with missing and left‐censored data: A simulation study comparing multiple‐imputation‐based procedures

Cited by 9 publications

References 40 publications

Clustering with missing data: which equivalent for Rubin's rules?

Clustering with missing data: which equivalent for Rubin's rules?

SARS-CoV-2 Antibodies in Hemodialysis Patients Six Months after Infection Compared to Healthcare Workers

Clustering Interval-Censored Time-Series for Disease Phenotyping

Contact Info

Product

Resources

About