Mixing Consistent Deep Clustering

Lutscher, Daniel; Hassouni, Ali el; Stol, Maarten; Hoogendoorn, Mark

doi:10.48550/arxiv.2011.01977

Cited by 2 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, some have recognised its importance 4-6,24-27 , although often focussing on reproducibility rather than generalisability 6,25 , or without using external data 7,26 . Furthermore, we built upon the growing body of research on deep clustering 3,17,19,[28][29][30] by creating X-DEC. The generalisability of clustering models on a general ICU patient population shows that clusters could potentially be used for patient strati cation in clinical research and, eventually, guide clinicians in designing patient-speci c care pathways.…”

Section: Discussionmentioning

confidence: 99%

Deep embedded clustering generalisability and adaptation for mixed datatypes: two critical care cohorts

de Kok,

Rosmalen,

Koeze

et al. 2023

Preprint

View full text Add to dashboard Cite

We propose X-DEC, a novel deep clustering technique that can integrate mixed datatypes (in this study numerical and categorical variables). Deep Embedded Clustering (DEC) is a promising technique capable of managing extensive sets of variables and non-linear relationships. Nevertheless, DEC cannot adequately handle mixed datatypes. Therefore, we created X-DEC by replacing the autoencoder with an X-shaped variational autoencoder (XVAE) and optimising hyperparameters for cluster stability. We compared DEC and X-DEC by reproducing a previous study that used DEC to identify clusters in a population of intensive care patients. We assessed internal validity based on cluster stability on the development dataset. Since generalisability of clustering models has insufficiently been validated on external populations, we assessed external validity by investigating cluster generalisability onto an external validation dataset. We concluded that both DEC and X-DEC resulted in clinically recognisable and generalisable clusters, but X-DEC produced much more stable clusters.

show abstract

Section: Discussionmentioning

confidence: 99%

Deep embedded clustering generalisability and adaptation for mixed datatypes: two critical care cohorts

de Kok,

Rosmalen,

Koeze

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

Deep embedded clustering generalisability and adaptation for integrating mixed datatypes: two critical care cohorts

de Kok,

van Rosmalen,

Koeze

et al. 2024

Sci Rep

View full text Add to dashboard Cite

We validated a Deep Embedded Clustering (DEC) model and its adaptation for integrating mixed datatypes (in this study, numerical and categorical variables). Deep Embedded Clustering (DEC) is a promising technique capable of managing extensive sets of variables and non-linear relationships. Nevertheless, DEC cannot adequately handle mixed datatypes. Therefore, we adapted DEC by replacing the autoencoder with an X-shaped variational autoencoder (XVAE) and optimising hyperparameters for cluster stability. We call this model “X-DEC”. We compared DEC and X-DEC by reproducing a previous study that used DEC to identify clusters in a population of intensive care patients. We assessed internal validity based on cluster stability on the development dataset. Since generalisability of clustering models has insufficiently been validated on external populations, we assessed external validity by investigating cluster generalisability onto an external validation dataset. We concluded that both DEC and X-DEC resulted in clinically recognisable and generalisable clusters, but X-DEC produced much more stable clusters.

show abstract

Mixing Consistent Deep Clustering

Cited by 2 publications

References 0 publications

Deep embedded clustering generalisability and adaptation for mixed datatypes: two critical care cohorts

Deep embedded clustering generalisability and adaptation for mixed datatypes: two critical care cohorts

Deep embedded clustering generalisability and adaptation for integrating mixed datatypes: two critical care cohorts

Contact Info

Product

Resources

About