Neural Temporal Point Processes For Modelling Electronic Health Records

Joseph Enguehard, Dan Busbridge, Adam Bozson, Claire Woodcock, Nils Hammerla
Proceedings of the Machine Learning for Health NeurIPS Workshop, PMLR 136:85-113, 2020.

Abstract

The modelling of Electronic Health Records (EHRs) has the potential to drive more efficient allocation of healthcare resources, enabling early intervention strategies and advancing personalised healthcare. However, EHRs are challenging to model due to their realisation as noisy, multi-modal data occurring at irregular time intervals. To address their temporal nature, we treat EHRs as samples generated by a Temporal Point Process (TPP), enabling us to model what happened in an event with when it happened in a principled way. We gather and propose neural network parameterisations of TPPs, collectively referred to as Neural TPPs. We perform evaluations on synthetic EHRs as well as on a set of established benchmarks. We show that TPPs significantly outperform their non-TPP counterparts on EHRs. We also show that an assumption of many Neural TPPs, that the class distribution is conditionally independent of time, reduces performance on EHRs. Finally, our proposed attention-based Neural TPP performs favourably compared to existing models, whilst aligning with real world interpretability requirements, an important step towards a component of clinical decision support systems.

Cite this Paper


BibTeX
@InProceedings{pmlr-v136-enguehard20a, title = {Neural Temporal Point Processes For Modelling Electronic Health Records}, author = {Enguehard, Joseph and Busbridge, Dan and Bozson, Adam and Woodcock, Claire and Hammerla, Nils}, booktitle = {Proceedings of the Machine Learning for Health NeurIPS Workshop}, pages = {85--113}, year = {2020}, editor = {Alsentzer, Emily and McDermott, Matthew B. A. and Falck, Fabian and Sarkar, Suproteem K. and Roy, Subhrajit and Hyland, Stephanie L.}, volume = {136}, series = {Proceedings of Machine Learning Research}, month = {11 Dec}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v136/enguehard20a/enguehard20a.pdf}, url = {https://proceedings.mlr.press/v136/enguehard20a.html}, abstract = {The modelling of Electronic Health Records (EHRs) has the potential to drive more efficient allocation of healthcare resources, enabling early intervention strategies and advancing personalised healthcare. However, EHRs are challenging to model due to their realisation as noisy, multi-modal data occurring at irregular time intervals. To address their temporal nature, we treat EHRs as samples generated by a Temporal Point Process (TPP), enabling us to model what happened in an event with when it happened in a principled way. We gather and propose neural network parameterisations of TPPs, collectively referred to as Neural TPPs. We perform evaluations on synthetic EHRs as well as on a set of established benchmarks. We show that TPPs significantly outperform their non-TPP counterparts on EHRs. We also show that an assumption of many Neural TPPs, that the class distribution is conditionally independent of time, reduces performance on EHRs. Finally, our proposed attention-based Neural TPP performs favourably compared to existing models, whilst aligning with real world interpretability requirements, an important step towards a component of clinical decision support systems.} }
Endnote
%0 Conference Paper %T Neural Temporal Point Processes For Modelling Electronic Health Records %A Joseph Enguehard %A Dan Busbridge %A Adam Bozson %A Claire Woodcock %A Nils Hammerla %B Proceedings of the Machine Learning for Health NeurIPS Workshop %C Proceedings of Machine Learning Research %D 2020 %E Emily Alsentzer %E Matthew B. A. McDermott %E Fabian Falck %E Suproteem K. Sarkar %E Subhrajit Roy %E Stephanie L. Hyland %F pmlr-v136-enguehard20a %I PMLR %P 85--113 %U https://proceedings.mlr.press/v136/enguehard20a.html %V 136 %X The modelling of Electronic Health Records (EHRs) has the potential to drive more efficient allocation of healthcare resources, enabling early intervention strategies and advancing personalised healthcare. However, EHRs are challenging to model due to their realisation as noisy, multi-modal data occurring at irregular time intervals. To address their temporal nature, we treat EHRs as samples generated by a Temporal Point Process (TPP), enabling us to model what happened in an event with when it happened in a principled way. We gather and propose neural network parameterisations of TPPs, collectively referred to as Neural TPPs. We perform evaluations on synthetic EHRs as well as on a set of established benchmarks. We show that TPPs significantly outperform their non-TPP counterparts on EHRs. We also show that an assumption of many Neural TPPs, that the class distribution is conditionally independent of time, reduces performance on EHRs. Finally, our proposed attention-based Neural TPP performs favourably compared to existing models, whilst aligning with real world interpretability requirements, an important step towards a component of clinical decision support systems.
APA
Enguehard, J., Busbridge, D., Bozson, A., Woodcock, C. & Hammerla, N.. (2020). Neural Temporal Point Processes For Modelling Electronic Health Records. Proceedings of the Machine Learning for Health NeurIPS Workshop, in Proceedings of Machine Learning Research 136:85-113 Available from https://proceedings.mlr.press/v136/enguehard20a.html.

Related Material