A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data

Abstract The widespread adoption of Electronic Health Records (EHRs) and deep learning, particularly through Self-Supervised Representation Learning (SSRL) for categorical data, has transformed clinical decision-making. This scoping review, following PRISMA-ScR guidelines, examines 46 studies publis...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuanyuan Zheng, Adel Bensahla, Mina Bjelogrlic, Jamil Zaghir, Hugues Turbe, Lydie Bednarczyk, Christophe Gaudet-Blavignac, Julien Ehrsam, Stéphane Marchand-Maillet, Christian Lovis
Format: Article
Language:English
Published: Nature Portfolio 2025-06-01
Series:npj Digital Medicine
Online Access:https://doi.org/10.1038/s41746-025-01692-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849225926324256768
author Yuanyuan Zheng
Adel Bensahla
Mina Bjelogrlic
Jamil Zaghir
Hugues Turbe
Lydie Bednarczyk
Christophe Gaudet-Blavignac
Julien Ehrsam
Stéphane Marchand-Maillet
Christian Lovis
author_facet Yuanyuan Zheng
Adel Bensahla
Mina Bjelogrlic
Jamil Zaghir
Hugues Turbe
Lydie Bednarczyk
Christophe Gaudet-Blavignac
Julien Ehrsam
Stéphane Marchand-Maillet
Christian Lovis
author_sort Yuanyuan Zheng
collection DOAJ
description Abstract The widespread adoption of Electronic Health Records (EHRs) and deep learning, particularly through Self-Supervised Representation Learning (SSRL) for categorical data, has transformed clinical decision-making. This scoping review, following PRISMA-ScR guidelines, examines 46 studies published from January 2019 to April 2024, sourced from PubMed, MEDLINE, Embase, ACM, and Web of Science, focusing on SSRL for unlabeled categorical EHR data. The review systematically assesses research trends in building computationally and data-efficient representations for medical tasks, identifying major trends in model families: Transformer-based (43%), Autoencoder-based (28%), and Graph Neural Network-based (17%) models. The analysis highlights scenarios where healthcare institutions can leverage or develop SSRL technologies. It also addresses current limitations in assessing the impact of these technologies and identifies research opportunities to enhance their influence on clinical practice.
format Article
id doaj-art-96ec731b920049a8b9233e23c66417d0
institution Kabale University
issn 2398-6352
language English
publishDate 2025-06-01
publisher Nature Portfolio
record_format Article
series npj Digital Medicine
spelling doaj-art-96ec731b920049a8b9233e23c66417d02025-08-24T11:52:07ZengNature Portfolionpj Digital Medicine2398-63522025-06-018111510.1038/s41746-025-01692-1A scoping review of self-supervised representation learning for clinical decision making using EHR categorical dataYuanyuan Zheng0Adel Bensahla1Mina Bjelogrlic2Jamil Zaghir3Hugues Turbe4Lydie Bednarczyk5Christophe Gaudet-Blavignac6Julien Ehrsam7Stéphane Marchand-Maillet8Christian Lovis9Division of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDivision of Medical Information Sciences, Geneva University HospitalsDepartment of Computer Science, University of GenevaDivision of Medical Information Sciences, Geneva University HospitalsAbstract The widespread adoption of Electronic Health Records (EHRs) and deep learning, particularly through Self-Supervised Representation Learning (SSRL) for categorical data, has transformed clinical decision-making. This scoping review, following PRISMA-ScR guidelines, examines 46 studies published from January 2019 to April 2024, sourced from PubMed, MEDLINE, Embase, ACM, and Web of Science, focusing on SSRL for unlabeled categorical EHR data. The review systematically assesses research trends in building computationally and data-efficient representations for medical tasks, identifying major trends in model families: Transformer-based (43%), Autoencoder-based (28%), and Graph Neural Network-based (17%) models. The analysis highlights scenarios where healthcare institutions can leverage or develop SSRL technologies. It also addresses current limitations in assessing the impact of these technologies and identifies research opportunities to enhance their influence on clinical practice.https://doi.org/10.1038/s41746-025-01692-1
spellingShingle Yuanyuan Zheng
Adel Bensahla
Mina Bjelogrlic
Jamil Zaghir
Hugues Turbe
Lydie Bednarczyk
Christophe Gaudet-Blavignac
Julien Ehrsam
Stéphane Marchand-Maillet
Christian Lovis
A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
npj Digital Medicine
title A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
title_full A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
title_fullStr A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
title_full_unstemmed A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
title_short A scoping review of self-supervised representation learning for clinical decision making using EHR categorical data
title_sort scoping review of self supervised representation learning for clinical decision making using ehr categorical data
url https://doi.org/10.1038/s41746-025-01692-1
work_keys_str_mv AT yuanyuanzheng ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT adelbensahla ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT minabjelogrlic ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT jamilzaghir ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT huguesturbe ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT lydiebednarczyk ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT christophegaudetblavignac ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT julienehrsam ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT stephanemarchandmaillet ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT christianlovis ascopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT yuanyuanzheng scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT adelbensahla scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT minabjelogrlic scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT jamilzaghir scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT huguesturbe scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT lydiebednarczyk scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT christophegaudetblavignac scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT julienehrsam scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT stephanemarchandmaillet scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata
AT christianlovis scopingreviewofselfsupervisedrepresentationlearningforclinicaldecisionmakingusingehrcategoricaldata