Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning
Abstract Introduction Unsupervised feature learning methods inspired by natural language processing (NLP) models are capable of constructing patient-specific features from longitudinal Electronic Health Records (EHR). Design We applied document embedding algorithms to real-world paediatric intensive...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-01-01
|
| Series: | BMC Medical Informatics and Decision Making |
| Online Access: | https://doi.org/10.1186/s12911-024-02812-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571589100568576 |
|---|---|
| author | Jingteng Li Kimberley R. Zakka John Booth Louise Rigny Samiran Ray Mario Cortina-Borja Payam Barnaghi Neil Sebire |
| author_facet | Jingteng Li Kimberley R. Zakka John Booth Louise Rigny Samiran Ray Mario Cortina-Borja Payam Barnaghi Neil Sebire |
| author_sort | Jingteng Li |
| collection | DOAJ |
| description | Abstract Introduction Unsupervised feature learning methods inspired by natural language processing (NLP) models are capable of constructing patient-specific features from longitudinal Electronic Health Records (EHR). Design We applied document embedding algorithms to real-world paediatric intensive care (PICU) EHR data to extract patient-specific features from 1853 patients’ PICU journeys using 647 unique lab tests and medication events. We evaluated the clinical utility of the patient features via a K-means clustering analysis. Results We trained a document embedding model under a unique evaluation pipeline and obtained latent patient feature vectors for all 1853 patients. We performed unsupervised clustering to the patient vectors as a downstream analysis and obtained 5 distinct clusters via hyperparameter optimisation. Significant variations (p<0.0001) within both patient characteristics and surgery intervention and diagnostic profiles were detected. Conclusion The K-means clustering results demonstrated the clinical utilities of the patient-specific features learned from the embedding algorithms. The latent patient features obtained via the embedding process enabled direct applications of other machine learning algorithms. Future work will focus on utilising the temporal information within EHR and extending EHR embedding algorithms to develop personalised patient journey predictions. |
| format | Article |
| id | doaj-art-7057d7bba11c4cd9a567c6905322979f |
| institution | Kabale University |
| issn | 1472-6947 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Informatics and Decision Making |
| spelling | doaj-art-7057d7bba11c4cd9a567c6905322979f2025-02-02T12:27:48ZengBMCBMC Medical Informatics and Decision Making1472-69472025-01-0125111310.1186/s12911-024-02812-9Discovering patient groups in sequential electronic healthcare data using unsupervised representation learningJingteng Li0Kimberley R. Zakka1John Booth2Louise Rigny3Samiran Ray4Mario Cortina-Borja5Payam Barnaghi6Neil Sebire7Great Ormond Street Institute of Child Health, University College LondonGreat Ormond Street Institute of Child Health, University College LondonGreat Ormond Street Institute of Child Health, University College LondonData Research Innovation and Virtual Environment, Great Ormond Street Hospital for ChildrenGreat Ormond Street Institute of Child Health, University College LondonGreat Ormond Street Institute of Child Health, University College LondonGreat Ormond Street Institute of Child Health, University College LondonGreat Ormond Street Institute of Child Health, University College LondonAbstract Introduction Unsupervised feature learning methods inspired by natural language processing (NLP) models are capable of constructing patient-specific features from longitudinal Electronic Health Records (EHR). Design We applied document embedding algorithms to real-world paediatric intensive care (PICU) EHR data to extract patient-specific features from 1853 patients’ PICU journeys using 647 unique lab tests and medication events. We evaluated the clinical utility of the patient features via a K-means clustering analysis. Results We trained a document embedding model under a unique evaluation pipeline and obtained latent patient feature vectors for all 1853 patients. We performed unsupervised clustering to the patient vectors as a downstream analysis and obtained 5 distinct clusters via hyperparameter optimisation. Significant variations (p<0.0001) within both patient characteristics and surgery intervention and diagnostic profiles were detected. Conclusion The K-means clustering results demonstrated the clinical utilities of the patient-specific features learned from the embedding algorithms. The latent patient features obtained via the embedding process enabled direct applications of other machine learning algorithms. Future work will focus on utilising the temporal information within EHR and extending EHR embedding algorithms to develop personalised patient journey predictions.https://doi.org/10.1186/s12911-024-02812-9 |
| spellingShingle | Jingteng Li Kimberley R. Zakka John Booth Louise Rigny Samiran Ray Mario Cortina-Borja Payam Barnaghi Neil Sebire Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning BMC Medical Informatics and Decision Making |
| title | Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| title_full | Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| title_fullStr | Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| title_full_unstemmed | Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| title_short | Discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| title_sort | discovering patient groups in sequential electronic healthcare data using unsupervised representation learning |
| url | https://doi.org/10.1186/s12911-024-02812-9 |
| work_keys_str_mv | AT jingtengli discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT kimberleyrzakka discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT johnbooth discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT louiserigny discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT samiranray discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT mariocortinaborja discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT payambarnaghi discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning AT neilsebire discoveringpatientgroupsinsequentialelectronichealthcaredatausingunsupervisedrepresentationlearning |