Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation

Abstract Aims Assessing the risk for HF rehospitalization is important for managing and treating patients with HF. To address this need, various risk prediction models have been developed. However, none of them used deep learning methods with real‐world data. This study aimed to develop a deep learn...

Full description

Saved in:
Bibliographic Details
Main Authors: Mi‐Na Kim, Yong Seok Lee, Youngmin Park, Ayoung Jung, Hanjee So, Joonwoong Park, Jin‐Joo Park, Dong‐Joo Choi, So‐Ree Kim, Seong‐Mi Park
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:ESC Heart Failure
Subjects:
Online Access:https://doi.org/10.1002/ehf2.14918
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850105503192973312
author Mi‐Na Kim
Yong Seok Lee
Youngmin Park
Ayoung Jung
Hanjee So
Joonwoong Park
Jin‐Joo Park
Dong‐Joo Choi
So‐Ree Kim
Seong‐Mi Park
author_facet Mi‐Na Kim
Yong Seok Lee
Youngmin Park
Ayoung Jung
Hanjee So
Joonwoong Park
Jin‐Joo Park
Dong‐Joo Choi
So‐Ree Kim
Seong‐Mi Park
author_sort Mi‐Na Kim
collection DOAJ
description Abstract Aims Assessing the risk for HF rehospitalization is important for managing and treating patients with HF. To address this need, various risk prediction models have been developed. However, none of them used deep learning methods with real‐world data. This study aimed to develop a deep learning‐based prediction model for HF rehospitalization within 30, 90, and 365 days after acute HF (AHF) discharge. Methods and results We analysed the data of patients admitted due to AHF between January 2014 and January 2019 in a tertiary hospital. In performing deep learning‐based predictive algorithms for HF rehospitalization, we use hyperbolic tangent activation layers followed by recurrent layers with gated recurrent units. To assess the readmission prediction, we used the AUC, precision, recall, specificity, and F1 measure. We applied the Shapley value to identify which features contributed to HF readmission. Twenty‐two prognostic features exhibiting statistically significant associations with HF rehospitalization were identified, consisting of 6 time‐independent and 16 time‐dependent features. The AUC value shows moderate discrimination for predicting readmission within 30, 90, and 365 days of follow‐up (FU) (AUC:0.63, 0.74, and 0.76, respectively). The features during the FU have a relatively higher contribution to HF rehospitalization than features from other time points. Conclusions Our deep learning‐based model using real‐world data could provide valid predictions of HF rehospitalization in 1 year follow‐up. It can be easily utilized to guide appropriate interventions or care strategies for patients with HF. The closed monitoring and blood test in daily clinics are important for assessing the risk of HF rehospitalization.
format Article
id doaj-art-b3f208eb489243c9aa5b6c64a1008bbb
institution OA Journals
issn 2055-5822
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series ESC Heart Failure
spelling doaj-art-b3f208eb489243c9aa5b6c64a1008bbb2025-08-20T02:39:04ZengWileyESC Heart Failure2055-58222024-12-011163702371210.1002/ehf2.14918Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validationMi‐Na Kim0Yong Seok Lee1Youngmin Park2Ayoung Jung3Hanjee So4Joonwoong Park5Jin‐Joo Park6Dong‐Joo Choi7So‐Ree Kim8Seong‐Mi Park9Department of Internal Medicine, Division of Cardiology, Anam Hospital Korea University Medicine Seoul KoreaData Analytics Group, Samsung SDS Seoul KoreaData Analytics Group, Samsung SDS Seoul KoreaData Analytics Group, Samsung SDS Seoul KoreaData Analytics Group, Samsung SDS Seoul KoreaData Analytics Group, Samsung SDS Seoul KoreaDepartment of Internal Medicine, Division of Cardiology Seoul National University Bundang Hospital Seongnam KoreaDepartment of Internal Medicine, Division of Cardiology Seoul National University Bundang Hospital Seongnam KoreaDepartment of Internal Medicine, Division of Cardiology, Anam Hospital Korea University Medicine Seoul KoreaDepartment of Internal Medicine, Division of Cardiology, Anam Hospital Korea University Medicine Seoul KoreaAbstract Aims Assessing the risk for HF rehospitalization is important for managing and treating patients with HF. To address this need, various risk prediction models have been developed. However, none of them used deep learning methods with real‐world data. This study aimed to develop a deep learning‐based prediction model for HF rehospitalization within 30, 90, and 365 days after acute HF (AHF) discharge. Methods and results We analysed the data of patients admitted due to AHF between January 2014 and January 2019 in a tertiary hospital. In performing deep learning‐based predictive algorithms for HF rehospitalization, we use hyperbolic tangent activation layers followed by recurrent layers with gated recurrent units. To assess the readmission prediction, we used the AUC, precision, recall, specificity, and F1 measure. We applied the Shapley value to identify which features contributed to HF readmission. Twenty‐two prognostic features exhibiting statistically significant associations with HF rehospitalization were identified, consisting of 6 time‐independent and 16 time‐dependent features. The AUC value shows moderate discrimination for predicting readmission within 30, 90, and 365 days of follow‐up (FU) (AUC:0.63, 0.74, and 0.76, respectively). The features during the FU have a relatively higher contribution to HF rehospitalization than features from other time points. Conclusions Our deep learning‐based model using real‐world data could provide valid predictions of HF rehospitalization in 1 year follow‐up. It can be easily utilized to guide appropriate interventions or care strategies for patients with HF. The closed monitoring and blood test in daily clinics are important for assessing the risk of HF rehospitalization.https://doi.org/10.1002/ehf2.14918Deep learningHeart failureRehospitalizationRisk assessment
spellingShingle Mi‐Na Kim
Yong Seok Lee
Youngmin Park
Ayoung Jung
Hanjee So
Joonwoong Park
Jin‐Joo Park
Dong‐Joo Choi
So‐Ree Kim
Seong‐Mi Park
Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
ESC Heart Failure
Deep learning
Heart failure
Rehospitalization
Risk assessment
title Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
title_full Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
title_fullStr Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
title_full_unstemmed Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
title_short Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation
title_sort deep learning for predicting rehospitalization in acute heart failure model foundation and external validation
topic Deep learning
Heart failure
Rehospitalization
Risk assessment
url https://doi.org/10.1002/ehf2.14918
work_keys_str_mv AT minakim deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT yongseoklee deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT youngminpark deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT ayoungjung deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT hanjeeso deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT joonwoongpark deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT jinjoopark deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT dongjoochoi deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT soreekim deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation
AT seongmipark deeplearningforpredictingrehospitalizationinacuteheartfailuremodelfoundationandexternalvalidation