Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble
Objective: To determine the most influential data features and to develop machine learning approaches that best predict hospital readmissions among patients with diabetes. Methods: In this retrospective cohort study, we surveyed patient statistics and performed feature analysis to identify the most...
Published in: | Asian Pacific Journal of Tropical Medicine |
---|---|
Main Authors: | , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Wolters Kluwer Medknow Publications
2021
|
Subjects: | |
Online Access: | https://doi.org/10.4103/1995-7645.326254 https://doaj.org/article/9b452c07ad8d4d189d637c3b373c071b |
id |
ftdoajarticles:oai:doaj.org/article:9b452c07ad8d4d189d637c3b373c071b |
---|---|
record_format |
openpolar |
spelling |
ftdoajarticles:oai:doaj.org/article:9b452c07ad8d4d189d637c3b373c071b 2023-05-15T15:07:38+02:00 Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble Xiayu Xiang Chuanyi Liu Yanchun Zhang Wei Xiang Binxing Fang 2021-01-01T00:00:00Z https://doi.org/10.4103/1995-7645.326254 https://doaj.org/article/9b452c07ad8d4d189d637c3b373c071b EN eng Wolters Kluwer Medknow Publications http://www.apjtm.org/article.asp?issn=1995-7645;year=2021;volume=14;issue=9;spage=417;epage=428;aulast=Xiang https://doaj.org/toc/2352-4146 2352-4146 doi:10.4103/1995-7645.326254 https://doaj.org/article/9b452c07ad8d4d189d637c3b373c071b Asian Pacific Journal of Tropical Medicine, Vol 14, Iss 9, Pp 417-428 (2021) electronic health records hospital readmissions feature analysis predictive models imbalanced learning diabetes Arctic medicine. Tropical medicine RC955-962 article 2021 ftdoajarticles https://doi.org/10.4103/1995-7645.326254 2022-12-30T20:39:31Z Objective: To determine the most influential data features and to develop machine learning approaches that best predict hospital readmissions among patients with diabetes. Methods: In this retrospective cohort study, we surveyed patient statistics and performed feature analysis to identify the most influential data features associated with readmissions. Classification of all-cause, 30-day readmission outcomes were modeled using logistic regression, artificial neural network, and EasyEnsemble. F1 statistic, sensitivity, and positive predictive value were used to evaluate the model performance. Results: We identified 14 most influential data features (4 numeric features and 10 categorical features) and evaluated 3 machine learning models with numerous sampling methods (oversampling, undersampling, and hybrid techniques). The deep learning model offered no improvement over traditional models (logistic regression and EasyEnsemble) for predicting readmission, whereas the other two algorithms led to much smaller differences between the training and testing datasets. Conclusions: Machine learning approaches to record electronic health data offer a promising method for improving readmission prediction in patients with diabetes. But more work is needed to construct datasets with more clinical variables beyond the standard risk factors and to fine-tune and optimize machine learning models. Article in Journal/Newspaper Arctic Directory of Open Access Journals: DOAJ Articles Arctic Asian Pacific Journal of Tropical Medicine 14 9 417 |
institution |
Open Polar |
collection |
Directory of Open Access Journals: DOAJ Articles |
op_collection_id |
ftdoajarticles |
language |
English |
topic |
electronic health records hospital readmissions feature analysis predictive models imbalanced learning diabetes Arctic medicine. Tropical medicine RC955-962 |
spellingShingle |
electronic health records hospital readmissions feature analysis predictive models imbalanced learning diabetes Arctic medicine. Tropical medicine RC955-962 Xiayu Xiang Chuanyi Liu Yanchun Zhang Wei Xiang Binxing Fang Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
topic_facet |
electronic health records hospital readmissions feature analysis predictive models imbalanced learning diabetes Arctic medicine. Tropical medicine RC955-962 |
description |
Objective: To determine the most influential data features and to develop machine learning approaches that best predict hospital readmissions among patients with diabetes. Methods: In this retrospective cohort study, we surveyed patient statistics and performed feature analysis to identify the most influential data features associated with readmissions. Classification of all-cause, 30-day readmission outcomes were modeled using logistic regression, artificial neural network, and EasyEnsemble. F1 statistic, sensitivity, and positive predictive value were used to evaluate the model performance. Results: We identified 14 most influential data features (4 numeric features and 10 categorical features) and evaluated 3 machine learning models with numerous sampling methods (oversampling, undersampling, and hybrid techniques). The deep learning model offered no improvement over traditional models (logistic regression and EasyEnsemble) for predicting readmission, whereas the other two algorithms led to much smaller differences between the training and testing datasets. Conclusions: Machine learning approaches to record electronic health data offer a promising method for improving readmission prediction in patients with diabetes. But more work is needed to construct datasets with more clinical variables beyond the standard risk factors and to fine-tune and optimize machine learning models. |
format |
Article in Journal/Newspaper |
author |
Xiayu Xiang Chuanyi Liu Yanchun Zhang Wei Xiang Binxing Fang |
author_facet |
Xiayu Xiang Chuanyi Liu Yanchun Zhang Wei Xiang Binxing Fang |
author_sort |
Xiayu Xiang |
title |
Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
title_short |
Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
title_full |
Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
title_fullStr |
Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
title_full_unstemmed |
Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble |
title_sort |
predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and easyensemble |
publisher |
Wolters Kluwer Medknow Publications |
publishDate |
2021 |
url |
https://doi.org/10.4103/1995-7645.326254 https://doaj.org/article/9b452c07ad8d4d189d637c3b373c071b |
geographic |
Arctic |
geographic_facet |
Arctic |
genre |
Arctic |
genre_facet |
Arctic |
op_source |
Asian Pacific Journal of Tropical Medicine, Vol 14, Iss 9, Pp 417-428 (2021) |
op_relation |
http://www.apjtm.org/article.asp?issn=1995-7645;year=2021;volume=14;issue=9;spage=417;epage=428;aulast=Xiang https://doaj.org/toc/2352-4146 2352-4146 doi:10.4103/1995-7645.326254 https://doaj.org/article/9b452c07ad8d4d189d637c3b373c071b |
op_doi |
https://doi.org/10.4103/1995-7645.326254 |
container_title |
Asian Pacific Journal of Tropical Medicine |
container_volume |
14 |
container_issue |
9 |
container_start_page |
417 |
_version_ |
1766339100341698560 |