Machine learning for identification of short-term all-cause and cardiovascular deaths among patients undergoing peritoneal dialysis

Although more and more cardiovascular risk factors have been verified in peritoneal dialysis (PD) populations in different countries and regions, it is still difficult for clinicians to accurately and individually predict death in the near future. We aimed to develop and validate machine learning-based models to predict near-term all-cause and cardiovascular death. Machine learning models were developed among 7539 PD patients, which were randomly divided into a training set and an internal test set by five random shuffles of 5-fold cross-validation, to predict the cardiovascular death and all-cause death in 3 months. We chose objectively collected markers such as patient demographics, clinical characteristics, laboratory data, and dialysis-related variables to inform the models and assessed the predictive performance using a range of common performance metrics, such as sensitivity, positive predictive values, the area under the receiver operating curve (AUROC), and the area under the precision recall curve. In the test set, the CVDformer models had a AUROC of 0.8767 (0.8129, 0.9045) and 0.9026 (0.8404, 0.9352) and area under the precision recall curve of 0.9338 (0.8134,0.9453) and 0.9073 (0.8412, 0.9164) in predicting near-term all-cause death and cardiovascular death, respectively. The CVDformer models had high sensitivity and positive predictive values for predicting all-cause and cardiovascular deaths in 3 months in our PD population. Further calibration is warranted in the future.

Authors

Institutions

Publication Details

Journal
Clinical Kidney Journal
Published
2024-08-29
DOI
https://doi.org/10.1093/ckj/sfae242
Citations
5
Primary Topic
Dialysis and Renal Disease Management
Type
article
Field-Weighted Citation Impact
1.70

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Machine learning for identification of short-term all-cause and cardiovascular deaths among patients undergoing peritoneal dialysis

Beiru Zhang, Shutong Du, Lihua Wang, Liping Duan et al.
5 citations
Clinical Kidney Journal
Dialysis and Renal Disease Management
1.70
article

Machine learning for identification of short-term all-cause and cardiovascular deaths among patients undergoing peritoneal dialysis

Beiru Zhang, Shutong Du, Lihua Wang, Liping Duan, Yan Li, Qiongzhen Lin, Tianrong Ji, Yumei Liao, Lirong Deng, Caili Wang, Yulan Shen, Zhigang Ma, Yirong Liu, Shanshan Guo, Jie Dong, Gang Fu, Fuyun Sun, Zhiyuan Xu, Li Zuo, Xuejian Wang, Zhanzheng Zhao, Hongyu Chen, Hao Li, Li Li, Lihua Wang, Xiaoli Chen, Yan Yan, Hongyi Li, Ying Li, Shanshan Chen, Yong Zhang, Xiao Xu, Jing Xiao, Zhonggao Xu, Wenbo Hu, Zhiqiang He, Yulin Li, Guiling Liu, Xiaoying Ma, Tiantian Ma, Kaijun Niu, Xuanyi Du, Jingwei Zhou, Yingping Li, Huiping Zhao, Shaomei Li, Xianchao Zhang, Yingchun Ma, Xinying Gao, Yingli Yue, Zhao Cui, Zhaoxia Zheng, Huayi Pei
article en
5 citations

Abstract

Although more and more cardiovascular risk factors have been verified in peritoneal dialysis (PD) populations in different countries and regions, it is still difficult for clinicians to accurately and individually predict death in the near future. We aimed to develop and validate machine learning-based models to predict near-term all-cause and cardiovascular death. Machine learning models were developed among 7539 PD patients, which were randomly divided into a training set and an internal test set by five random shuffles of 5-fold cross-validation, to predict the cardiovascular death and all-cause death in 3 months. We chose objectively collected markers such as patient demographics, clinical characteristics, laboratory data, and dialysis-related variables to inform the models and assessed the predictive performance using a range of common performance metrics, such as sensitivity, positive predictive values, the area under the receiver operating curve (AUROC), and the area under the precision recall curve. In the test set, the CVDformer models had a AUROC of 0.8767 (0.8129, 0.9045) and 0.9026 (0.8404, 0.9352) and area under the precision recall curve of 0.9338 (0.8134,0.9453) and 0.9073 (0.8412, 0.9164) in predicting near-term all-cause death and cardiovascular death, respectively. The CVDformer models had high sensitivity and positive predictive values for predicting all-cause and cardiovascular deaths in 3 months in our PD population. Further calibration is warranted in the future.

Clinical Kidney JournalVol. 17(9)
Hebei Medical University (CN), King University (US), Handan College (CN), Ministry of Education of the People's Republic of China (CN), Beijing University of Posts and Telecommunications (CN), Army Medical University (CN), Pingdingshan University (CN), Harbin Medical University (CN), Shanxi Medical University (CN), Jilin University (CN), Anhui Medical University (CN), Peking University (CN), China Rehabilitation Research Center (CN), Third Hospital of Hebei Medical University (CN), Hospital of Hebei Province (CN), Ganzhou People's Hospital (CN), Second Hospital of Hebei Medical University (CN), Cangzhou Central Hospital (CN), People's Hospital of Cangzhou (CN), Ministry of Health of the Russian Federation (RU), Xinqiao Hospital (CN), Dongzhimen Hospital Affiliated to Beijing University of Chinese Medicine (CN), Peking University People's Hospital (CN), First Hospital of Jilin University (CN), Peking University Shenzhen Hospital (CN), Peking University First Hospital (CN), Second Affiliated Hospital of Harbin Medical University (CN), Second Hospital of Shanxi Medical University (CN), Taiyuan Central Hospital (CN), First People's Hospital of Nanning (CN), Second Affiliated Hospital of Anhui Medical University (CN), First Affiliated Hospital of Zhengzhou University (CN), Second People’s Hospital of Yibin (CN), Baotou Medical College (CN), Beijing Haidian Hospital (CN), China Medical University (CN)
Peking University First Hospital, Peking University, Army Medical University, Hebei Medical University
Good health and well-being
Openalex Percentile: Top 14%
Dialysis and Renal Disease Management
1.70
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.