Explainable machine-learning predictions for complications after pediatric congenital heart surgery

Abstract The quality of treatment and prognosis after pediatric congenital heart surgery remains unsatisfactory. A reliable prediction model for postoperative complications of congenital heart surgery patients is essential to enable prompt initiation of therapy and improve the quality of prognosis....

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Xian Zeng, Yaoqin Hu, Liqi Shu, Jianhua Li, Huilong Duan, Qiang Shu, Haomin Li
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/fd38e8681dfe4d99b77c9bb6b1928ada
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:fd38e8681dfe4d99b77c9bb6b1928ada
record_format	dspace
spelling	oai:doaj.org-article:fd38e8681dfe4d99b77c9bb6b1928ada2021-12-02T18:53:09ZExplainable machine-learning predictions for complications after pediatric congenital heart surgery10.1038/s41598-021-96721-w2045-2322https://doaj.org/article/fd38e8681dfe4d99b77c9bb6b1928ada2021-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-96721-whttps://doaj.org/toc/2045-2322Abstract The quality of treatment and prognosis after pediatric congenital heart surgery remains unsatisfactory. A reliable prediction model for postoperative complications of congenital heart surgery patients is essential to enable prompt initiation of therapy and improve the quality of prognosis. Here, we develop an interpretable machine-learning-based model that integrates patient demographics, surgery-specific features and intraoperative blood pressure data for accurately predicting complications after pediatric congenital heart surgery. We used blood pressure variability and the k-means algorithm combined with a smoothed formulation of dynamic time wrapping to extract features from time-series data. In addition, SHAP framework was used to provide explanations of the prediction. Our model achieved the best performance both in binary and multi-label classification compared with other consensus-based risk models. In addition, this explainable model explains why a prediction was made to help improve the clinical understanding of complication risk and generate actionable knowledge in practice. The combination of model performance and interpretability is easy for clinicians to trust and provide insight into how they should respond before the condition worsens after pediatric congenital heart surgery.Xian ZengYaoqin HuLiqi ShuJianhua LiHuilong DuanQiang ShuHaomin LiNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Xian Zeng Yaoqin Hu Liqi Shu Jianhua Li Huilong Duan Qiang Shu Haomin Li Explainable machine-learning predictions for complications after pediatric congenital heart surgery
description	Abstract The quality of treatment and prognosis after pediatric congenital heart surgery remains unsatisfactory. A reliable prediction model for postoperative complications of congenital heart surgery patients is essential to enable prompt initiation of therapy and improve the quality of prognosis. Here, we develop an interpretable machine-learning-based model that integrates patient demographics, surgery-specific features and intraoperative blood pressure data for accurately predicting complications after pediatric congenital heart surgery. We used blood pressure variability and the k-means algorithm combined with a smoothed formulation of dynamic time wrapping to extract features from time-series data. In addition, SHAP framework was used to provide explanations of the prediction. Our model achieved the best performance both in binary and multi-label classification compared with other consensus-based risk models. In addition, this explainable model explains why a prediction was made to help improve the clinical understanding of complication risk and generate actionable knowledge in practice. The combination of model performance and interpretability is easy for clinicians to trust and provide insight into how they should respond before the condition worsens after pediatric congenital heart surgery.
format	article
author	Xian Zeng Yaoqin Hu Liqi Shu Jianhua Li Huilong Duan Qiang Shu Haomin Li
author_facet	Xian Zeng Yaoqin Hu Liqi Shu Jianhua Li Huilong Duan Qiang Shu Haomin Li
author_sort	Xian Zeng
title	Explainable machine-learning predictions for complications after pediatric congenital heart surgery
title_short	Explainable machine-learning predictions for complications after pediatric congenital heart surgery
title_full	Explainable machine-learning predictions for complications after pediatric congenital heart surgery
title_fullStr	Explainable machine-learning predictions for complications after pediatric congenital heart surgery
title_full_unstemmed	Explainable machine-learning predictions for complications after pediatric congenital heart surgery
title_sort	explainable machine-learning predictions for complications after pediatric congenital heart surgery
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/fd38e8681dfe4d99b77c9bb6b1928ada
work_keys_str_mv	AT xianzeng explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT yaoqinhu explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT liqishu explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT jianhuali explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT huilongduan explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT qiangshu explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery AT haominli explainablemachinelearningpredictionsforcomplicationsafterpediatriccongenitalheartsurgery
_version_	1718377365120221184

Explainable machine-learning predictions for complications after pediatric congenital heart surgery

Ejemplares similares