Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning

BackgroundPatients with diabetes face an increased risk of postoperative pulmonary infection (PPI). However, precise predictive models specific to this patient group are lacking.ObjectiveTo develop and validate a machine learning model for predicting PPI risk in patients with diabetes.MethodsThis re...

Full description

Saved in:
Bibliographic Details
Main Authors: Chunxiu Zhao, Bingbing Xiang, Jie Zhang, Pingliang Yang, Qiaoli Liu, Shun Wang
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-12-01
Series:Frontiers in Physiology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fphys.2024.1501854/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850138607067594752
author Chunxiu Zhao
Bingbing Xiang
Jie Zhang
Pingliang Yang
Qiaoli Liu
Shun Wang
author_facet Chunxiu Zhao
Bingbing Xiang
Jie Zhang
Pingliang Yang
Qiaoli Liu
Shun Wang
author_sort Chunxiu Zhao
collection DOAJ
description BackgroundPatients with diabetes face an increased risk of postoperative pulmonary infection (PPI). However, precise predictive models specific to this patient group are lacking.ObjectiveTo develop and validate a machine learning model for predicting PPI risk in patients with diabetes.MethodsThis retrospective study enrolled 1,269 patients with diabetes who underwent elective non-cardiac, non-neurological surgeries at our institution from January 2020 to December 2023. Predictive models were constructed using nine different machine learning algorithms. Feature selection was conducted using Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression. Model performance was assessed via the Area Under the Curve (AUC), precision, accuracy, specificity and F1-score.ResultsThe Ada Boost classifier (ADA) model exhibited the best performance with an AUC of 0.901, Accuracy of 0.91, Precision of 0.82, specificity of 0.98, PPV of 0.82, and NPV of 0.82. LASSO feature selection identified six optimal predictive factors: postoperative transfer to the ICU, Age, American Society of Anesthesiologists (ASA) physical status score, chronic obstructive pulmonary disease (COPD) status, surgical department, and duration of surgery.ConclusionOur study developed a robust predictive model using six clinical features, offering a valuable tool for clinical decision-making and personalized prevention strategies for PPI in patients with diabetes.
format Article
id doaj-art-cebcaffb6d834bad8f19a94cea383ebb
institution OA Journals
issn 1664-042X
language English
publishDate 2024-12-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Physiology
spelling doaj-art-cebcaffb6d834bad8f19a94cea383ebb2025-08-20T02:30:32ZengFrontiers Media S.A.Frontiers in Physiology1664-042X2024-12-011510.3389/fphys.2024.15018541501854Predicting postoperative pulmonary infection risk in patients with diabetes using machine learningChunxiu Zhao0Bingbing Xiang1Jie Zhang2Pingliang Yang3Qiaoli Liu4Shun Wang5Department of Critical Care Medicine, Affiliated Hospital of Southwest Jiaotong University, The Third People's Hospital of Chengdu, Chengdu, Sichuan, ChinaDepartment of Anesthesiology, West China Hospital, Sichuan University, Chengdu, ChinaDepartment of Anesthesiology, Clinical Medical College and The First Affiliated Hospital of Chengdu Medical College, Chengdu, Sichuan, ChinaDepartment of Anesthesiology, Clinical Medical College and The First Affiliated Hospital of Chengdu Medical College, Chengdu, Sichuan, ChinaDepartment of Anesthesiology, Clinical Medical College and The First Affiliated Hospital of Chengdu Medical College, Chengdu, Sichuan, ChinaDepartment of Anesthesiology, Clinical Medical College and The First Affiliated Hospital of Chengdu Medical College, Chengdu, Sichuan, ChinaBackgroundPatients with diabetes face an increased risk of postoperative pulmonary infection (PPI). However, precise predictive models specific to this patient group are lacking.ObjectiveTo develop and validate a machine learning model for predicting PPI risk in patients with diabetes.MethodsThis retrospective study enrolled 1,269 patients with diabetes who underwent elective non-cardiac, non-neurological surgeries at our institution from January 2020 to December 2023. Predictive models were constructed using nine different machine learning algorithms. Feature selection was conducted using Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression. Model performance was assessed via the Area Under the Curve (AUC), precision, accuracy, specificity and F1-score.ResultsThe Ada Boost classifier (ADA) model exhibited the best performance with an AUC of 0.901, Accuracy of 0.91, Precision of 0.82, specificity of 0.98, PPV of 0.82, and NPV of 0.82. LASSO feature selection identified six optimal predictive factors: postoperative transfer to the ICU, Age, American Society of Anesthesiologists (ASA) physical status score, chronic obstructive pulmonary disease (COPD) status, surgical department, and duration of surgery.ConclusionOur study developed a robust predictive model using six clinical features, offering a valuable tool for clinical decision-making and personalized prevention strategies for PPI in patients with diabetes.https://www.frontiersin.org/articles/10.3389/fphys.2024.1501854/fulldiabetes mellituspostoperative pulmonary infectionmachine learningrisk predictionAda Boost classifier
spellingShingle Chunxiu Zhao
Bingbing Xiang
Jie Zhang
Pingliang Yang
Qiaoli Liu
Shun Wang
Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
Frontiers in Physiology
diabetes mellitus
postoperative pulmonary infection
machine learning
risk prediction
Ada Boost classifier
title Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
title_full Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
title_fullStr Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
title_full_unstemmed Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
title_short Predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
title_sort predicting postoperative pulmonary infection risk in patients with diabetes using machine learning
topic diabetes mellitus
postoperative pulmonary infection
machine learning
risk prediction
Ada Boost classifier
url https://www.frontiersin.org/articles/10.3389/fphys.2024.1501854/full
work_keys_str_mv AT chunxiuzhao predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning
AT bingbingxiang predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning
AT jiezhang predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning
AT pingliangyang predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning
AT qiaoliliu predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning
AT shunwang predictingpostoperativepulmonaryinfectionriskinpatientswithdiabetesusingmachinelearning