Related Experiment Video
Updated: Jun 29, 2025
05:39
Author Spotlight: A Non-Intubated Video-Assisted Thoracoscopic Surgery with Multimodal Analgesia and Sevoflurane Inhalation Anesthesia
Published on: May 26, 2023
1.5K
A predictive model for post-thoracoscopic surgery pulmonary complications based on the PBNN algorithm.
Cheng-Mao Zhou1,2, Qiong Xue3, HuiJuan Li3
1Big Data and Artificial Intelligence Research Group, Department of Anaesthesiology, Central People's Hospital of Zhanjiang, Zhanjiang, Guangdong, China. zhouchengmao187@foxmail.com.
Scientific Reports
|March 26, 2024
Summary
Machine learning models predict pulmonary complications after thoracoscopic surgery. The pruning Bayesian neural network shows promise for identifying high-risk patients before their procedure.
Area of Science:
- Medical Informatics
- Thoracic Surgery
- Artificial Intelligence in Medicine
Background:
- Postoperative pulmonary complications (PPCs) are a significant concern after thoracoscopic surgery.
- Accurate prediction of PPCs can aid in risk stratification and patient management.
Purpose of the Study:
- To develop and evaluate machine learning and deep learning models for early prediction of PPCs after thoracoscopic surgery.
- To identify key factors associated with PPCs.
Main Methods:
- Construction of artificial intelligence (AI) prediction models using Python, incorporating machine learning and deep learning algorithms.
- Correlation analysis to identify factors associated with PPCs.
- Evaluation of various AI algorithms including Logistic Regression, Light Gradient Boosting Machine (LGBM), and Pruning Bayesian Neural Network (PBNN).
Main Results:
- Age, surgery duration, and serum albumin were correlated with PPCs.
- Key predictive factors identified by LGBM included single lung ventilation duration, smoking history, surgery duration, ASA score, and blood glucose.
- Pruning Bayesian Neural Network (PBNN) demonstrated strong performance in AUC (0.869) and F1 score (0.566).
Conclusions:
- AI models, particularly PBNN, can effectively predict the likelihood of PPCs after thoracoscopy.
- These models can assist in identifying high-risk individuals preoperatively, enabling timely interventions.
