A Real-Time Artificial Intelligence-Assisted System to Predict Weaning from Ventilator Immediately after Lung...

Area of Science:

• Anesthesiology
• Artificial Intelligence
• Medical Informatics

Background:

• Pre-operative risk assessment for lung resection surgery is crucial for determining post-operative ventilator weaning.
• Time constraints in pre-anesthetic clinics limit comprehensive risk evaluations by anesthesiologists.
• Accurate prediction of ventilator weaning is essential for patient management and resource allocation.

Purpose of the Study:

• To develop and validate an artificial intelligence (AI) model for predicting immediate post-lung resection ventilator weaning.
• To create a practical AI application to assist anesthesiologists in pre-anesthetic clinics.
• To enhance physician-patient communication regarding surgical risks and outcomes.

Main Methods:

• Retrospective analysis of electronic medical records from 709 lung resection patients (January 2017 - July 2019).
• Construction and evaluation of seven supervised machine learning algorithms for prediction.
• Identification of the Naïve Bayes Classifier as the optimal algorithm for the AI model.

Main Results:

• The AI model utilizing the Naïve Bayes Classifier demonstrated the highest predictive accuracy.
• The developed AI application effectively utilizes patient data for risk assessment.
• The model's performance indicates potential for improved clinical decision-making.

Conclusions:

• An AI-driven approach can accurately predict ventilator weaning post-lung resection.
• The AI application offers a valuable tool for anesthesiologists in pre-anesthetic assessments.
• Digitalized risk assessment through AI can enhance patient understanding and communication.