USE OF ARTIFICIAL INTELLIGENCE FOR PREOPERATIVE ANAESTHESIA EVALUATION – A SYSTEMATIC REVIEW

Abstract

AIM:

This systematically review aims to evaluate the application of artificial intelligence (AI) in pre-operative anaesthesia evaluation and assess its effectiveness in enhancing clinical decision-making and patient safety.

MATERIALS AND METHOD:

A systematic review with total number of 662 articles were searched using PubMed, SCOPUS, Elsevier Science Direct, Wiley online library, Cochrane library, Prospero, and web of science was conducted using PRISMA guidelines. Studies focused on AI applications in pre-operative anaesthesia evaluation using machine learning (ML) or deep learning (DL). Data extracted includes study design, AI model used, sample size, and clinical application. Eight studies meeting inclusion criteria were included for analysis. Bias assessment was done using Cochrane-based assessment tool 2 (ROB2).

RESULTS:

The included studies demonstrated a wide range of AI applications, such as ASA classification, difficult airway prediction, and risk stratification using platforms like MySurgeryRisk. Machine learning was the predominant technique, with deep learning used in select studies. AI models achieved high sensitivity and specificity in predicting perioperative complications and were accepted by clinicians. Sample sizes varied, reflecting both exploratory and large-scale validation efforts.

CONCLUSION:

AI shows strong potential to enhance the accuracy, efficiency, and personalization of pre-operative anaesthesia assessments. Integration of AI into routine practice may improve perioperative outcomes, streamline workflows, and support global standardization in anaesthetic care.