Artificial intelligence and machine learning applications in ambulatory surgery - A systematic review.

Area of Science:

• Medical Informatics
• Surgical Technology
• Machine Learning Applications

Background:

• Artificial intelligence (AI) technologies are increasingly explored for optimizing patient care in ambulatory surgery settings.
• A systematic review was conducted to understand the current landscape of AI applications in this domain.

Purpose of the Study:

• To systematically review and synthesize the applications of artificial intelligence (AI) technologies specifically for ambulatory surgical patients.
• To identify key themes, methodologies, and outcomes of AI use in ambulatory surgery.

Main Methods:

• A systematic literature search was performed across major databases: PubMed, Scopus, Web of Science, and EBSCOhost, covering publications from 2015 to 2025.
• Studies were included if they utilized artificial intelligence techniques within ambulatory surgical populations.

Main Results:

• Out of 26 identified studies, 25 employed machine learning (ML), with a significant focus on orthopaedic procedures (65.3%).
• Key application themes included preoperative patient selection (e.g., Random Forest, XGBoost), same-day discharge prediction (ensemble models showing high AUC), postoperative management (e.g., predicting opioid refill needs), and cost prediction (ensemble models superior to single models).
• The majority of studies originated in the USA, and common ML algorithms included Random Forest, eXtreme Gradient Boost, and Artificial Neural Networks.

Conclusions:

• Machine learning, especially ensemble methods, demonstrates significant potential for enhancing ambulatory surgery.
• Identified challenges include model inconsistencies, data-related issues, and a need for robust external validation to ensure generalizability and reliability.
• Further development and validation are crucial for the widespread clinical integration of AI in ambulatory surgical care.