Leveraging AI-based prediction in perioperative and critical care: from model development to clinical implementation

Meijden, S.L. van der (2025)

Doctoral Thesis

In this dissertation, the potential of Artificial Intelligence (AI) for predicting patient outcomes in perioperative and critical care was investigated. The chapters of this dissertation align with the AI development and evaluation trajectory, ranging from (data) preparations to implementing and governing AI. The main use case discussed throughout this dissertation was PERISCOPE, an AI tool predicting the risk of postoperative infection within 7- and 30-days of surgery.

• Artificial Intelligence
• Machine Learning
• Prediction model
• Decision support
• Surgery
• Perioperative care
• Intensive care
• Postoperative infection
• Validation
• Fairness

• All authors: Meijden, S.L. van der
• Supervisor: Arbous, M.S.; Steyerberg, E.W.
• Co-supervisor: Geerts, B.F.
• Committee: Jonge, E. de; Peul, W.C.; Hernandez-Boussard, T.; Boermeester, M.A.
• Qualification: Doctor (dr.)
• Awarding Institution: Leiden University Medical Center (LUMC), Leiden University
• Date: 2025-05-06
• ISBN (print): 9789465220505
• Sponsorship: Healthplus.ai; ChipSoft