Abstract
Effective postoperative pain relief is crucial for the recovery of pediatric patients. While artificial intelligence (AI) is increasingly being applied in pain assessment, there is a notable lack of data regarding its role in managing postoperative pain in children. This systematic review aims to address this gap by focusing on AI's use in predicting and evaluating pediatric postoperative pain. We conducted a comprehensive search of relevant studies from January 2000 to November 2023, identifying 4,491 studies, which were narrowed down to eight based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. These selected studies included 4,470 pediatric patients assessed using various pain measurement tools. The AI models used, primarily deep learning and machine learning, demonstrated accuracy rates ranging from 79% to 85.62% and area under the receiver operating characteristic curve values between 84.00% and 94.00%. Although these AI-based pain assessment tools are still in the early stages, they often focus on single parameters. The heterogeneity of the available publications prevented the conduct of a meta-analysis. Our findings underscore the need for multimodal, multicentric research to improve the performance of AI-based tools for assessing postoperative pain in the pediatric population. Such advancements could significantly enhance the future of pediatric pain management.