Artificial intelligence in perioperative pain: a scoping review protocol.

چکیده اصلی

INTRODUCTION: Perioperative pain is a major determinant of patients' experience and may influence long-term outcomes. Artificial intelligence (AI) is applied to perioperative datasets to predict acute postoperative pain, opioid requirements, analgesic-related adverse events, chronic postsurgical pain and pain trajectories. However, the evidence is dispersed across perioperative phases, clinical outcomes, data modalities and AI methodologies, while the extent to which current models address validation, interpretability, uncertainty and clinical implementation remains unclear. This scoping review protocol aims to characterise the existing evidence on AI applications related to perioperative pain to identify methodological features that affect its clinical credibility and implementation. METHODS AND ANALYSIS: This protocol will follow the Joanna Briggs Institute methodology for scoping reviews and will be reported in line with Preferred Reporting Items for Systematic Reviews and Meta-Analysis Extension for Scoping Reviews (PRISMA-ScR) guidance. Eligible studies will include human research evaluating AI methods in relation to pain-related perioperative outcomes, including observational studies, interventional studies and predictive modelling studies. AI approaches include machine learning, deep learning, natural language processing, computer vision, large language models and hybrid or ensemble methods. Searches will be conducted from PubMed/MEDLINE, Embase, the Cochrane Library, medRxiv, arXiv and ClinicalTrials.gov, with supplementary screening of reference lists. Results will be synthesised descriptively. The formal literature searches are planned for September 2026, with completion of study selection, data charting, evidence synthesis and preparation of the final review expected by December 2026. ETHICS AND DISSEMINATION: We will chart ethical aspects reported in included studies, such as governance of retrospective electronic health record use, consent waivers and privacy protections, where available. Findings will be disseminated through submission to a peer-reviewed journal, presentation at scientific meetings and open sharing of search strategies on the Open Science Framework. PROSPERO REGISTRATION NUMBER: This protocol was prospectively registered on the Open Science Framework in March 2026: Cascella M et al 'Artificial intelligence in perioperative pain: a scoping review protocol'. DOI: 10.17605/OSF.IO/SYZH8.

ترجمه فارسی چکیده

درد دوران عمل عامل اصلی تجربه بیمار است و ممکن است بر نتایج بلندمدت تأثیر بگذارد. هوش مصنوعی (AI) بر داده‌های دوران عمل برای پیش‌بینی درد حاد پس از عمل، نیاز به مسکن‌های اپیوئید، رویدادهای نامطلوب مرتبط با مسکن، درد مزمن پس از جراحی و مسیرهای دردی کاربرد دارد. با این حال، شواهد پراکنده در فازهای دوران عمل، نتایج بالینی، حالات داده و روش‌های AI هستند و میزان مدل‌های فعلی که شامل اعتبارسنجی، قابلیت تفسیر، عدم قطعیت و پیاده‌سازی بالینی هستند، روشن نیست. این پروتکل مرور محدود هدف دارد ویژگی‌های موجود شواهد در مورد کاربردهای AI مرتبط با درد دوران عمل را توصیف کند تا ویژگی‌های روش‌شناختی که بر اعتبار بالینی و پیاده‌سازی آن تأثیر می‌گذارند شناسایی شود.

متن کامل اصلی