Fuchs Juri, Rabaux-Eygasier Lucas, Guerin Florent
Department of General, Visceral and Transplantation Surgery, Heidelberg University Hospital, 69120 Heidelberg, Germany.
Department of Pediatric Surgery, Université Paris-Saclay, Assistance Publique Hôpitaux de Paris (AP-HP), Bicêtre Hospital, 94270 Le Kremlin Bicêtre, France.
Children (Basel). 2024 Aug 15;11(8):996. doi: 10.3390/children11080996.
Historically, pediatric liver transplantation has achieved significant milestones, yet recent innovations have predominantly occurred in adult liver transplantation due to higher caseloads and ethical barriers in pediatric studies. Emerging methods subsumed under the term artificial intelligence offer the potential to revolutionize data analysis in pediatric liver transplantation by handling complex datasets without the need for interventional studies, making them particularly suitable for pediatric research. This review provides an overview of artificial intelligence applications in pediatric liver transplantation. Despite some promising early results, artificial intelligence is still in its infancy in the field of pediatric liver transplantation, and its clinical implementation faces several challenges. These include the need for high-quality, large-scale data and ensuring the interpretability and transparency of machine and deep learning models. Ethical considerations, such as data privacy and the potential for bias, must also be addressed. Future directions for artificial intelligence in pediatric liver transplantation include improving donor-recipient matching, managing long-term complications, and integrating diverse data sources to enhance predictive accuracy. Moreover, multicenter collaborations and prospective studies are essential for validating artificial intelligence models and ensuring their generalizability. If successfully integrated, artificial intelligence could lead to substantial improvements in patient outcomes, bringing pediatric liver transplantation again to the forefront of innovation in the transplantation community.
从历史上看，小儿肝移植已取得了重大里程碑，但由于小儿研究中病例数量较多以及存在伦理障碍，近期的创新主要发生在成人肝移植领域。归入人工智能这一术语的新兴方法有潜力通过处理复杂数据集而无需进行干预性研究来彻底改变小儿肝移植中的数据分析，使其特别适用于小儿研究。本综述概述了人工智能在小儿肝移植中的应用。尽管早期有一些令人鼓舞的结果，但人工智能在小儿肝移植领域仍处于起步阶段，其临床应用面临若干挑战。这些挑战包括需要高质量、大规模的数据，以及确保机器学习和深度学习模型的可解释性和透明度。还必须解决伦理考量，如数据隐私和偏差可能性等问题。小儿肝移植中人工智能的未来方向包括改善供体 - 受体匹配、管理长期并发症以及整合多种数据源以提高预测准确性。此外，多中心合作和前瞻性研究对于验证人工智能模型并确保其通用性至关重要。如果成功整合，人工智能可能会大幅改善患者预后，使小儿肝移植再次成为移植界创新的前沿领域。
