Gastro-entero-pancreatic neuroendocrine neoplasms (GEP-NENs) represent a challenging disease. Their large heterogeneity limits the possibility of providing accurate risk assessments or standardizing the most effective therapies for these patients. In recent years, artificial intelligence (AI), and in particular machine learning approaches, have shown real promise in addressing these complexities. By analyzing large volumes of clinical, imaging, and pathological data, AI-based tools can significantly improve the accuracy of survival predictions and guide more tailored treatment strategies. In this narrative review, we examine the potential applications of AI to develop effective prognostic models in GEP-NENs, and how these models may help clinicians in predicting survival and optimizing patient management. While early results are encouraging, important limitations remain, since available data stem from small, retrospective datasets, sometimes lacking external validation, and concerns around transparency and ethics still represent an open issue. Addressing these gaps will be key to moving from research applications to practical tools that can support everyday clinical decision-making.
胃肠胰神经内分泌肿瘤（GEP-NENs）是一种具有挑战性的疾病。其高度异质性限制了为患者提供准确风险评估或标准化有效治疗方案的可能性。近年来，人工智能（AI），特别是机器学习方法，在应对这些复杂性方面展现出巨大潜力。通过分析大量临床、影像和病理数据，基于AI的工具能够显著提高生存预测的准确性，并指导制定更个体化的治疗策略。本文通过叙述性综述，探讨AI在构建GEP-NENs有效预后模型中的潜在应用，以及这些模型如何帮助临床医生预测生存期和优化患者管理。尽管早期成果令人鼓舞，但仍存在重要局限：现有数据多来源于小规模回顾性数据集，部分缺乏外部验证，且透明度和伦理问题仍是待解决的议题。弥补这些不足将是推动AI从研究应用转化为支持日常临床决策实用工具的关键。
Artificial Intelligence for Prognosis of Gastro-Entero-Pancreatic Neuroendocrine Neoplasms
肿瘤(癌症)患者之家