Artificial intelligence in the diagnosis of gastro-entero-pancreatic neuroendocrine neoplasms: Potential benefits and current limitations.

Abstract

Neuroendocrine neoplasms (NENs), once considered rare, are now increasingly diagnosed worldwide, with gastro-entero-pancreatic neuroendocrine tumors (GEP-NETs) accounting for the majority of cases (55%-70%). NENs are characterized by considerable heterogeneity, driven by factors such as tumor differentiation, Ki-67 index, primary tumor location, somatostatin receptor status, and disease stage. International guidelines advocate for a multidisciplinary approach to ensure individualized treatment strategies. Given the disease's complexity, artificial intelligence (AI) may offer substantial support in the management of NENs. AI is playing an increasingly prominent role in medicine by enabling advanced diagnostic capabilities through machine learning and deep learning algorithms, particularly in imaging. However, current literature on AI applications in NENs is limited, and their routine use in clinical practice has yet to be established. This narrative review aims to provide a comprehensive overview of the potential roles of AI in the diagnosis of GEP-NENs, while also addressing the associated biases and ethical considerations of medical AI implementation.