Artificial intelligence in gastroenterology: Ethical and diagnostic challenges in clinical practice
Davide Ramoni, Alessandro Scuricini, Federico Carbone, Luca Liberale, Fabrizio Montecucco
Abstract
, which explores the application of deep learning models in decision-making processes via wireless capsule endoscopy. Integrating artificial intelligence (AI) into gastrointestinal disease diagnosis represents a transformative step toward precision medicine, enhancing real-time accuracy in detecting multi-category lesions at earlier stages, including small bowel lesions and precancerous polyps, ultimately improving patient outcomes. However, the use of AI in clinical settings raises ethical considerations that extend beyond technological potential. Issues of patient privacy, data security, and potential diagnostic biases require careful attention. AI models must prioritize diverse and representative datasets to mitigate inequities and ensure diagnostic accuracy across populations. Furthermore, balancing AI with clinical expertise is crucial, positioning AI as a supportive tool rather than a replacement for physician judgment. Addressing these ethical challenges will support the responsible deployment of AI, through equitable contribution to patient-centered care.