The significant advancements in artificial intelligence (AI) in recent years have established it as a crucial innovation in medicine and gastroenterology, with the potential to transform medical practice¹. AI’s remarkable capability to analyze complex algorithms and develop self-learning technologies-incorporating new concepts similarly to human learning-can enhance diagnostic accuracy, streamline management flowcharts, stratify risks, and optimize treatments, thereby facilitating personalized and efficient care¹. One of the most promising applications of AI in gastroenterology and digestive endoscopy is its immense ability to analyze endoscopic images with speed and precision.

Initial applications included computer-aided diagnosis (CAD) for the detection, differentiation, and characterization of neoplastic and non-neoplastic polyps². A randomized controlled trial involving 1058 patients demonstrated a significant increase in adenoma detection rates with the use of CAD compared to standard colonoscopy (29% vs. 20%, p < 0.001), with higher detection rates of diminutive adenomas (185 vs. 102, p < 0.001) and hyperplastic polyps (114 vs. 52, p < 0.001). There was no statistical difference in the detection of larger adenomas². Local experiences, such as those of Dr. Gómez and colleagues published in our journal³, contribute to understanding how AI techniques can improve the accuracy of endoscopic image interpretation, which often suffers from significant interobserver variability, as also noted in a previous issue⁴.

In this one, we publish the article by Dr. Martín Gómez and colleagues⁵, winner of the ACADI 2023 National Research Award in Gastroenterology. This article exemplifies how AI can contribute to the quality audit of upper digestive tract endoscopy through an AI-based endoscopic exploration protocol of the stomach. This protocol allowed the identification of gastric areas, achieving a sensitivity of 85.5% and a specificity of 98.8% in interpreting the 13 specific areas described. These results demonstrate AI’s potential effectiveness as a valuable tool for assisting endoscopists in performing a thorough and meticulous examination of the entire stomach.

The development of AI systems that can assist endoscopists in real-time during procedures can provide instant analyses of endoscopic images, highlight areas of interest, and suggest differential diagnoses. By integrating AI into clinical practice, endoscopists can improve their efficiency and accuracy, leading to better patient outcomes^6,7.

High-quality datasets and robust algorithms are indispensable for this integration. This necessitates close collaboration among physicians, researchers, and IT experts. However, AI algorithms and their applications require further studies and validation. Additionally, more clinical data are needed to demonstrate their efficacy, value, and impact on patient care and outcomes. Finally, cost-effective AI models and products must be developed to enable physicians, clinics, and hospitals to incorporate AI into daily clinical use.

We must view the use of AI as a tool to further improve clinical outcomes for patients with gastrointestinal diseases while also considering the risks of its potential misuse⁷.