Artificial intelligence application in diagnostic gastrointestinal endoscopy - Deus ex machina?

Area of Science:

• Gastroenterology
• Medical Technology
• Artificial Intelligence

Background:

• Medicine's integration with technology, particularly artificial intelligence (AI), is rapidly advancing.
• AI systems are increasingly developed for medical applications, demonstrating potential for accurate diagnoses.

Purpose of the Study:

• To review the current applications and state of the art of artificial intelligence in gastroenterological pathology.
• To summarize AI's role in diagnosing digestive tract pathologies, including colorectal polyps, Barrett's esophagus, and ulcerative colitis.

Main Methods:

• This review synthesizes findings from various studies on AI applications in gastroenterology.
• The focus is on AI's diagnostic accuracy, sensitivity, and specificity compared to human endoscopists.

Main Results:

• AI systems achieve high accuracy, sensitivity, and specificity in diagnosing gastroenterological conditions, often matching or exceeding human performance.
• AI shows promise in detecting colorectal polyps, assessing dysplasia in Barrett's esophagus, and evaluating colon inflammation in ulcerative colitis.

Conclusions:

• Artificial intelligence is a valuable tool in gastroenterological pathology, enhancing diagnostic capabilities and clinical decision-making.
• While AI demonstrates significant potential, further studies are required to validate its applicability across broader indications.