The EAGLE Trial, a multicenter randomized controlled study published in npj Digital Medicine, has demonstrated that Olympus's cloud-based CADDIE™ AI application significantly aids endoscopists in detecting high-risk and hard-to-detect colorectal lesions during colonoscopy. The study, conducted across eight centers in four European countries, involved 841 patients and 22 endoscopists performing screening and surveillance procedures. Patients were randomized to receive either standard colonoscopy or CADDIE-assisted colonoscopy.
Key findings from the trial indicate a 7.3% absolute increase in adenoma detection rate (ADR) when using the CADDIE application compared to standard colonoscopy. More significantly, the study showed substantial relative increases in the detection of clinically relevant lesion subtypes: 93% for large adenomas (>10 mm), 57% for non-polypoid (flat) adenomas, and 230% for sessile serrated lesions (SSLs). These lesion types are particularly important because SSLs and flat lesions are difficult to detect visually and are associated with increased risk of post-colonoscopy colorectal cancer.
The CADDIE application is the first cloud-based Computer-Aided Detection (CADe) system for real-time polyp detection that has received both FDA clearance and CE marking. As part of the OLYSENSE™ Intelligent Endoscopy Ecosystem, the software is trained on a dataset specifically enriched with clinically relevant and hard-to-detect lesions. The cloud-based architecture offers hospitals flexibility by reducing hardware dependency and enabling subscription-based procurement models, which could democratize access to advanced AI tools in endoscopy.
"This study marks a pivotal shift in the clinical translation of AI-assisted endoscopy," said Rawen Kader, Principal Investigator of the EAGLE Trial and GI Researcher at University College London. "Cloud deployment can remove hardware barriers and give hospitals access to the latest AI innovations, which has the potential of improving detection of the lesions that matter most for reducing colorectal cancer risk." The complete study is available at https://doi.org/10.1038/s41746-025-02270-1.
The trial's results address concerns raised in recent clinical guidelines about AI-assisted colonoscopy by demonstrating increased detection of clinically relevant lesions without increasing unnecessary resections. The ability to reliably detect SSLs is increasingly viewed as a critical quality consideration in colonoscopy, as highlighted in recent quality indicator publications. The system demonstrated real-time performance and operational efficiency across diverse testing environments, suggesting feasible integration into clinical workflows.
"The EAGLE trial demonstrates how cloud‑based AI can be translated into routine endoscopy at scale," said Peter Mountney, CEO of Odin Vision and Vice President, AI Unit at Olympus Corporation. By delivering AI in real time through the cloud, this approach could help accelerate innovation and enable hospitals worldwide to benefit from evidence-based technologies that support clinicians and enhance care quality. The cloud architecture utilizes industry standard security controls while providing the foundation for future AI applications in endoscopy.
