Post written by Joel Troya, MSc, from Interventional and Experimental Endoscopy (InExEn), Internal Medicine II, University Hospital Würzburg, Würzburg, Germany.

The combination of artificial intelligence (AI) and side optics might help detect adenomas that otherwise could be missed.

Several studies have demonstrated that the adenoma detection rate is the critical parameter for reducing the incidence of colorectal cancer. However, there is evidence that the adenoma miss rate is 26%. The objective of this novel prototype was to reduce this parameter using the combination of side optic microcameras and AI.

In this feasibility study, we were able to automatically detect polyps in adenomatous polyposis coli gene—mutated pigs using side optic microcameras, which were installed in an endoscope using an add-on cap printed with 3D technology. The developed AI system was able to detect, mark, and alert the endoscopist of adenomas outside the conventional field of view.

The prototypic concepts in this study have proven valuable with potential perspectives to be integrated into clinical practice once it has been approved as a medical device. Future studies must be performed to extensively evaluate the performance of the device and testing if it can lead to the detection of additional polyps.

Overall, we believe our new concept for colonoscopy combining AI with side optics might help detect adenomas that otherwise could be missed without significantly disturbing the conventional colonoscopic workflow.

Image of a correct detection. Left, Left-sided camera showing normal mucosa. Center, Standard endoscopic view image. The arrowhead alerts the endoscopist that the right camera has detected an adenoma. Right, Right-sided microcamera highlighting the region where the adenoma appears.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.