EndoCartoScope
Spatial AI for Medical Endoscopy
We are building a spatial AI perception layer for clinical endoscopes — software that enables real-time 3D reconstruction, automated coverage assessment, AR-guided navigation, and quantitative lesion measurement. An early-stage R&D programme at the University of Zaragoza, funded by the EU EIC Transition programme.
EIC Transition Programme
Grant No. 101211633
Technology
We are developing a software-only spatial intelligence layer that makes standard flexible endoscopes spatially aware — transforming a passive camera into an active navigational instrument, without hardware modifications.
Real-time 3D Mapping
Dense 3D reconstruction of the explored cavity from the camera feed alone, while simultaneously localising the endoscope within the map. A live SLAM core without hardware modifications.
Coverage Assessment
Objective measurement of how much mucosa has been inspected. Total coverage scores the whole procedure; local coverage flags unexamined regions live, guiding the endoscopist back before withdrawal.
Quantitative Measurement
Metric size estimation of polyps and lesions from monocular video, with no additional hardware.
CLiMB Challenge
Colonoscopy Localization and Mapping Benchmark
We are organizing CLiMB as part of the Endoscopic Vision Challenge (EndoVis) at MICCAI 2026. An open benchmark for 3D mapping and spatial AI methods in colonoscopy.
Team
A blend of young entrepreneurs and world-class experts in Spatial AI, Computer Vision, and clinical endoscopy.
Carlos Sostres
Medical Advisor
Advised by the founders of Odin Vision — a leading AI endoscopy company acquired in 2023.
Consortium Partners
Get in touch
Campus Río Ebro, Universidad de Zaragoza, Spain
This project has received funding from the European Union's Horizon Europe research and innovation programme under the EIC Transition grant agreement No. 101211633. Views expressed are those of the project team and do not necessarily reflect the EU's position.
View project on CORDIS ↗