EndoNaut

EndoNaut is an advanced image fusion and computerized navigational system designed to enhance precision in fluoroscopy-guided endovascular procedures by overlaying preoperative 3D anatomical models onto real-time 2D X-ray images. Utilizing technologies such as 3D/2D registration and machine learning for motion and contrast detection, it assists physicians in accurately positioning surgical instruments and devices, with recent developments introducing a server-based architecture for potentially improved performance and integration.

EndoNaut, developed by Therenva SAS, is an image fusion software solution and computerized navigational system designed to assist physicians during X-ray fluoroscopy-guided procedures. It aids in the precise positioning of surgical instruments and endovascular devices by overlaying preoperative 3D anatomical data onto live fluoroscopic images. The system is indicated for use in procedures involving the chest, abdomen, pelvis, neck, and lower limbs, such as aneurysm repair, artery/vein embolization, and peripheral artery disease treatment. EndoNaut is interoperable with the EndoSize software, which facilitates preoperative case planning, device selection, and visualization of vascular structures from CT scans. The system comprises software components, including standalone and server variants, and hardware accessories like the EndoNaut Workstation. The core methodology of EndoNaut involves advanced image processing techniques, including 3D/2D and 2D/2D registration, panorama creation from fluoroscopic sequences, and automatic motion detection, with some features utilizing machine learning algorithms. It processes DICOM images and C-Arm video streams to provide enhanced visualization and measurements, though it does not produce X-rays itself. This particular iteration introduces new system configurations and a server-based software architecture (EndoNaut Server SW) alongside the existing monolithic standalone version. Despite the architectural change to a server/client model, verification and validation activities confirmed that the server variant maintains equivalent performance to its predicate, offers robust safety and security mechanisms, and in some instances, demonstrates improved execution speed, such as in recalibration processes, without introducing new unacceptable risks.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the performance of the machine learning algorithms for 3D/2D registration and motion detection in EndoNaut Server (client/server architecture) quantitatively compare to the standalone version, particularly in complex anatomical regions or scenarios with significant patient movement?; What specific benefits or trade-offs (e.g., latency, scalability, integration flexibility) does the new server/client architecture of EndoNaut offer compared to the monolithic standalone system in a real-world clinical workflow involving multiple users or integrated operating room systems?; Beyond aneurysm repair and PAD treatment, what is the potential utility and performance of EndoNaut in other complex fluoroscopy-guided endovascular interventions, such as fenestrated/branched stent-grafting or structural heart procedures, where precise navigation is critical?; What are the typical recalibration times observed with the EndoNaut Server version compared to the standalone version, and how does this impact intraoperative workflow efficiency?; Given the reliance on preoperative CT data, how does EndoNaut's image fusion accuracy and navigational guidance adapt to or account for intraoperative anatomical changes (e.g., vessel deformation due to instrument manipulation, physiological changes) not present in the initial scan?