Lung Vision

The LungVision System is designed to support catheter and device navigation within the lungs during pulmonary procedures by segmenting pre-operative 3D CT scans and overlaying this anatomical data onto live fluoroscopic images. Its methodology leverages PC-based image processing for DICOM CT segmentation and co-registration with real-time fluoroscopy, offering visual guidance for endobronchial tool placement.

The LungVision System is an image processing tool developed to assist physicians during pulmonary procedures by improving the navigation of endobronchial tools. Its core function is to enable users to segment previously acquired 3D CT datasets, plan a target location and an access path within the lungs, and then overlay these segmented 3D datasets onto live fluoroscopic X-ray images of the patient's anatomy. This co-registration and overlay provide real-time visual support for guiding catheters or other medical devices toward the intended area of interest during the procedure. The system utilizes a PC-based hardware platform running proprietary image processing algorithms and software. Its main components include a laptop, a video capture box, an IEC 60601-1 compliant isolation transformer, a passive LungVision Board, and the LungVision software. Key technological capabilities encompass the segmentation of DICOM 3D CT image data, registration of this CT data with live fluoroscopic X-ray images, and the subsequent overlay of the segmented 3D CT dataset onto the live fluoroscopic view. Non-clinical bench testing confirmed that the system met all requirement specifications, including hazard mitigation and overall system performance, with specific tests using pig lungs to evaluate accuracy in deformable tissue. The system hardware complies with IEC 60601-1:2005 and IEC 60601-1-2:2014 standards.

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 system's accuracy in deformable tissue, as tested with pig lungs, translate to the dynamic environment of human breathing during a live procedure?; What is the typical time required for pre-procedure planning, including CT segmentation and path definition, using the LungVision System?; How does the LungVision System account for and mitigate potential registration inaccuracies between the static pre-acquired CT and the dynamic live fluoroscopy caused by patient movement or respiratory motion?; What specific image processing algorithms underpin the segmentation of 3D CT datasets and their registration with live fluoroscopic images, particularly concerning complex bronchial anatomies?; Are there comparative data on procedural efficiency, diagnostic yield, or patient outcomes when using the LungVision System versus unguided procedures or other navigation technologies for similar pulmonary interventions?