Room: Track 1
Purpose: To measure the accuracy of co-registering a transthoracic echocardiography (TTE) probe to C-arm fluoroscopy coordinate space by tracking x-ray visible fiducials attached to the TTE probe. Accurate 3D x-ray/TTE fusion displays can guide catheter device deployment in structural heart interventions such as Transcatheter Aortic Valve Replacement (TAVR).
Methods: A TTE probe attachment was fabricated using a thermoplastic material with 3 and 5-mm diameter steel fiducials arranged in two 4-cm diameter rings. A pose estimation algorithm deduces 3D probe pose from a 2D x-ray image of the fiducials by iteratively rotating/translating a 3D virtual fiducial model until its forward-projection matches the image; a match is identified by minimizing a cost function computing the average sum-of-square of image grayscale values. Pose estimation performance was tested on fluoroscopic images of an anthropomorphic chest phantom designed to mimic clinical TAVR cases. A TTE probe, with fiducials, was placed on the phantom in four standard acoustic windows. 2D x-rays were then obtained from several TAVR-relevant gantry orientations. C-arm cone-beam CT established 3D reference fiducial locations, enabling pose estimation performance quantification with 3D fiducial registration error (FRE). Upon verifying pose estimation accuracy, a second phantom experiment was conducted to demonstrate 3D x-ray/TTE fusion.
Results: Pose estimation was performed on x-ray images acquired at 20°RAO, 0°PA, 20°LAO, and 20°LAO/15°CRA for each probe acoustic window, with 100 trials per set-up. The 90th-percentile 3D FREs ranged between 0.37–1.34 mm for different set-ups. Overall, 98% of pose estimates achieved 3D FRE = 2.00 mm. 3D co-registration of an x-ray-based TAVR device image and echocardiography was demonstrated in a dual-modality phantom.
Conclusion: TTE probe tracking with x-ray-visible fiducials enables 3D co-registration of ultrasound images of patient anatomy and x-ray images of interventional devices. This technology supports a minimally-invasive TTE-based approach to x-ray/echo image fusion in structural heart interventions.
Funding Support, Disclosures, and Conflict of Interest: Research reported in this work was partly supported by funding received from Siemens Healthineers.