First Steps into Catheter Guidance Including Shape Sensing for Endovascular Aneurysm Repair Procedures

Abstract

Introduction: Currently, endovascular aortic repair (EVAR) procedures are guided using fluoroscopy and 2D digital subtraction angiography. The research project Nav EVAR (funded by the German Federal Ministry of Education and Research, grant number 13GW0228C) aims to reduce the x-ray exposure and the administration of contrast agent. This project includes the development of a guidance system based on electromagnetic (EM) tracking for catheter location and Fiber Bragg Gratings (FBG) for catheter shape sensing. The goals of this study were to evaluate the accuracies of the catheter location and the catheter shape sensing separately, and to carry out an initial experiment combining both technologies. Methods: A custom-made 9F catheter was build containing an Aurora EM sensor (Northern Digital Inc.) at its tip and a FBG system (DTG©, FBGS Technologies GmbH). The EM sensor position was obtained by the EM tracking system and then transformed into the computed tomography (CT) planning data by using metallic markers placed on the patient's torso. An optimized shape sensing model was used to reconstruct the first 38 cm of the catheter. The combination of both technologies allowed the localization of the reconstructed shape. The EM tracking system was evaluated with a human torso model including a 3D-printed patient-specific aortic system. The catheter was inserted into the iliac artery and the EM sensor accuracy was assessed at five positions (Figure 1) by obtaining the ground truth from CT scans. The FBG system was evaluated by inserting the catheter into a 3D-printed patient-specific aortic system. The accuracy was assessed by comparing the reconstructed shape with the ground truth from a CT scan. The feasibility of the combined system was tested by visualizing the located shape with MeVisLab software (MeVis Medical Solutions AG). Results: For the EM tracking system, the error was 1.60 ± 0.86 mm (mean ± standard deviation, maximum error 2.56 mm). For the FBG system, the error was 1.13 ± 0.43 mm (maximum error 2.11 mm) over the whole 38 cm. Figure 2 shows the CT scan with the reconstructed shape. Regarding the combined guidance, the located shape moved in the same way as the catheter in the real scenario. Conclusion: Both technologies provided an error of less than 5 mm in realistic endovascular scenarios. This accuracy is required to ensure that the renal arteries are not blocked by the stent placement. The EM sensor accuracy was comparable to other studies [1-2]. Previous studies of FBG systems focused on medical needle shape sensing [3-4]. The shape reconstruction error was higher compared to previous studies, but our endovascular was more complex and realistic [5]. The combined system showed promising results. However, these are initial experiments that do not take into account any anatomical deformations during EVAR procedures. Future work will include the optimization of our catheter guidance system, intraoperative imaging for updating the actual anatomy and further evaluations with patient-specific models. Disclosure: Nothing to disclose