World's First Artificial Intelligence-Based Evaluation of Rist Catheter Stability in Transradial Procedures: A Feasibility Study

Abstract

Objective: Artificial intelligence (AI) holds promise for advancing neuroendovascular therapy through device evaluation, but its application in real-world clinical settings remains limited. We aimed to validate the feasibility of AI-based quantitative device evaluation during actual procedures by assessing the stability of the Rist radial access guide catheter (Medtronic, Dublin, Ireland), a novel device designed for the increasingly adopted transradial approach (TRA), during flow diverter stent (FDS) placement.

Methods: We retrospectively analyzed 4 cases of FDS placement using Rist via the TRA. Rist was tracked in recorded fluoroscopic videos using the AI technology of Neuro-Vascular Assist (iMed Technologies, Tokyo, Japan). The movement distance of Rist during FDS placement was calculated as a stability indicator.

Results: All procedures were successfully completed without any complications. Rist was introduced from the radial artery and positioned in the distal internal carotid artery. The maximum movement distances of the Rist during the procedures were 3.36, 6.63, 1.79, and 0.33 mm for each case, respectively, with an average of 3.03 mm. The maximum movement distances per minute were 1.68, 2.34, 1.19, and 0.46 mm/min, respectively, with a mean of 1.42 mm/min. These measurements suggest sufficient stability for the FDS procedures.

Conclusion: This study demonstrates the feasibility of using AI technology to quantitatively analyze Rist stability in TRA procedures. To the best of our knowledge, this is the 1st clinical evaluation of device function in a clinical setting using AI technology. Further studies with more cases are required to validate these findings. This method is promising for real-world device evaluation and development.

Keywords: Rist catheter; artificial intelligence; guide catheter stability; neuroendovascular treatment; radial access.

Fig. 1. Example of device recognition by artificial intelligence (AI) (Case 2). (A) Original subtraction image. The wire tip of the flow-diverter stent (arrowhead), tip of the Navien catheter (black arrow), and tip of the Rist catheter (white arrow) can be identified. (B) AI recognition. Navien and Rist are recognized as catheter tips (squares). The wire tip is displayed as a triangle.

Fig. 2. Trajectories and temporal changes in the positions of Navien and Rist tips during flow-diverter stent deployment. (A–D) The upper row shows the trajectories of Navien (orange) and Rist (blue) for each case. (E–H) The lower row represents the temporal changes in Navien and Rist positions from their initial positions. The initial position is set as the origin, with negative values indicating proximal movement. The movement of Rist is relatively small compared with that of Navien, demonstrating the stability of Rist.

Fig. 3. Movement distances of Navien and Rist during flow-diverter stent deployment. (A) Maximum movement distances of Navien and Rist during flow-diverter stent deployment for each case. (B) Maximum movement distances per minute. Rist shows an average movement of 1.74 mm per minute, suggesting stability.