An MRI-Guided Robotic System to Improve Hippocampal Access for Epilepsy Interventions
Authors: John E. Peters, Abby M. Grillo, Mahshid Mansouri, Daniel S. Esser, Sarah Garrow, Nithin S. Kumar, Dario J. Englot, Joseph S. Neimat, William A. Grissom, Robert J. Webster, Eric J. Barth
Organizations: Department of Mechanical Engineering, Vanderbilt University, Nashville, TN, USA. · Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA. · Department of Neurological Surgery, Vanderbilt University Medical Center, Nashville, TN, USA. · Department of Neurological Surgery, University of Louisville Health, Louisville, KY, USA.
This paper presents an MRI-guided robotic system that improves hippocampal access by delivering a curved needle-like laser ablator through the foramen ovale, a natural opening in the base of the skull. Both the delivery path and the curved nature of the needle improve upon current clinical straight-line laser interstitial thermal therapy (LITT) as measured by hippocampal cannulation percentage. We describe the design of the robotic system which includes positioning, aiming, and curved needle deployment stages, followed by experimental results assessing cannulation percentages using MRI images in phantoms. In three curvilinear, single-insertion experiments, we achieved cannulation percentages of 93.5%, 96.5%, and 60.4%, exceeding reported clinical averages of 50-60%. Seizure control is believed by physicians to be a function of hippocampal volume treated, and our system provides a means of treating a greater volume of the hippocampus with LITT-based interventions.
Figures & tables
Figure 1: Transforaminal ablation approach in a patient skull. (A) Concentric docking tube, precurved tube, and laser fiber traversed through the foramen ovale and hippocampus. (B) Percutaneous approach of the docking tube to the foramen ovale.
Figure 2: Complete CAD rendering of the entire robotic system depicted for use in the scanner bore with a patient positioned for a percutaneous, transforaminal approach.
Figure 3: The actuation unit of the concentric tube robot from different perspectives. (A) Side view (docking tube hidden for simplicity) detailing the tubes, encoders, and fiber actuator. (B) Isometric view illustrating the concentric tubes, each capable of translation and rotation, as well as the precurved tube actuator. Docking tube is rigidly fixed to the CTAU housing. Inset shows the degrees of freedom (T: translation, R: rotation) associated with the fiber (red) and precurved tube (blue).
Figure 4: Detailed view of the robot’s 3D-printed, pneumatic, stepper actuators. (A) Isometric view of the precurved tube actuator (cyan) highlighting the translational DOF (pink arrow) including a linear gripper (grey) and its deflection (yellow arrows). (B) Rear view of the precurved tube actuator detailing the rotational DOF (red arrows) including a rotary gripper (grey) and its deflection (yellow arrows), a pair of rotary bellows and their deflection (green arrows), and various air inlets. (C) Rear view of the RPA actuator (cyan) detailing the rotational DOF (red arrows) including a rotary gripper (grey) and its deflection (yellow arrows), a pair of rotary bellows and their deflection (green arrows), and various air inlets. Each RPA actuator includes a second gripper (not shown) positioned on the opposing end of a common driveshaft. All actuators operate in a stepwise fashion executing either a translational or rotational step using alternating gripper states. Additional detail on the stepwise operation can be found in Peters et al. (2023) .
Figure 5: (A) Robotic Parallelogram Arm (RPA) used to adjust the angle of the initial entry vector of the docking tube. The body of each actuator (shown in Fig. 4 C) is affixed perpendicular to its axis of rotation in order to generate rotation in both the pitch and yaw directions. (B) A 6-DOF manual positioning platform aligns the RCM of the RPA to the foramen ovale.
Figure 6: Experimental setup with the assembled robotic system. (A) Robotic hardware for optical tracker experiments. (B) Robotic system inside the MRI scanner.
Figure 7: Free-space targeting results for four desired trajectories. The final tip errors were 1.70 mm, 1.75 mm, 1.18 mm, and 2.47 mm for Trials 1-4, respectively.
Figure 8: MRI‑guided phantom evaluation of the curved, transforaminal approach (n=3). For each trial, four spatial performance metrics are plotted versus arc length along the hippocampal medial axis (s = 0 at the head, increasing toward the tail; total length ≈ 40 mm). Corresponding 3D renderings depict the segmented hippocampus (semi‑transparent blue), its medial axis (dark blue), the planned trajectory (green), and the actual trajectory (magenta). Pale green shading marks the percent of the hippocampus cannulated by the actual path.
Performance Metric
Trial
Percent of Hippocampus Cannulated (%)
Plan Error (mm)
Tracking Error (mm)
Medial Axis Placement Error (mm)
Mean
Max
Mean
Max
Mean
Max
1
93.5
1.91
4.19
2.90
4.85
2.28
4.81
2
96.5
1.67
5.04
2.64
5.43
1.95
5.16
3
60.4
1.83
4.95
2.79
5.73
2.31
4.01
Mean
83.5
1.80
4.73
2.78
5.34
2.18
4.66
Table 1: Performance Metrics along the Hippocampal Arc Length for MRI-Guided Targeting Experiments
Department of Surgical & Interventional Engineering, School of Biomedical Engineering & Imaging Sciences, Faculty of Life Sciences & Medicine, King’s College London, London WC2R 2LS, United Kingdom · Department of Computer Science & Biomedical Informatics, University of Thessaly, Lamia 35131, Greece · School of Engineering and Materials Science, Queen Mary University London, London E1 4NS, United Kingdom +2