Steerable Robots for MRI-Guided Intracranial Delivery of Therapeutics 

Motivation: Glioblastoma multiforme (GBM) is one of the most aggressive and deadly brain tumors, with limited treatment options and poor prognosis, particularly in recurrent cases. Traditional treatment methods, including surgery, chemotherapy, and radiation, offer only marginal survival benefits, and most patients face high rates of tumor recurrence. Immunotherapy has emerged as a promising alternative, but its effectiveness is hampered by poor cell trafficking and limited delivery efficiency, especially in large tumors. To overcome these challenges, we have developed an MRI-conditional, steerable neurosurgical robot designed to deliver therapeutic agents directly to multiple locations within the tumor. This innovative robotic system aims to improve the accuracy and efficacy of therapeutic delivery while minimizing damage to healthy brain tissue, enhancing the potential of immunotherapy in treating GBM.

Figure 1. Therapeutic delivery at multiple locations inside a brain tumor

Project Highlights:

In this project, we have designed and prototyped a highly dexterous needle robot capable of multi-site intracranial delivery of therapeutic agents. The system integrates a novel MR-conditional robotic actuation mechanism with an injection system to achieve precise navigation and targeted therapeutic delivery under MRI guidance.

Figure 2. Design of the robotic system: (a) Overall view of the robot and its features, (b) Mounting of the nonmagnetic actuators and sliders, (c) Connection of the syringes in injection system, and (d) Tubes arrangement and tendon wire routing.

The steerable needle robot includes a rigid outer tube and a flexible, tendon-driven inner tube made of superelastic nitinol. This design allows for dexterous, curved motion and precise steering, enabling the robot to reach multiple locations within the tumor. Performance tests demonstrated a tip positioning error of less than 1 mm, validating the accuracy and safety of the robot’s navigation within soft brain tissue. 

The robot’s injection system is designed to deliver therapeutic agents through a long hydraulic transmission tube connected to motorized syringes. This setup enables the controlled release of immunotherapy agents, ensuring accurate delivery to multiple sites within the tumor under MRI guidance.
 

Figure 3. (i) Snapshots of demonstration of injecting phantom therapeutic agent (blue) into the phantom tumor (red area) at nine different locations using the developed robotic system. The injection locations are illustrated by yellow crosses. (ii) Experimental setup consisting of the needle robot, actuation system, and injection system. (iii) Results of in vitro demonstration: (a) segmentation of the result by straight and steerable needle

This approach enhances the potential for immunotherapy by maximizing agent distribution, increasing therapeutic efficacy, and minimizing damage to healthy tissue. Future work will focus on in vivo trials and further refinement of the robot’s control system to improve performance in clinical settings.

Figure 4. (i) Signal and noise regions used for the analysis of MR-conditionality of the robotic system in scenarios including: (a) water phantom with the robot powered on but not extended inside and (b-d) water phantom with the robot inside extended to three different lengths in FTL motion but powered off during scans. The green and red squares represent the choose noise and signal regions, respectively. (ii) 3D renderings of DICOM series at various FTL lengths in the MRI bore with customized alphamaps. (iii) Experimental setup with a 15◦ angle between the outer tube and phantom surface to facilitate inserting the robot into the phantom container.

 Refereed Conference Publications:

  • S. Rezaeian, H. Chen, C. Sidhu, B. Bartnik-Olson, B. Badie, and J. Sheng, “Neurosurgical robot for mri-guided intracranial therapeutic delivery,” Authorea Preprints, 2024.