Soft robot for MRI-guided transoral laser microsurgery of head and neck cancers

Researchers from the Chinese University of Hong Kong (CUHK) and the University of Hong Kong (HKU) have jointly developed an MRI-guided soft robotic manipulator for transoral laser microsurgery for head and neck cancer, which is designed to ensure adequate ablation margins as well as protect critical neurovascular structures.

“Laser-based tumour ablation would benefit greatly from MRI guidance, because 3D resection margins and thermal distributions could be evaluated in real time to protect critical structures while ensuring adequate resection margins. However, few studies to date have investigated the use of projection-based lasers such as those for transoral laser microsurgery, likely due to the dexterity of laser steering required at the ablation site, which raises substantial challenges in the confined MRI bore and its strong magnetic field,” wrote the researchers. [Sci Robot 2021;doi:10.1126/scirobotics.abg5575]

Through close collaboration between clinicians and engineers, the team developed a magnetic resonance (MR)–safe soft robotic system with some unique features. The soft robot’s miniature size (diameter of 12 mm and length of 100 mm) and five degrees of freedom enable dextrous operation within the confines of oral and pharyngeal cavities, without the need for extreme positioning including retraction of the tongue, opening of the mouth and/or extension of the neck of patients to achieve adequate tumour exposure for laser delivery.

Hydraulic actuation ensures MR compatibility while the whole system is fabricated with MR-safe materials. “The laser manipulator is rapidly fabricated with hybrid soft and hard structures and powered by microvolume [<0.004 mL] fluid flow to enable laser steering with enhanced stiffness and lowered hysteresis,” noted the researchers.

The system generates zero electromagnetic interference, allowing the introduction of intraoperative MRI guidance for evaluating laser–tissue interaction, which is essential to balancing adequate tumour resection with functional preservation. During laser ablation, MR thermal imaging can also be conducted to evaluate the heat diffusion in tissue, thus offering real-time monitoring of the ablation margin.

“A learning-based controller is used to accommodate the inherent nonlinear robot actuation and is validated with laser path–following tests, which demonstrated a steering accuracy with a mean tracking error <200 μm,” added the researchers.

Experimental validation of intraoperative MRI-guided robotic laser ablation on ex vivo tissue and cadaveric oropharyngeal tissue showed that the robot achieved precision in an automated path following the laser spot along a trajectory defined by the surgeon.

“The system enabled direct exposure of the target lesion to the ablation laser, without the need for positioning patients with extreme neck extension to accommodate rigid laser micromanipulators. Together with MRI guidance, clinicians could give intuitive and precise feedback to the ablation process, which is critical for function preservation of surrounding structures,” explained Dr Jason Chan of the Department of Otorhinolaryngology, Head and Neck Surgery at CUHK.

For their next steps, the researchers are planning to explore the potential for fully automated surgery, refine the workflow of the robot, conduct live animal trials and further reduce the size of the robot. “Miniaturizing the size of the robot will allow it to access more confined sites such as the nasal cavity and sinus cavity. We will also implement some specific image sequences to enable faster feedback of images to the robot,” said Dr Ka-Wai Kwok of the Department of Mechanical Engineering, Faculty of Engineering at HKU.