My Ph.D. research is focused on the development of dexterous surgical robots to assist physicians. During my first year as an NSF Graduate Research Fellow, my primary focus has been on the development of dexterous robotic surgical assistants for CT-guided Interventional Radiology procedures.
Robotic Assistant for CT Guided Interventional Radiology Procedures:
With the guidance of my advisor, Professor Michael Yip, I am developing and testing CT compatible robotic needle placement assistants to perform clinical testing by the end of this calendar year. We are primarily focusing on transthoracic needle lung biopsy due to its wide prevalence and high mortality rate. Diagnosis requires tissue samples, and early diagnosis is strongly correlated with improved survival rates. However, nodules under 6mm in diameter are infrequently biopsied due to targeting difficulties. A robotic assistant will help physicians target these smaller nodules, enabling earlier diagnosis. Few robotic assistants are capable of lung biopsy due to the lungs' large physical size and significant physiological motion. This past year, I published a paper on and released designs for an Open-Source 7-Axis, Robotic Platform to Enable Dexterous Procedures within CT Scanners. To the best of our knowledge, this is the first open-hardware image-guided biopsy robot of its kind. This system uses a low-profile serial link design, which provides a large dexterous working area for needle placement within the CT scanner bore. By using a large, fully active external positioning stage and a low profile in bore cable-driven needle manipulator, the platform's in-bore form factor in minimized. This maximizes the free space within the scanner for needle positioning and manipulation. The system's performance was evaluated: collision-free reachable workspace in simulation, repeatability using a camera-based tracking system, and physician teleoperated performance on a phantom-biopsy under CT guidance. Currently, I am developing the second generation of this system, CRANE, with significantly improved dexterity and robustness. This revision involves a full hardware redesign, increases the joint travels by at least forty percent, adds a novel clutching needle driver, and adds a Z-axis. The system has two redundant degrees of freedom, which allow for null-space control to optimize for secondary objectives, including dexterity maximation and avoiding collisions. I am collaborating with physicians to collect clinically relevant data. Through this project, I mentor several masters and undergraduate students.
Planetary Exploration Serpentine Robot:
Also, this past summer, I mentored (along with Professor Yip and Florian Richter) several undergraduate students for an exploratory robot project. Together, we developed the first serpentine robot proof-of-concept demonstration for the NASA-JPL Exobiology Extant Life Surveyor
(EELS) program. This serpentine robot is composed of several repeated links of active universal-joints and Archimedes screws for positioning and propulsion. Resulting in a paper at the end of summer, this work is continuing with an increased focus on controls and kinematics modeling.
Finally, in collaboration with UCSD Structural Engineers, I am continuing to develop earthquake simulation tables for educational outreach.
I completed UC San Diego's Ph.D. program's course requirements, finishing ECE276B (Planning and Learning Robotics) and BENG280C (Imaging Cardiovascular Disease Spring quarter. Additionally, I passed my Preliminary Exam with a research presentation titled "Development of dexterous MR/CT compatible Needle Placement Robots" and course examination on ECE276B (Planning and Learning Robotics) and ECE276C (Robot Reinforcement Learning).