UC Santa Barbara mechanical engineer Elliot Hawkes and a national team of collaborators are developing an autonomous soft robotic system designed to perform time-sensitive thrombectomies, procedures that remove dangerous blood clots. The initiative, funded by the federal government’s Advanced Projects Research Agency for Health’s (ARPA-H) Autonomous Interventions and Robotics (AIR) program, seeks to expand access to this life-saving treatment across the United States. Hundreds of thousands of eligible stroke patients annually currently lack access to advanced surgical expertise, with only 12% receiving the necessary intervention, according to researchers.
Thromboses, or blood clots in veins or arteries, are a significant health concern, leading to hundreds of deaths daily in the U.S. While often a normal part of healing, these clots can obstruct circulation, become dislodged, and cause severe events such as strokes and heart attacks. Rapid intervention is crucial, as delays in treatment dramatically increase the risk of serious long-term disability or death and contribute significantly to healthcare costs. The highly technical nature of thrombectomies, which involve locating and carefully removing clots blocking vital blood flow, typically requires surgeons with years of specialized training. Hawkes noted that while the surgery can "basically reperfuse the brain," access is limited due to the difficulty of the procedure and a shortage of available surgeons.
The ARPA-H AIR program’s vision is to deploy autonomous systems in hospitals nationwide to meet the demand for these time-critical surgeries. A common approach to robotic surgery might involve training an automaton to wield existing rigid surgical tools. However, Hawkes explained that teaching a robot to safely replicate human surgeons' years of training within the project's approximately five-year timeframe was not feasible. Such an approach, using rigid tools, carries a significant risk of puncturing blood vessels or causing unforeseen problems due to patient anatomy or autonomous errors, potentially worsening the patient's condition.
Instead, Hawkes and his team chose to "reinvent the tools that are going inside the body," focusing on soft robotics. The resulting soft, vine-like robot mimics natural vine growth, extending from its tip when pressurized while its body remains stationary. This tip extension minimizes friction and force compared to instruments pushed from their base. The robot’s inherent softness and tip-based steering allow it to navigate complex, tortuous blood vessel paths while remaining compliant with the vessel walls. Once at its target, the robot can deploy specific tools to address the clot.
A key advantage of this design is its inherent safety, described by Hawkes as "safe by physics rather than by AI." While the physical design ensures safety, artificial intelligence will be integral to the robot's autonomy. AI will develop patient-specific optimized treatment plans and enable the robot to make necessary decisions throughout the procedure.
The primary hardware challenge for Hawkes and his colleagues involves miniaturizing the technology to fit within blood vessels. Hawkes has long envisioned diverse applications for soft, growing robots, including medicine and search and rescue. The project includes a two-year milestone check and a capstone demonstration at the end of the fifth year. Following successful research investigation, the robot could transition to clinical testing and, "shortly thereafter," potentially be used with live people. This development aims to ensure every person in the U.S. has access to life-saving thrombectomies, fundamentally changing outcomes for individuals affected by severe blood clots.
