Two research teams at the Johns Hopkins Whiting School of Engineering will collaborate with device manufacturer Philips North America and software developer Kitware on newly funded efforts to broaden access to curative treatments for stroke, a leading cause of death in the U.S. that kills more than 100,000 people each year. The teams are part of projects funded by the Advanced Research Projects Agency for Health (ARPA-H) program for Autonomous Interventions and Robotics (AIR).
John C. Malone Associate Professor Mathias Unberath is leading a scientific group under the direction of software company Kitware, which received up to $17.5 million from ARPA-H to create virtual simulations and validation testing on four ARPA-H-funded robotic solutions for stroke.
Unberath and the Johns Hopkins team will develop simulation environments and testing datasets for robotic devices that account for the physical variability inherent in human anatomy and blood vessels, as well as changes in device movement, the flow of imaging contrast through blood vessels, blood flow, and other physical properties.
Mechanical engineer Rajat Mittal, who is working with Unberath on the project, says that while many physical systems rely on how water and air flow in an environment, the biology of blood vessels comprises a complex mix of substances that must navigate through highly flexible structures. He plans to use computing methods that rely on GPUs—or graphics processing units, which can handle math and data processing much faster than standard CPUs—to train the robots in real time.
“Our goal is to create publicly available resources that other scientists can build on to test and validate autonomous robotic medical devices,” says collaborator Gregory D. Hager, the Mandell Bellmore Professor of Computer Science and director of the Laboratory for Computational Sensing and Robotics.
Teams will use virtual simulations to train the robots; Hager says it’s essential to validate those simulations for accuracy and to gauge how well they reflect realistic scenarios among stroke patients.