This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore.
Until we come up with something less unpleasant, all of the most reliable ways of determining whether you’re playing host to even the teeniest tiniest little bit of COVID-19 that we have right now involve some flavor of jamming a sampling device into a mucus-y orifice that doesn’t appreciate having sampling devices jammed into it. Whether it’s farther up your nose than you thought possible or down your throat to the point at which you involuntarily attempt to punch the nurse in the face (I’m very sorry about that), taking samples of disgusting goo is no fun for patients and probably even less fun for medical professionals, especially when there is a reasonable chance that the mucus you’re now elbow-deep in is effervescing with COVID.
The process of taking oropharyngeal-swab sampling is dull, dirty, and dangerous all at once, making it an ideal task for a robot to do instead—as long as that robot can avoid violently stabbing you in the throat.
In a paper published in IEEE Robotics and Automation Letters, researchers from the Shenzhen Institute of Artificial Intelligence and Robotics for Society introduce the robot in the video above, designed to protect the safety of medical staff during the sampling process as well as free them up to do more useful and less gross things, since let’s face it, taking these samples is not fun for anyone.
Yes, force sensing is enabled that absolutely should stop motion of the arm if it runs into something, but that’s in software, right?
The researchers note that “the human throat is fragile and easily injured…. [E]nd-effectors designed on the basis of a rigid body structure may cause physical injuries of the oral cavity during improper operation.” To make things a little less unsafe, this system combines a UR5 arm with a disposable soft micropneumatic actuator hooked up to a force sensor to do the actual in-throat swabbing action, with a vision sensor to determine where to prod. You can see in the video that excessive force on the actuator will cause the robot to stop moving, and that it has no trouble safely taking samples from actual human volunteers.
The concern that I personally would have with this system is that as designed it appears as though the back of my skull would be well within the UR5’s workspace, or at least within the reach of rigid components attached to the arm. Behind that deformable actuator is a very nondeformable linear actuator backed up by a powerful robotic arm with joints that can move at 180 degrees per second. Yes, force sensing is enabled that absolutely should stop motion of the arm if it runs into something, but that’s in software, right? I would feel better if the system was designed so that the arm, the linear actuator, and anything else rigid were arranged in such a way that it would be physically impossible for them to injure me—say, by setting things up so that I have to adjust how far I am from the system rather than the system being able to adjust itself to reach me and beyond. The researchers do acknowledge that “mechanical [constraint] implementation is usually safer for an open minimally invasive surgery,” but they feel (quite correctly) that software safety is cheaper and more convenient. It just may be less, you know, safe.
To be fair, step one here is to make sure that the system can reliably take samples, because if it can’t do that, then why even bother, right? PCR testing (using a reference gene rather than COVID) to verify the effectiveness of the sampling technique showed that samples taken by the robot were fully qualified, on par with samples taken by a professional human. Next, the researchers plan to “improve the robustness of the system and move forward to clinical testing.”
- Video Friday: Robots Help Keep Medical Staff Safe at COVID-19 ... ›
- Autonomous Robots Are Helping Kill Coronavirus in Hospitals ... ›
- Telepresence Robots Are Helping Take Pressure Off Hospital Staff - IEEE Spectrum ›
- How Robots Became Essential Workers in the COVID-19 Response - IEEE Spectrum ›