Keep pulling the thread on Yossi Bar.
LEM Surgical is developing a surgical robot with an upper torso humanoid architecture.
Yossi Bar states that to his knowledge, no existing surgical robot has the capability of proprioception, which would allow it to operate multiple tools without direct, continuous visual guidance on its manipulators.
LEM Surgical's strategy is to create a general-purpose hard tissue robot that is not tied to proprietary implants and can operate with a wide variety of tools.
LEM Surgical has received FDA clearance for its system to be used with various screws in spine surgery, demonstrating its tool-agnostic capability.
In soft tissue robotic systems like the Da Vinci, the surgeon acts as the controller and closes the feedback loop, which means the robot itself does not need to be independently accurate.
Yossi Bar characterizes the progress of hard tissue robotics as being like the "slow brother" compared to the rapid advancement of soft tissue robotics like Intuitive's Da Vinci.
Yossi Bar believes the full humanoid form factor, like the Tesla robot, is the wrong design for surgery because complex parts like fingers are unnecessary when surgeons use tools like needle drivers.
Yossi Bar believes the success of the Da Vinci robot is due to its architecture as a general-purpose soft tissue robot.
A key challenge for surgical robotics companies is that FDA clearances are granted for specific clinical applications, not for general technological capabilities.
Yossi Bar asserts that companies with successful knee surgery robots will struggle to create even basic applications in spine surgery due to the specialized and limited architecture of their systems.
Yossi Bar believes many large robotics companies have "maxed out" the capabilities of their current platforms and are now focused on incremental optimizations like making them smaller or cheaper, rather than adding significant clinical value.
LEM Surgical has a partnership with NVIDIA to leverage developments in physical AI for its surgical robotics platform.