AI Gets Hands: What’s New in Healthcare Robotics?

Jivesh Sharma, M.D.

Welcome to Health I T dot com. I'm Jivesh Sharma.

A blood draw. An ultrasound scan. A delivery down a hospital corridor. Robotics is taking on very different kinds of work in healthcare, and the last few months have brought developments worth understanding.

In July, the FDA authorized a surgical system with robotic arms integrated into the operating table. In August came authorization for a supervised blood-draw robot. In September, researchers reported a robot that acquires thyroid ultrasound scans. And in October, Google demonstrated an AI model guiding a real robot arm through a physical task.

Let's look at what changed, how the technology works, and where patients and care teams could notice it.

Talia

I'm Talia, your AI cohost. Hospitals have used robots for years. What makes this wave different?

Jivesh Sharma, M.D.

The change is in how machines interpret a situation and choose their next action. A carefully programmed movement works well when everything stays in the same place. A hospital is full of variation. Someone moves a cart. A doorway becomes crowded. A patient shifts position.

Researchers are connecting models that understand images and language with systems that choose physical actions. Robots can learn from demonstrations, and simulation lets developers practice variations before testing the real machine.

Google's July Gemini Robotics 2 announcement described advances in whole-body control and dexterity. Its October Argon demonstration made the idea easier to see: a real robotic arm set a fallen mug upright and placed a napkin inside it. The model used camera observations and a sequence of available actions.

That was a general-robotics demonstration, not a medical procedure. But watch the pattern: look, act, then look again. The next move depends on what actually happened. That feedback is a crucial part of turning AI into a machine that can work around people.

Talia

What actually happens between a robot seeing my neck and touching it with an ultrasound probe?

Jivesh Sharma, M.D.

First, it needs to connect what it sees with where its tool is. A camera can help locate the body; ultrasound can reveal structures beneath the skin. The robot needs a way to translate between those views and its own coordinates.

That alignment is called registration. Think of lining up a map with where you are standing before following the directions.

Then comes contact. The probe needs enough contact to produce an image, with limits on how firmly it presses. Different devices solve this differently. An FDA-cleared remote ultrasound system called Melody, for example, has a specified limit on the force it applies. The FDA also requires testing of motion accuracy, speed and force for the blood-draw robot.

Finally, the machine needs feedback. The patient may have shifted, or soft tissue may have moved. Finding a target once and reaching it accurately throughout a procedure are different engineering problems. Perception, controlled movement and repeated observation have to work together.

Jivesh Sharma, M.D.

Some of the most practical progress is happening outside the procedure room.

In August, Diligent Robotics announced the rollout of Moxi 2.0 to hospital customers. The company describes upgraded computing and sensors, learning from fleet experience, and better recovery when the robot encounters an unexpected situation.

In September, NEURA Robotics announced a healthcare business focused initially on moving empty hospital beds and sterile goods. That is equipment transport, with the robot taking on part of the physical work around care.

These applications matter because a delivery still has to happen when a nurse or technician has something more urgent to do. A robot that gets a needed item to the right place can potentially give that person time back. Whether it does depends on the whole job, including loading, handoffs and interruptions.

We often imagine a humanoid at the bedside. A useful hospital robot may simply be a very dependable courier.

Talia

Which of these robots actually touches a patient today?

Jivesh Sharma, M.D.

Aletta is authorized for that specific job. The FDA authorized it in August to collect blood from adult outpatients under a trained phlebotomist's supervision.

It combines near-infrared imaging with ultrasound, including Doppler information about blood flow, to find a suitable vein. Without a suitable target, it does not attempt the draw. If the arm moves too much, the FDA says the needle automatically detaches and the draw stops.

A trained professional starts the session, supervises and handles exceptions. That is bounded autonomy: the machine performs a defined procedure within a larger service run by people.

Jivesh Sharma, M.D.

Ultrasound shows another important step. In September, a study in Nature Partner Journals, Digital Medicine evaluated autonomous thyroid scanning in two hundred and sixty-two participants at three institutions.

Here, the robot acquired the images, and separate AI software helped assess them. That distinction matters. Software that can interpret a scan still needs someone or something to obtain the right images.

The researchers reported agreement with conventional ultrasound in thyroid measurements, and faster AI-assisted analysis. Robotic scanning itself took longer. The combined screening workflow also missed roughly one in five biopsy recommendations, so expert oversight remained important. Several authors worked for the companies involved.

The technical opportunity is to extend automation to image acquisition. If that becomes reliable across more patients and settings, imaging could become available in places that struggle to provide it today.

Jivesh Sharma, M.D.

Surgery shows another kind of progress. In July, the FDA authorized Johnson and Johnson's OTTAVA system. The company describes four robotic arms built into the operating table itself.

According to the FDA, staff can reposition the patient by adjusting the table without undocking the system. The company calls the coordinated movement of table and arms twin motion. The surgeon remains in control of the operation.

The engineering idea is easy to miss: the table, patient and instruments have to work as a coordinated physical system. Improving that coordination is a different advance from teaching a machine to decide how to operate.

For that more independent future, DARPA launched a surgical robotics competition in September. One track calls for collaboration with a surgeon; another calls for an independent trauma procedure. The finals are planned for March 2027. Those are research targets, including care in settings where surgical help may be scarce.

Talia

So when would I actually notice any of this?

Jivesh Sharma, M.D.

Over the next few years, I would watch the ordinary parts of a healthcare visit. Does the sample reach the lab sooner? Can a local clinic offer a scan that previously required a long trip? Can a team add another testing session?

Those are opportunities to test as these systems develop. They depend on reliable equipment, trained staff and the rest of the service being ready to use the capacity.

As an oncologist, I think about patients who return for blood tests and imaging repeatedly. Less waiting and fewer journeys would matter to them. The robot would not need to resemble a doctor for that improvement to be valuable.

Here is the deeper point: clinical capacity can grow one task at a time. Better AI helps machines interpret a changing world. Better engineering lets them act in it. The useful result is a completed delivery, scan or blood draw that helps the next step of care happen.

That is what I will be watching as healthcare AI moves beyond the screen.

Jivesh Sharma, M.D.

Thank you for listening. If this was useful, follow Health I T dot com for the next conversation. On YouTube, you can like this episode and subscribe. And tell us what you'd like us to explore next. This episode uses an authorized AI-generated version of my voice. It is general education, not personal medical advice.