Have we just got a sneak peek at how Ottava Surgical Robotic arms might move?

does McGill show insights into Ottava

So one of the theories is that Ottava will have the underpinnings of the Kinova arms for the Ottava system. It’s just a theory, but because the Monarch system uses bespoke Kinova robotic arms; it could be logical.

Monarch platform Kinova Robotic Arms

Kinova Auris Health Ethicon Johnson & Johnson

So JNJ Ethicon already knows how to deal with those arms and the control software etc etc.

It could be that they have dropped back to the original VERB design and underpinnings of arms - and understanding there are been a few agreements signed including SRI with VERB. That is possible. So the assumption of them using Kinova is speculation at best.

But when I look at the Ottava robot image I have it seems to me to be a potential Kinova based design. (or DNA under it.)

Sneak previews of Ottava at SRS

McGill department of surgery have been showing for over 7 months (it’s only come to my attention recently- so my bad.)

A set of table mounted Kinova based arms that use a daVinci instrument as the payload.

Plus you can see other videos of Kinova arms if you google.

https://www.surgicalroboticstechnology.com/articles/enhancing-robotic-surgery-with-a-plug-and-play-multi-robot-surgical-platforms/

Watch their video (below) as there is a lot to unpack here - but some points to consider for this particular system are:

Notice there is no port docking so this is more like and Asensus or Versius in terms of no port docking and arms locking to trocars.

It uses a remote centre by virtual pivot points. (Software.)

It uses side docking instruments much like a daVinci, Senhance or Versius

It has arm detection and collision detection - this uses a navigation style fiducial visual detection system. So combining some of the features of spine robots (Ortho robots)

But what it also demonstrates very very well - is that when you hold an instrument “like a dagger” - with this type of arm design... you end up with huge movements outside the dome, to get small movements inside the dome. That not only becomes an issue for bedside staff, but it can make the total reach of arms to become limited.

What the collision detection system shows - is with just two arms - how you can quickly end up in collision poses. -And that is a major concern of mine for a) bed mounted systems b) systems that do not have a V-wrist like Versius - (Which is a massive differentiator of that system.)

The V-wrist was born to allow smaller arm movements outside, but bigger instrument movements inside (much like the human arm does.) The original Ottava system (see below) has end joints that massively reduced overall arm movements: They were holding the ports and the instruments (in the Z axis) passed though the terminal joint. Very smart - but bulky in reality. And very complicated on how to do cabling of instruments.

Original Ottava Robot design

For the McGill system...

You can also clearly see the arrival at joint extension limits even when instruments are working in a fairly small internal working area. And this my friends is where “real world” positioning of trocars will get systems like this into trouble fast - in my opinion.

I think you either end up needing new port placements - moving the arms... or both. That combination of joint limits and clash possibilities is a double whammy for any bed mounted system that holds instruments side on. "Ya can't change the laws of physics Jim!"

So what about Ottava robotic system?

Well it is very hard for me to be sure if it is a Kinova based system or not - images are scant. But if it is, then I think the kinematic movements of these arms should be very similar to how Ottava arms potentially move.

If they have gone with a standard Z rail - as per the VERB patent below - then the terminal joints will be different - and how they achieve in and out Z axis movement of the instruments will be more akin to a daVinci or a Carina (Ronovo Surgical)

The z-rails will dock onto a port and use a remote centre as a pivot point. So the terminal joint movement might be quite different. Plus the arm kinematics to recreate Z axis movement might also be different. So keep that in mind.

VERB robotic patent diagram

However if they have ended up with a system closer to the original Ottava - or more like a McGill system above - with a static Z rail that is created virtually through software - then it should move exactly like we see in the video.

I am very very keen to see the full arm of Ottava so I can comment more. But based on what I see in the video, I would say a few things (My assumption is Ottava uses a Kinova arm as underpinnings.)

First - the quality will be excellent. -You can see the stability and smoothness of the movements of these arms. They have excellent kinematics. As they have very decent industrial underpinnings. I think that the joints look super smooth - and I can imagine quite a good durability for the joints. Perhaps with good long-term joint survival between servicing. Which is a big plus. The less you have to adjust joints - replace joints - the better.

I think they will have “Okay” reach (by the image I saw of Ottava above) but I do wonder at extremes of body size if they will struggle. Very large abdomens may be an issue for joint limits and reach.

Very small abdomens may likely be an issue for trocar placement and arm clash.

You can see this clearly in the video above - and they have the 1st joints mounted up at / above the table height. The fact Ottava comes from under (unless you can jack it up) - I worry a little about extreeme reach. (It's an issue for all robots, but bed mounting reduces your ability to move the spatial location of joint 1 as much.)

These arms are quite slim, and Ottava looks similar in diameter - which is nice. Again they look very very nice for McGill and Ottava. So that diamter will help a little with clash avoidance. But the fact that McGill have already got clash avoidance on “this” system makes me suspicious about clash management (again and again). Especially when you hold the instrument like a dagger (Z rail or not) and you constrain the arms to the bed.

(I do bang on... I'm sorry. But having lived hundreds of cases I know what this can mean for the teams in the OR.)

On thing I did notice at about 0.48 in the video is a latency gap between inputs and movements. (Take a look.) To me, it looked quite laggy. Now if Ottava is using this type of arms - and if this is an inherent built in issue of the arms... then Ottava may feel laggy?

But if not - and it is just software between inputs and outputs. And the engineers have worked hard on the software - and it really is just a software issue then this may not be an issue for Ottava. Knowing how long they’ve had to work on this (and the smart people that have), and the mass of KOLs they’ve had working on it - I must assume that the lag is absolutely acceptable. But as I see it in the video above... I'll have my eye out.

Final thoughts

As Ottava is now under review for IDE (well done to all for getting that submission in) - and it will soon be used in first humans; then we will all see if all the “lab data” will translate into real world patients. And if I’m just overblowing the concerns over clash - and joint limits. I think that will become clearer as they come out of IDE in a few years, and Ottava gets FDA clearances. (They will.) The thing to watch for will be BMI constraints on the clearances.

If I’m being pessimistic, there will be a full open and wide range of BMIs from the very small to the very big patients.

If my suspicions become true - then the system will get cleared, but with maybe some strict guardrails on BMI, and some limitations / rules on port placements. And that will then translate into a commercial acceptability issue not a regulatory issue.

I think - if - and it’s a big if - they have gone with Kinova, then they made a smart choice to not try and build bespoke arms. Especially if they have nothing novel about Z rails - or terminal joints and "wrists."

Using a lot of "off the shelf" adds some complexity during development; it may not give 100% of all the characteristics that they wanted. BUT… when it becomes commercial - having Kinova underpinnings would pay dividends in arm durability, servicing and servicing costs. If they have gone with one of the major manufacturers (Kinova or other) - then that will indeed show to have been a very smart move long term.

Very keen to actually see the real Ottava out in the wild, or even a video, or a few more images... It will tell me a lot about any challenges they might or might not have in the future.

These are just opinions of the author for educational purposes only based off publically available information and a deep experience with surgical robotics. All trademarks and product names belong to their respective owners.