Yaw, pitch and roll not behaving as expected

[DanEgert]

Hi, great tool, it really helps to see the structures in 3D.

I'd like to look into inserting a NP2 probe angled in the coronal plane. I thought yaw should control this. But it seems like at 90deg pitch (when probes are inserted along the DV axis as usual), changing yaw and roll has the the same effect. These angles behave as expected when the probe pitch is 0, though.

Hope this makes sense, please let me know if I missed something.
Dan

[dbirman]

Hey Dan -- this is a common confusion, and it's a little more confusing because I'm just noticing that the probe-axis graphics are not showing up right now. I'll get that fixed.

What's happening is that Yaw is an extrinsic rotation around the dorsal-ventral axis. Roll is an intrinsic rotation around the probe's depth axis. When you set the pitch to 90 you are aligning these two axes, so the two kinds of rotation have the same effect. It's much more obvious that they are independent if you set a probe to 45 degree pitch and observe how yaw and roll affect the probe.

I'll keep this open to remind me to fix the issue with the probe-axis lines not showing up.

[Sepidak]

@dbirman Hi Dan, it's great to read that this bug is reported and flagged to be fixed. I just wanted to add that the ability to tilt the probe in roll while keeping the pitch at 90 is critical, particularly for the 4-shank probe. Changing the pitch from 90 to 0 will completely change the planned trajectory for the 4-shank configuration. I hope this will be considered in the fix. We are looking forward to using the tool once the tilt issue is resolved. Thank you

[dbirman]

@Sepidak I believe Pinpoint already does this correctly, I left this issue open just to remind myself that there's a graphics issue that I need to fix.

If there's an existing issue with the probe controls can you show an image of what you're seeing? From the description I'm not sure what the problem is. All three controls of the probe (pitch/yaw/roll) are independent except when the depth axis of the probe is aligned with the vertical axis (when pitch=90) because then rotation around the vertical axis (yaw) has the same effect as rotating around the depth axis (roll).

[Sepidak]

Thank you for your reply Dan. The issue is the same as you mentioned and is also flagged above. We need to tilt the 4-shank probe in roll while the probe is aligned with the vertical axis (pitch = 90 and not less). When pitch is set at 90, the tilt option behaves the same as yaw (just rotates the probe around the vertical axis). For further clarification, I'm attaching a picture of the trajectory we would like to achieve. I believe this is a common application for many other users too. If there is any way to achieve this trajectory in pinpoint I'd appreciate you sending us the details. We tried everything but didn't manage.

[dbirman]

The key thing to understand in Pinpoint is that the roll is applied independently from the other axes. If you position the tip of the reference shank at your target location and set the appropriate angles you can then independently rotate the 4 shanks to any orientation around the probe axis.

Here is your insertion:

The steps to achieve this were:

(1) Pitch the probe to the angle you want (here ~75 deg)
(2) Yaw the probe so that the reference probe shank is at the position you want, so in this case we have to yaw by 90 degrees to get the reference tilted off the sagittal plane
(3) Roll the probe to align the 4 shanks into the coronal axis

[dbirman]

I know it's the yaw/pitch/roll system is confusing for people used to doing planning using angles off the coronal/sagittal planes, which is why I want to add those options to the planner. It's on my to-do list.

[Sepidak]

The problem with the example you suggested is that pitch is at 75, meaning the 4 shanks are not all aligned in the same plane rostro-caudally, so the tip of those shanks are going to be at a different coronal plane than the top. This is not what we want to achieve and we certainly never do that in our experiments. If you could also send a top view it could be more obvious. That's why I mentioned that the pitch should be 90 degree and not less, yaw at 0 (we don't want any rotation around the vertical access), and then tilt the probe by 20 degree with roll.

[dbirman]

@Sepidak I'm not sure I understand your concern fully, the four shanks are on the same coronal plane. In Pinpoint the pitch is the axial angle (i.e. angle relative to the axial plane, independent of yaw). You're absolutely right that with a rotation of (0, 90, 20) the shanks will not be on the same coronal plane. In the example I provided, as you can see in a top-down view, the 4 shanks are aligned in the same coronal plane:

Please just re-create it in Pinpoint using the coordinates/angles I showed above and play with it, it should be the identical insertion to what you showed in your image above.

[Sepidak]

Thank you @dbirman. This seems to work, although we're still trying to get our head around it! :) Is it correct to say that for a 18 degree tilt angle we should set pitch at 72 ( = 90-18) with this configuration set for yaw and roll? This is how I understand it.

[dbirman]

Yes that's right the tilt angle is the complement of the pitch in our definition.