Interpolation/Extrapolation 3D Points

6 次查看（过去 30 天）

显示更早的评论

Oliver Kumar 2015-10-16

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/248914-interpolation-extrapolation-3d-points

评论： Huy Le Van 2021-8-30

Hello

I have 5 markers (L1, L2, L3, L4, L5) all on the lumbar spine. L1 is missing.

From L2, L3, L4, L5 i have the x,y,z coordinates for 360frames. Is it possible to extrapolate the coordinates for L1?

The markers can be assumed to be equaly distributed.

Thank you!

7 个评论
显示 5更早的评论隐藏 5更早的评论

Walter Roberson 2015-10-16

Oliver, please keep in mind that the volunteers here know MATLAB better than they know how the spine works, and especially we know MATLAB better than we know your experimental setup.

1) Are the 5 markers to be assumed to be on a single short part of the spine that is completely rigid and flat, like trying to predict where the "1" is on a wooden ruler when we know where "2", "3", "4", and "5" are?

2) Are the markers to be assumed to be on a segment of the spine that has a maximum bend from the other segments, and you are fine with getting back coordinates that are anywhere in the range of that maximum bend, or alternately getting back constraints on the range of positions where the L1 might be?

3) Are the markers to be assumed to be on a segment of the spine that moves in consistent ways, and that what is needed is to track the markers over time forward and backwards in order to figure out where the L1 must have been because the other markers on the spine would have moved differently if it had been elsewhere?

Option #1 is not difficult to implement and does not require knowledge of anything other than some easy geometry.

Option #2 is a bit more tricky but should not be too bad

Option #3 requires a lot more knowledge of spinal mechanics and would likely not be all that easy to implement.