Tweet
Last month, I posed the following question to the Biomch-L community:
I am trying to measure lower limb kinematics during a hip exam on a supine patient. Traditional motion analysis techniques estimate the hip joint center from four skin markers placed on the anterior and posterior superior iliac spines (ASIS/PSIS). On a supine subject, the PSIS markers are not visible. Are there alternative bony landmarks or different methods that could be used?
Ashley Kapron
Graduate Research Assistant
University of Utah
Department of Bioengineering
work: (801) 587-5200
email: Ashley.Kapron@utah.edu
Thanks to all who replied! Here is a summary of the responses I received:
Here are some options:
1) As long as you have at least three visible non-collinear points, you could digitize the landmarks you want to use in a standing trial. Then you can calculate the position of these landmarks during the supine trials. The problem with this is that the pelvis markers will probably not remain in the same position relative to the pelvis during the standing and supine trials.
2) Kinematic fit using inverse kinematics algorithms like in OpenSim.
3) Functional joint center methods.
I'd go with either 2 or 3, as they don't rely on existing landmarks. The markers just need to remain stationary with respect to the underlying bone.
-Brian Schulz, Ph.D.
You can do this with a new version 4.1 of the AnyBody Modeling System that is about to be released. It will allow you to place the markers anywhere you want and determine the hip joint center by means of optimization.
The technique was recently published: http://dx.doi.org/10.1080/10255840903067080
A computationally efficient optimisation-based method for parameter identification of kinematically determinate and over-determinate biomechanical systems. Michael Skipper Andersen ; Michael Damsgaard ; Bruce Mac-Williams ; John Rasmussen.
In a few days you will be able to download a free trial version from www.anybodytech.com. Make sure you get version 4.1. If you do so, the easiest way to try it out is to go to Help -> Tutorials and select the tutorial on "Making things move". Free support is available at forum.ayscript.org
-John Rasmussen, Ph.D.
I developed a method for locating the HJC with seated subjects (the posterior markers are obscured when a person is seated in a chair/wheelchair/car seat). The method is published in:
TR Bush and P Gutowski. An approach for Hip Joint Center Calculation for Use in Seated Postures. Journal of Biomechanics, volume 36, 2003, pp. 1739-1743
I believe that you could use this approach with a flexed knee in a supine position - or you may be able to build from the approach to suit your needs. The issue you will have is the fact there are two possible solutions, but if your subject is staying in one location, you can determine which solution is appropriate.
-Tamara Bush, Ph.D.
You can do a "static" calibration with all the required markers (ie, ASIS, PSIS) plus one more placed somewhere on the pelvis such as the iliac crest. Then remove the PSIS markers prior to supine capture. The two "virtual" PSIS markers can then be referenced to a pelvic reference frame defined by the other 3.
- Arnel Aguinaldo, MA, ATC
In that sort of circumstance, if 3D motion is of interest, anatomical landmarks are no longer feasible references, i.e. a technical reference frame shall be defined, and locations for these landmarks identified by calibration This can be either with direct mounted markers, or with a pointer, or by a functional approach, apparently very appropriate in the case of the hip joint (you can find more details on these general techniques over our relevant papers; also much has been done in Rome more explicitly and recently on the functional approach, by Cappozzo-Camomilla-Cereatti).
In addition, I shall point out to you that hip joint centre definition in supine position is an issue addressed brilliantly a long ago in surgical navigation. These systems implement algorithms for functional-based calculation of this point in a femoral reference frame with very good accuracy (though mainly because of the bone pins utilized), exactly in that position because of the operative techniques in joint replacement. This is a field worth to be investigated, though little has been published because relevant companies are very jealous of these techniques.
- Alberto Leardini, DPhil
NIH had a method using the greater trochanters - Each hip center is 1/4 of the way between the greater trochanters. I don't know that they ever published it though - maybe someone else can give you a reference for it.
- Dustin Bruening, Ph.D.
Could you establish the segment and joint center(s) in the standing position and then remove the PSIS or Sacral marker for your movement in supine. You would use the remaining markers to ID the movements. You could if necessary add a marker or two that would be visible during your task.
- Robert Streb, PT, Ph.D.
I happened upon the same problem when quantifying supine hip rotations and flexion. Although I initially thought that having a trio of markers slightly lateral to the ASIS's would help solve the problem, I also found that the skin movement artifact is a huge problem here.
Perhaps that will not be the case if you are not planning on flexing the hip near 90. But if you are, be forewarned. The flexion of the hip causes a lot of crimping of the skin and hiding of the markers. Frankly, I would not trust it. Nor would I trust angles calculated from a rigid body (with a cluster of markers) that is attached via a fin, due to the skin motion.
It is extremely difficult to monitor pelvis motion, especially in the sagittal plane. We resorted to using a blood pressure cuff under the lumbar spine, and encouraged participant feedback, to monitor motion in the sagittal plane. The other planes were less of a problem, as pelvis rotation or side bending are less subtle.
- Janice Moreside, Registered Physiotherapist, PhD Candidate
I don't know which kind of motion analysis system you are using, but if you use visual 3D software, in its modeling part you are able to locate hip center with this simple formula: 0.25*DISTANCE(RGT,LGT) in which RGT and LGT, are the right/left greater trochanters.
- Farzaneh Yazdani, Research Officer
I have used Vicon to do a supine project. For that one, we used ASIS markers and iliac crest markers to define the pelvis, and makers on the anterior, lateral and distal thigh to define the "femur" and brought the data into C-Motion's Visual 3D. The other way that I did it with a Motion Analysis Corporation system (for a prone project) was to again have makers on the iliac crest and ASIS and PSIS to start. We collected a standing trial with the markers on to establish the relationship between the iliac crest and ASIS. We then removed the ASIS before the subject got on their stomach. Then we used "virtual markers" within MAC's EVaRT to "recreate" the ASIS. Sounds like your big challenge is subject size. To do any of the techniques I'm suggesting, you wouldn't want a lot of skin marker movement, so it might not work Visual 3D does have a functional definition of hip joint center. I'm not sure if you could get enough movement in supine to establish it, but you might.
- Cara Lewis, PT, PhD
I am trying to measure lower limb kinematics during a hip exam on a supine patient. Traditional motion analysis techniques estimate the hip joint center from four skin markers placed on the anterior and posterior superior iliac spines (ASIS/PSIS). On a supine subject, the PSIS markers are not visible. Are there alternative bony landmarks or different methods that could be used?
Ashley Kapron
Graduate Research Assistant
University of Utah
Department of Bioengineering
work: (801) 587-5200
email: Ashley.Kapron@utah.edu
Thanks to all who replied! Here is a summary of the responses I received:
Here are some options:
1) As long as you have at least three visible non-collinear points, you could digitize the landmarks you want to use in a standing trial. Then you can calculate the position of these landmarks during the supine trials. The problem with this is that the pelvis markers will probably not remain in the same position relative to the pelvis during the standing and supine trials.
2) Kinematic fit using inverse kinematics algorithms like in OpenSim.
3) Functional joint center methods.
I'd go with either 2 or 3, as they don't rely on existing landmarks. The markers just need to remain stationary with respect to the underlying bone.
-Brian Schulz, Ph.D.
You can do this with a new version 4.1 of the AnyBody Modeling System that is about to be released. It will allow you to place the markers anywhere you want and determine the hip joint center by means of optimization.
The technique was recently published: http://dx.doi.org/10.1080/10255840903067080
A computationally efficient optimisation-based method for parameter identification of kinematically determinate and over-determinate biomechanical systems. Michael Skipper Andersen ; Michael Damsgaard ; Bruce Mac-Williams ; John Rasmussen.
In a few days you will be able to download a free trial version from www.anybodytech.com. Make sure you get version 4.1. If you do so, the easiest way to try it out is to go to Help -> Tutorials and select the tutorial on "Making things move". Free support is available at forum.ayscript.org
-John Rasmussen, Ph.D.
I developed a method for locating the HJC with seated subjects (the posterior markers are obscured when a person is seated in a chair/wheelchair/car seat). The method is published in:
TR Bush and P Gutowski. An approach for Hip Joint Center Calculation for Use in Seated Postures. Journal of Biomechanics, volume 36, 2003, pp. 1739-1743
I believe that you could use this approach with a flexed knee in a supine position - or you may be able to build from the approach to suit your needs. The issue you will have is the fact there are two possible solutions, but if your subject is staying in one location, you can determine which solution is appropriate.
-Tamara Bush, Ph.D.
You can do a "static" calibration with all the required markers (ie, ASIS, PSIS) plus one more placed somewhere on the pelvis such as the iliac crest. Then remove the PSIS markers prior to supine capture. The two "virtual" PSIS markers can then be referenced to a pelvic reference frame defined by the other 3.
- Arnel Aguinaldo, MA, ATC
In that sort of circumstance, if 3D motion is of interest, anatomical landmarks are no longer feasible references, i.e. a technical reference frame shall be defined, and locations for these landmarks identified by calibration This can be either with direct mounted markers, or with a pointer, or by a functional approach, apparently very appropriate in the case of the hip joint (you can find more details on these general techniques over our relevant papers; also much has been done in Rome more explicitly and recently on the functional approach, by Cappozzo-Camomilla-Cereatti).
In addition, I shall point out to you that hip joint centre definition in supine position is an issue addressed brilliantly a long ago in surgical navigation. These systems implement algorithms for functional-based calculation of this point in a femoral reference frame with very good accuracy (though mainly because of the bone pins utilized), exactly in that position because of the operative techniques in joint replacement. This is a field worth to be investigated, though little has been published because relevant companies are very jealous of these techniques.
- Alberto Leardini, DPhil
NIH had a method using the greater trochanters - Each hip center is 1/4 of the way between the greater trochanters. I don't know that they ever published it though - maybe someone else can give you a reference for it.
- Dustin Bruening, Ph.D.
Could you establish the segment and joint center(s) in the standing position and then remove the PSIS or Sacral marker for your movement in supine. You would use the remaining markers to ID the movements. You could if necessary add a marker or two that would be visible during your task.
- Robert Streb, PT, Ph.D.
I happened upon the same problem when quantifying supine hip rotations and flexion. Although I initially thought that having a trio of markers slightly lateral to the ASIS's would help solve the problem, I also found that the skin movement artifact is a huge problem here.
Perhaps that will not be the case if you are not planning on flexing the hip near 90. But if you are, be forewarned. The flexion of the hip causes a lot of crimping of the skin and hiding of the markers. Frankly, I would not trust it. Nor would I trust angles calculated from a rigid body (with a cluster of markers) that is attached via a fin, due to the skin motion.
It is extremely difficult to monitor pelvis motion, especially in the sagittal plane. We resorted to using a blood pressure cuff under the lumbar spine, and encouraged participant feedback, to monitor motion in the sagittal plane. The other planes were less of a problem, as pelvis rotation or side bending are less subtle.
- Janice Moreside, Registered Physiotherapist, PhD Candidate
I don't know which kind of motion analysis system you are using, but if you use visual 3D software, in its modeling part you are able to locate hip center with this simple formula: 0.25*DISTANCE(RGT,LGT) in which RGT and LGT, are the right/left greater trochanters.
- Farzaneh Yazdani, Research Officer
I have used Vicon to do a supine project. For that one, we used ASIS markers and iliac crest markers to define the pelvis, and makers on the anterior, lateral and distal thigh to define the "femur" and brought the data into C-Motion's Visual 3D. The other way that I did it with a Motion Analysis Corporation system (for a prone project) was to again have makers on the iliac crest and ASIS and PSIS to start. We collected a standing trial with the markers on to establish the relationship between the iliac crest and ASIS. We then removed the ASIS before the subject got on their stomach. Then we used "virtual markers" within MAC's EVaRT to "recreate" the ASIS. Sounds like your big challenge is subject size. To do any of the techniques I'm suggesting, you wouldn't want a lot of skin marker movement, so it might not work Visual 3D does have a functional definition of hip joint center. I'm not sure if you could get enough movement in supine to establish it, but you might.
- Cara Lewis, PT, PhD