1) choose locations with little muscle mass underneath individual markers

2) Consider what motions or gestures are to be recorded.  Test locations you consider appropriate with a point of reference (fixed) and measure the variations.  I did this by measuring inter-marker distance variations over the variety of motions expected. 

3) consider inter-marker placement error too - try to find easy to locate placements which are well defined (ex. bony landmarks are great when possible; if not - 'half-way between the top of the shoulder and the armpit along the depression associated with tendon of the long head of the biceps, maybe sufficient).

Best wishes

Sarah,

In addition to the ideas already mentioned (which I'd agree are very useful), you can also configure segment markers into a cluster with known geometry, and use various methods to filter out motion artifact on the segment by using cluster geometry and some kind of shape analysis or filter. One possible approach:

https://www.ncbi.nlm.nih.gov/pubmed/21607885

Hope that helps.

Best,

Sean Osis

Thank you! 

Kind regards,

Sarah 

I second the idea of using clusters. If you do that, make sure the markers are affixed to a hard surface which you wrap onto the segment. For one project we were looking at joint accelerations (which will dramatically amplify any motion artifact) and these data became much cleaner upon switching to hard plastic clusters.

Also, using markers located at the ends of segments can increase the amount of tissue motion artifact (even thought this is the best practice if not using a cluster model). Using clusters can reduce this issue by tracking the segment only with the mid-segment cluster during motion trials and using the end-of-segment markers for defining the joint positions. Vicon may not let you do that very easily, but if you have access to other motion analysis software, such as Visual3d, that could be an option for you.

Hi Sarah,

I also second the idea of using cluster - rigid or not.  

Since Nexus 2, Vicon has an implemented algorithm that helps to reduce soft tissue artefact.  it is called "optimum common shape technique".  The link below gives a brief description:

https://www.vicon.com/faqs/software/what-is-optimum-common-shape-technique-ocst 

I highly recommend you to have a look into the work that has been done by Cappozzo and his team (Capello, Learddini, Fantozzi and Stagni), especially their idea of double calibration.  Here are few links: 

https://www.ncbi.nlm.nih.gov/pubmed/15698706

https://www.ncbi.nlm.nih.gov/pubmed/15977729

https://www.ncbi.nlm.nih.gov/pubmed/18768320

Some extremely good work has been done by Cheze and Dumas. This article summarises few existing soft tissue artefact compensation techniques:Add your answer

https://www.ncbi.nlm.nih.gov/pubmed/19640540

Hope that helps you.

Best regards,

Bertrand

Dear Bertrand,

Thank you very much. I was searching some ideas in Leardini as well :-) 

Sarah