Results 1 to 8 of 8

Thread: Visual3d Gait Event Detection

Hybrid View

Previous Post Previous Post   Next Post Next Post
  1. #1

    Question Visual3d Gait Event Detection

    Hello all
    I'm having a problem with Visual3d automatic Gait Event Detection. I'm using it for couple of force-plate treadmill running trials, and as you can see in the CMO attached file some of detected events are not exactly correct. How I can solve this problem?
    Thanks in advance

    https://drive.google.com/file/d/1EzC...ew?usp=sharing

  2. #2
    Join Date
    Jan 2011
    Location
    Motion Analysis Research Center, Samuel Merritt University, 400 Hawthorne Ave Oakland, CA 94609, USA
    Posts
    107

    Cool Re: Visual3d Gait Event Detection

    Two possibilities: (1) you may need to increase the FP threshold from 20N to 40N for instrumented treadmill measurements; and/or (2) you're using a pre-version 5 version of V3D, so you might want to upgrade since the Automatic_Gait_Events command has been changed slightly.
    Cheers,

    Drew Smith, PhD

    Professor
    Director, Motion Analysis Research Center (MARC)
    https://www.samuelmerritt.edu/marc
    Samuel Merritt University
    400 Hawthorne Avenue
    Suite 101
    Oakland, CA 94609, USA
    Office: (510) 879-9242
    Lab: (510) 879-0337
    FAX: (510) 457-2620
    Email: asmith3@SamuelMerritt.edu

    LinkedIn: http://bit.ly/AWS_li
    Google Scholar: http://bit.ly/AWS_gs
    Researchgate: http://bit.ly/AWS_rg
    ResearcherID: http://bit.ly/AWS_rid

  3. #3

    Re: Visual3d Gait Event Detection

    Dear Drew
    I'm using the last update of the software, and by applying your suggestion (increase the FP threshold from 20N to 40N) nothing changed. I have another question: Is it possible to export reports (for example gait cycle normalized joints moment or power) into a spreadsheet for each trial or every detected cycle Separatly using Visual3D?

  4. #4
    Join Date
    Jan 2011
    Location
    Motion Analysis Research Center, Samuel Merritt University, 400 Hawthorne Ave Oakland, CA 94609, USA
    Posts
    107

    Cool Re: Visual3d Gait Event Detection

    Quote Originally Posted by Ali1226 View Post
    Dear Drew
    I'm using the last update of the software, and by applying your suggestion (increase the FP threshold from 20N to 40N) nothing changed. I have another question: Is it possible to export reports (for example gait cycle normalized joints moment or power) into a spreadsheet for each trial or every detected cycle Separatly using Visual3D?
    The reason why I thought your V3D may be an older version can be seen here: http://www.c-motion.com/v3dwiki/inde...ic_Gait_Events. Your file only had LHS, LTO, etc but the latest version will create LON, LOFF, etc. when there are FP data.

    You need to use the Export_Data_To_ASCII command: http://www.c-motion.com/v3dwiki/inde..._To_ASCII_File. You can opt for Mean and SD in addition to each cycle which can be time-normalized or not.
    Cheers,

    Drew Smith, PhD

    Professor
    Director, Motion Analysis Research Center (MARC)
    https://www.samuelmerritt.edu/marc
    Samuel Merritt University
    400 Hawthorne Avenue
    Suite 101
    Oakland, CA 94609, USA
    Office: (510) 879-9242
    Lab: (510) 879-0337
    FAX: (510) 457-2620
    Email: asmith3@SamuelMerritt.edu

    LinkedIn: http://bit.ly/AWS_li
    Google Scholar: http://bit.ly/AWS_gs
    Researchgate: http://bit.ly/AWS_rg
    ResearcherID: http://bit.ly/AWS_rid

  5. #5

    Re: Visual3d Gait Event Detection

    Ali,

    Just a quick note from me as the moderator. You posted a CMO file, but not everyone can read those. If you can illustrate your problem with an image or screenshot, you may get more replies. There is a lot of expertise on event detection among Biomch-L subscribers, but most of them are not able to see what problem you are having.

    Ton van den Bogert, Biomch-L co-moderator

  6. #6
    Join Date
    Jan 2011
    Location
    India
    Posts
    78

    Re: Visual3d Gait Event Detection

    Logically, for automatic Gait event detection algorithm to work correctly, you need to have dual belt Instrumented Treadmill, with one Force Plate for each foot. It is true that for running trials, double limb support phase duration in a cycle is very short (20% or even less for higher speeds of running), but that is where errors in Gait Detection appear to be happening.
    Visual3D also provides easy modification in Gait Events, manually.
    If data is indeed from dual belt, Force Platform definition does not indicate that.
    It is better to define Instrumented treadmills as Force Structure, even if data does not involve any inclination changes.
    Last edited by msingh27; 05-18-2018 at 03:37 AM.

  7. #7

    Re: Visual3d Gait Event Detection

    Quote Originally Posted by bogert View Post
    Ali,

    Just a quick note from me as the moderator. You posted a CMO file, but not everyone can read those. If you can illustrate your problem with an image or screenshot, you may get more replies. There is a lot of expertise on event detection among Biomch-L subscribers, but most of them are not able to see what problem you are having.

    Ton van den Bogert, Biomch-L co-moderator

    Dear all
    I'm using a set of treadmill running data. After applying Gait_Events_Detection function from pipeline, for some of trials it dose not work correctly. I'm send you a picture of sagittal ankle joint power. As you can see heel strike did not detect correctly. as suggested, I'm used different tricks to solve this problem like filtering the force plate data before event detection, set minimum force plate threshold to 40N but nothing changes. The sampling frequency has been 300 Hz, could this be the reason for this problem?

    Untitled.jpg
    Last edited by Ali1226; 05-19-2018 at 03:52 PM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •