PDA

View Full Version : summary of software download



tpribanic57
07-01-2005, 02:37 AM
Dear All,

earlier I have posted a question about not being able to download the software accompanying (DOS version) of the book "Dynamics of Human Gait", from ISB software pages. Eventually. the author himself prof. Vaughan was kind to send me those. Also I would like to point out another important issue regarding the versions of the book second copy edition. Somehow, my version of second edition has some mistakes that are corrected in the version that is NOW downloadable from publisher site. It cost me two days to realize that since both second edition versions appear exactly the same on first couple of pages. Since publisher did not bother to point that out maybe biomech web master can put some note next to publisher link for all those that my have second edition copy like I've had, where some obvious mistakes are present.

The second of part of my original question concerned the other sources of similar software. It is rather unfortunate that I have received only one replay (probably due to holiday season), still I am very grateful to Dr. Chris Kirtley for it.

Regards, Tomo

ORIGINAL POST:
I have been trying unsuccessfully to download from ISB software pages. the following
GaitLab software package for MSDOS (ZIP file, 580 kB) (based on the book "Dynamics of Human Gait", C.L. Vaughan et al). The other three files: man, woman and clinical are downloadable with no problem. Thus, I was wondering if someone has done it before and is willing to email me the mentioned software package.
There are supposedly some printing mistakes in the mentioned book (second edition). In case someone has those neatly summarized, or any other criticism concerning the book (Helen Hayes protocol etc) would be very much appreciated.

Also, I would like to take this opportunity to ask you for other similar software sources, free for comparison and evaluation purposes, that calculates kinematic&kinetic data from markers set. Any accompanying document which clearly explains, or at least points to references, how various quantities are calculated (particularly BSP parameters) are welcome too, in order to make comparison easier.

Thank you, the summary will follow.

Tomislav Pribanic, M.Sc., EE
Department for Electronic Systems and Information Processing
Faculty of Electrical Engineering and Computing
3 Unska, 10000 Zagreb, Croatia
tel. ..385 1 612 98 67, fax. ..385 1 612 96 52
E-mail : tomislav.pribanic@fer.hr


REPLAY:

Dear Tomislav,

It ocurred to me some time ago that for teaching purposes it would be
nice to make an Excel spreadsheet for inverse dynamics. After much
trouble and frustration, I nearly have it working:

http://www.univie.ac.at/cga/Gait.xls

Basically it works like this:

1. You put raw marker data in the first sheet, tell it how many frames
and the frame rate, then hit Ctrl-m to normalise the marker data to 1%
time intervals. Not that you need to turn on macros for this (it uses
an algorithm suggested by Li Li some time ago on BIOMCH-L). I have put
in David Winter's data from the back of his book to check that it is
working OK. The reason I time-normalise at the start like this is so
that the spreadsheet can then cope with data of any number of samples,
since all the other worksheets are then 100 lines long.

2. It filters the data using a 2nd order Butterworth filter forwards
then backwards to eleiminate phase shift (this is a useful exercise
for students to see). You can specify the cutoff (e.g. 6 Hz), and I
have used a bit of a fiddle, wrapping round data at the beginning and
end of the cycle to cope with the end points - if anyboy can suggest a
better method I would be very interested to hear it, as it's not
ideal.

3. It then calculates all kinematic data, including derviatives by a
difference equation. Once again, there is a bit of a fiddle at the
endpoints, which results in some discontinuities. These relly need a
bit more work.

4. Ground reaction (force plate) data is entered and normalised by
hitting Ctrl-f. I show the vector to help in visualizing the GRF in
relation to the joint centres.

5. Body segment parameters are calculated from Depster regressions,
and the motion of the centres of masses derived, as well as segment
energies and powers.

6. Reactions are then calculated by inverse dynamics. I needed to
filter again here to remove transients introduced by the end-point
problem mentioned earlier - it would be nice to avoid this. The
moments and powers look pretty god as far as the ankle and knee are
concerned, but for the life of me I can't find out where I've gone
wrong with the hip. Once again, I'd be very grateful to anyone who can
fix it!

I do think Excel is ideal for an exercise such as this because the
student can see all the data - also because Excel can be found on
almost all PCs these days.

Look forward to your feedback and hope you or someone else can fix the
last remaining problems with the sheet!

Chris