search

u0078867 / ulema-ul-analyzer

U.L.E.M.A. is a MATLAB-based GUI tool (for Windows) for quickly processing motion capture data (C3D files) relative to upper limb movements
18 stars 7 forks source link

Questions about the segmentation, variables and output #16

Closed gabrielalsantos closed 10 years ago

gabrielalsantos commented 10 years ago

Hi Davide and Ellen,

I have a question about the segmentation because my dynamic trials were performed separately. In other words, my dynamics trials are in separate files and each file has full movement. But, I cut part of the movement that I intend to analyze in this files in the Nexus. For this, in accordance with the concept of segmentation, I would like to know if I can use the cut file or if I should mark the star and end of the movement that I intend to analyze in the nexus as events?

Other question is about the variables. I woud like if the ULEMA calculates angular velocities and movements units?

I would know if the program generates a table with the angles of each joint during movement with a column of time?

Thanks!!!

Gabriela.

ejaspers commented 10 years ago

Hi Gabriela,

I'll answer your questions in different "points", so that Davide can add/correct where needed.

  1. You can process different dynamic trials at once, no matter if they contain only 1 movement repetition or if it's a dynamic trial with 10 repetitions. The segmentation method is crucial for the calculation of the "best cycles", which is a part of the APS calculation. You can analyse you dynamic trials without marking the start and end of the movement.
  2. If you wish to calculate the spatiotemporal parameters, I would suggest to define the start and stop of a dynamic task in Nexus (rather than removing data from you c3d files), using method "2" (so GE1 for start and GE2 for stop). The semantics for this can be found in the "EventsConfig.txt" file in the Seg-folder.
  3. On page 8-9 of the UserGuide, you have a list of the spatiotemporal parameters. Angular velocities are not calculated. However, the joint angles as a function of time can be outputted (in a format of choice), so you could easily calculate angular velocities in a second step - outside the U.L.E.M.A. environment. We offer "jerk" as a measure of movement smoothness, as it characterises the average rate of change of acceleration in a movement. However, the data needed for the MU is again outputted, so you could also calculate this parameters outside of U.L.E.M.A.
  4. You can save the calculated joint angles as csv or mat-files, which gives you the angles per measurement of time-unit. Davide, do we output in real-time or is the data already normalised?

Hope this already helps. Cheers Ellen

u0078867 commented 10 years ago

Hello Gabriela,

sorry for my late reply.

1) I agree 2) I agree. Plus, the meaning of EventsConfig.txt is also desribed in the full documentation 3) I agree. It can be easily done both in MATLAB / Excel. If you have problems with this, let us know. 4) in the csv exporte files per trial, you have both raw data, as a table, each columns being an original time frame (time step can be reconstructed using the Camera freq field); and cycles data normalized between 0 and 100%.

Cheers, Davide