You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The issues that need to be addressed regarding the ICD algorithm before submitting the JOSS paper could be listed as follows:
Add dt_data as an optional input to the ICD algorithm as we did in GSD
We need to incorporate "dt_data" into the ICD algorithm to provide real-time information whenever it's available and given as an input to the algorithm. The onset should be updated accordingly.
Since the GSD algorithm requires at least 5 consecutive steps to detect the event as a gait sequence, the algorithm fails to identify the gait sequence for this special case. However, still we can apply ICD to the referred data. The issue should be solved in a way that 'gait_sequences' in https://github.com/neurogeriatricskiel/NGMT/blob/main/ngmt/modules/icd/_paraschiv.py should be optional.
For this purpose, we need to change the algorithm in the following way. When 'gait_sequences' are not provided, the ICD should create artificial gait sequences that cover the active parts of the data and then identify initial contacts within defined limits.
We should try to avoid hard-coding of "acc_vertical = data["LowerBack_ACCEL_x"]" in the initial contact detection algorithm and use some methods to detect x component of the acceleration and then use it inside the algorithm to detect initial contacts.
The text was updated successfully, but these errors were encountered:
I have completed tasks 2 and 3. @JuliusWelzel could you transfer the "tracking_systems" and "tracked_points" algorithms for the GSD and ICD to the detect method as optional arguments??
Dear all,
The issues that need to be addressed regarding the ICD algorithm before submitting the JOSS paper could be listed as follows:
For this purpose, we need to change the algorithm in the following way. When 'gait_sequences' are not provided, the ICD should create artificial gait sequences that cover the active parts of the data and then identify initial contacts within defined limits.
The text was updated successfully, but these errors were encountered: