Tweet
Good morning,
I have seen a number of excellent discussions regarding the use
of filtering and smoothing of kinematic and other data in recent time.
However a problem which we face in our lab, and I assume is faced by
many others, is the synchronisation on kinematic and kinetic data. My
question is, if kinematic data is collected at, say 100Hz whilst
kinetics data is collected at, say 1000Hz, what is the most appropriated
fitting method to employ. One choice may be to apply a linear
interpolation to the kinematic data to allow the determination of
kinematics at other points. Another may be to apply data reduction to
the kinematics and reduce everything to the lowest sampling rate.
Another alternative put forward by some of our IT people is to use a
graphics morphing package to create a transformation based on nearby
data points (trends) to interpolate kinematic points. It appears few
papers which collect simultaneous kinematic and kinetic data collected
at different rates report the methods by which interpolation or
reduction of data is obtained. Is there a consensus of opinion among
biomechanists worldwide?
Kind Regards,
Rob
Robert Stanton BHMSc (Hons) MAAESS AEP
School of Health and Human Performance
Central Queensland University
Bruce Highway
Rockhampton, QLD, 4702
Phone: +61 07 49309763
Fax: +61 07 49309871
I have seen a number of excellent discussions regarding the use
of filtering and smoothing of kinematic and other data in recent time.
However a problem which we face in our lab, and I assume is faced by
many others, is the synchronisation on kinematic and kinetic data. My
question is, if kinematic data is collected at, say 100Hz whilst
kinetics data is collected at, say 1000Hz, what is the most appropriated
fitting method to employ. One choice may be to apply a linear
interpolation to the kinematic data to allow the determination of
kinematics at other points. Another may be to apply data reduction to
the kinematics and reduce everything to the lowest sampling rate.
Another alternative put forward by some of our IT people is to use a
graphics morphing package to create a transformation based on nearby
data points (trends) to interpolate kinematic points. It appears few
papers which collect simultaneous kinematic and kinetic data collected
at different rates report the methods by which interpolation or
reduction of data is obtained. Is there a consensus of opinion among
biomechanists worldwide?
Kind Regards,
Rob
Robert Stanton BHMSc (Hons) MAAESS AEP
School of Health and Human Performance
Central Queensland University
Bruce Highway
Rockhampton, QLD, 4702
Phone: +61 07 49309763
Fax: +61 07 49309871