Towards validation and standardization of automatic gait event identification algorithms for use in paediatric pathological populations

To analyse and interpret gait patterns in pathological paediatric populations, accurate determination of the timing specific events (e.g. initial contract – IC, or toe-off TO) is essential. As currently used clinical identification methods are generally subjective, time-consuming, limited to steps with force platform data, several techniques have been proposed based on processing marker kinematics. However, until now, validation standardization these for use diverse remains lacking. 1) What accuracy available kinematics-based algorithms determining IC TO signatures? 2) Does automatic affect interpretation spatio-temporal parameters?. 3D kinematic kinetic data 90 children were retrospectively analysed from a database. Participants classified into 3 categories: group A (toe-walkers), B (flat IC) C (heel IC). Five (one modified) implemented two different foot configurations both compared (visual force-plate) using Bland-Altman analysis. The best-performing algorithm-marker configuration was compute parameters (STP) all trials. establish whether error associated this would interpretation, bias limits agreement determined against inter-trial variability established visual identification. Sagittal velocity heel (Group C) toe B) most reliable indicator while sagittal hallux performed best TO. Biases walking speed, stride time length within respective values. Automatic dependent configuration, results obtained when optimized towards patterns. Our suggest that correct selection event detection approach will ensure misinterpretation STPs avoided.