A polynomial function of gait performance

A mathematical data processing method is presented
that represents a further step in gait analysis. The proposed
polynomial regression analysis is reliable in assessing
differences in the same patient and even on
the same day. The program also calculates the confidence
interval of the whole curve. The procedure was
applied to normal subjects in order to collect normative
data. When a new subject is tested, the polynomial
function obtained is graphically superimposed on
control data. Should the new curve fall within the limits
described by normative data, it is considered to be
equivalent. The procedure can be applied to the same
subject, either normal or pathological, for retesting
kinematic characteristics. The gait cycle is analyzed
as a whole, not point-by-point. Ten normal subjects
and two patients, one with recent- and the other with
late-onset hemiplegia, were tested. Multiple baseline
evaluation is recommended before the start of a rehabilitation
program.