A new approach to determine the hip rotation profile from clinical gait analysis data

Conventional models for determining joint rotation angles from marker posit ions as part of three-dimensional clinical gait analysis are susceptible to errors arising from mis-placement of the thigh markers. An analysis of ide alised data reveals how the measured variables are affected by different an gular offsets of the thigh marker from its true position. An artefact on th e varus-valgus signal arising from the projection of true knee flexion onto a mal-aligned thigh segment axis is the most characteristic feature of thi s problem. If this is observed then the hip rotation profiles are also erro neous. A technique is proposed to determine a correction factor which can be appli ed to gait data to correct for this mal-alignment. Its use is demonstrated on a single case study and a subjective assessment of its use on a cohort o f 40 patients is reported. A detailed discussion of the assumptions on whic h the method is founded is included as well as guidelines as to when the te chnique is likely to be successful. It is recommended that the technique is used as an aid to training staff in marker placement rather than as a routine retrospective correction of data for erroneously placed markers. (C) 1999 Elsevier Science B.V. All rights reserved. PsycINFO classification. 2330.