Quantitative gait analysis requires attachment of optical targets to the skin's surface. Data derived from monitoring the locations of these targets while a patient walks provide the basis for an objective study of the movement. The movement of the skin and underlying soft tissues relative to the bone is known to be a source of error in the measurements, but the magnitude of the errors is unknown and goes uncorrected in standard analyses. The purpose of this study is to evaluate the magnitude and pattern of error produced by tracking the skeletal system using surface mounted targets. As a subject walks in the laboratory, a video-based target tracking system is used to measure simultaneously the movement of two sets of reflective targets attached to each body segment being studied -- one set mounted to the surface (standard technique) and one mounted to the bone. Skeletal fixation of the bone-mounted targets is accomplished using a percutaneous skeletal tracker (PST). Each target set is used separately in calculations of segment motions (kinematics) and the net joint forces and moments (inverse dynamics analysis) that cause the motions. The differences in results represent the errors introduced by the surface target attachment technique. Inverse dynamics analyses were performed on data from three subjects who walked with the PST attached to the shank (lower leg). The magnitudes of the differences in forces and moments would not likely affect the clinical interpretation of the data, but may be important for other applications such as musculoskeletal modeling. Additional data were also collected to measure the motion with respect to bone of a single surface mounted target on the lateral aspect of the knee. The motion of this target (used by some laboratories to estimate the location of the knee joint center) was compared with the joint center estimate derived from the set of surface mounted targets at the mid-shank. The movement of the single lateral knee target with respect to bone during gait was more than twice that of the knee joint center estimated from the surface mounted targets. The PST is currently undergoing modification and cadaveric testing to establish a design and attachment technique for use on the femur.
