Spectral analysis of standing balance using medical stereo images

Serial 3D measured data on human motion provides important information for the medical evaluation of a person's stability during his daily movements. This paper presents a new approach employing a personal computer-aided system for acquiring digital data about human movement. This study is based on three components: a synchronized hardware system linking two CCD video cameras, a simple 3D videogrammetric and image processing system, named "New Bird Man", for non-metric cameras using bundle adjustment with self-calibration and biomechanics, and spectral analysis utilizing autoregressive modeling for determining the recovery of a patient's standing balance. After therapy, the fluctuation of the patient's COG (center of gravity) was improved and its power spectrum was found to vary in inverse proportion to the frequency, which was uneven and unstable before therapy in patients having lower-limb disorders. This study provides useful information for medical rehabilitation.