Oral Presentations / Gait & Posture 24S (2006) S7–S97 S83 O-51 A comparison of methods for using center of pressure progression in the classification of foot deformity Gene Jameson ∗ , Jason Anderson, Roy Davis, Jon Davids, Lisa Christopher Shriners Hospitals for Children, Greenville, SC, USA 1. Introduction The pedobarograph, a system that measures plantar pres- sure, has been successfully applied to patients with neuropa- thy where the information can be used to prevent pressure- related problems [1,2]. In attempts to utilize the pedobaro- graph in the evaluation of children with cerebral palsy and associated foot deformities, previous researchers used nor- malized plantar pressure within different regions of the foot to classify the feet into varus or valgus categories [3,4]. One recent method of pedobarograph analysis used the normal center of pressure progression (COPP) during stance to estab- lish medio-lateral divisions of the foot [5]. The COPP for pathological feet could then be compared to COPP from a normal population. This method required the longitudinal axis of the foot to be subjectively determined by the analyst so the pedobabrograph output could be oriented consistently and comparisons made. The current study expands on that method by introducing simultaneous motion capture to eval- uate the difference between subjective and objective methods of long axis determination. 2. Statement of clinical significance This study reports a new pedobarograph analysis pro- cedure as well as the results for a population of pediatric individuals without gait impairment and compares it to pre- viously published data. This information can then be used in the evaluation and treatment of patients with foot deformity. 3. Methods Five successful walking trials were collected for each foot of 23 normal subjects (mean age 11.4 ± 3.3 years, range of 6–17 years). One representative trial was selected for each foot for COPP analysis. Because the ultimate goal was to apply this technique to feet with pathology, the custom soft- ware developed in our laboratory allows for the analysis of the COPP if the foot shape is atypical and if the entire foot does not make contact with the floor surface. A graphical computer interface allowed the four analysts to specify the long axis of each foot, which was defined as a line connect- ing a point between the second and third metatarsal and the midpoint of the heel. The medial and lateral borders of the foot were also identified. In addition to this analysis, the 3D Fig. 1. Mean COPP. Dark gray band is mean ± 1 S.D.; light gray is mean ± 2 S.D. kinematic data was used to identify foot landmarks consistent with standard gait analysis, i.e. medial and lateral malleoli, and the midline of the foot at the base of the toes. The cen- ter point between the malleoli markers and the toe marker established the long axis of the foot. The length of the foot was divided into three regions lon- gitudinally (Fig. 1). One and two standard deviations from the mean COPP were calculated (Fig. 1). These fore/aft and medio-lateral regions were numbered to facilitate classifica- tion (Fig. 1). The time (% of stance) and location of the COP within the heel, midfoot, and forefoot regions were deter- mined so that the COPP could be described temporally as well as in both anteroposterior and mediolateral directions (Fig. 2). The rotation angles of the long axis as determined by the two methods was compared, as well as the rotation angles between the four observers. 4. Results The average COPP in normal individuals begins in the middle of the heel region at initial contact then progresses for- ward for 23.8% of stance phase. It proceeds forward through