Effect of low-pass filtering on the reliability of center of pressure data in static posturography in dogs

Center of pressure (COP) analysis is widely used to assess postural stability in dogs; however, methodological variability—particularly regarding data filtering and measurement duration—limits comparability across studies. This study evaluated the effect of low-pass filtering on the reliability of COP parameters and its potential to reduce measurement duration and sampling requirements. COP data from 16 clinically sound dogs during quiet standing were analyzed using a pressure platform. A previously published dataset was expanded and re-analyzed by applying a 4th-order Butterworth low-pass filter with cutoff frequencies of 6 Hz and 10 Hz. COP parameters were calculated across multiple measurement durations (1–10 s) and sequence combinations. Reliability was assessed using intraclass correlation coefficients (ICC) for inter-observer agreement and test–retest consistency. Inter-observer reliability was excellent for all parameters regardless of filtering. Low-pass filtering improved test–retest reliability, particularly for short measurement durations, and reduced the number of sequences required to achieve the predefined reliability threshold (ICC ≥ 0.60). The 6 Hz filter showed a consistent but modest advantage over 10 Hz, especially for shorter recordings. However, significant differences between measurement durations persisted despite filtering. In conclusion, low-pass filtering enhances the reliability and efficiency of canine COP analysis, enabling shorter and thus more feasible measurement protocols without compromising data consistency. However, measurement duration remains a critical factor, emphasizing the need for standardized protocols.