Test-retest reliability of multiscale fractal dimension measurements of plantar pressure maps during dynamic tasks

The tools used to evaluate foot types are divergent since they adopt classic linear analyzes, based on anthropometric or image measurements, which do not dynamically contemplate the variability of foot shape. The use of newer techniques such as multiscale fractal dimension (MFD) may be a key to this type of problem. However, for these measures to be used safely and consistently, it is essential to evaluate their reliability. The aim of this study was to evaluate the test-retest reliability of MFD measurements of adult plantar pressure maps during gait, as well as the standard error of measurement (SEM), and minimal detectable change (MDC90). Seventy-two subjects were included in the test-retest, with a one week interval. The plantar pressure maps were constructed using a pressure platform. The data were processed in a routine for extracting the MFD curve measurements (maximum and integral values). The Intraclass Correlation Coefficient results (ICC3,k) were excellent for both measurements (maximum value 0.96, 95% confidence interval [0.93–0.97], and integral 0.95 [0.92–0.97]) with low SEM and MDC90 values below 10% of the mean. The application of MFD to the plantar pressure data generated by the pressure platform is reliable and could allow exploration of the complexity of foot shapes, enabling their classification.