Human gait analysis can provide an excellent source for identifying and predicting pathologies and injuries. In this respect, sensorized insoles also have a great potential for extracting gait information. This, combined with mathematical techniques based on machine learning (ML), can potentialize biomechanical analyses. The present study proposes a proof-of-concept of a system based on vertical ground reaction force (vGRF) acquisition with a sensorized insole that uses an ML algorithm to identify different patterns of vGRF and extract biomechanical characteristics that can help during clinical evaluation. The acquired data from the system was clustered by an immunological algorithm (IA) based on vGRF during gait. These clusters underwent a data mining process using the classification and regression tree algorithm (CART), where the main characteristics of each group were extracted, and some rules for gait classification were created. As a result, the system proposed was able to collect and process the biomechanical behavior of gait. After the application of IA and CART algorithms, six groups were found. The characteristics of each of these groups were extracted and verified the capability of the system to collect and process the biomechanical behavior of gait, offering verification points that can help focus during a clinical evaluation.

Sensorized insoles (SIs) have been used as a wearable instrument to study human gait and have the potential to identify and predict pathologies and injuries. However, most of these sensorized insoles are only statically calibrated, relying on a scale and known weights to establish a relationship between electrical signals and the load applied on laboratory benches while ignoring the dynamic interaction between person and instrument. This study proposes and verifies a calibration method complementary to static calibration to compensate for different dynamic interactions between the insole and the individual during gait. In order to perform this comparison, a laboratory test was proposed with 32 volunteers (18 men and 14 women). Each volunteer walked on a double-belt instrumented treadmill (Bertec at 1000 Hz, Bertec Corp, Columbus, OH) while wearing an experimental resistive sensorized insole (SI). The SI data were compared with the instrumented treadmill and adjusted using an optimization algorithm to create a dynamic coefficient to complement and optimize the results. This study also verifies the impact of the method considering three different types of gait: pronated, neutral, and supinated. After using this technique and considering static calibration, the Pearson correlation coefficient between the SI and the instrumented treadmill improved by 12%.

Context : Handheld dynamometry is common for isometric strength testing, but the high cost and import barriers of commercial devices (eg, MicroFET II) limit access. Three-dimensional (3D) printing enables low-cost, customizable instruments. We developed the 3D-printed PhysioCode and examined its measurement properties for knee strength. Design : Methods and Technical Report evaluating concurrent validity and test–retest reliability versus a reference handheld dynamometer. Methods : Forty-nine healthy adults (23.2 [3.9] y; body mass index 21.9 [2.8] kg/m ² ) performed 3 isometric trials of dominant-limb knee extension and flexion with both devices in randomized order, assessed by 2 trained novice raters under blinded analysis. Concurrent validity was evaluated using Pearson correlation coefficients ( r ) and root mean square error, and agreement between devices was assessed using Bland–Altman analyses. Reliability was assessed using intraclass correlation coefficients (ICC[2,1] and ICC[2, k ]), along with the standard error of measurement and minimal detectable change at 95% confidence. We compared single and repeated test strategies. Results : All mean values differed by ≤21 N between devices across knee extensors and flexors. Concurrent validity was strong–very strong ( r = .88–.95; P < .001), with lower root mean square error for averaged repetitions. Bland–Altman showed small, direction-specific bias: PhysioCode slightly overestimated extensor force values (−17 to 21 N) and underestimated flexor force values (−12 N). Reliability was excellent: PhysioCode ICC(2,1) = .93 to .95 and ICC(2, k ) = .98 to .98; MicroFET II ICC(2,1) = .89 to .94 and ICC(2, k ) = .96 to .98. Averaging 2 to 3 trials reduced standard error of measurement and minimal detectable change at 95% confidence relative to single trials or maxima. Conclusions : The 3D-printed PhysioCode showed adequate concurrent validity and excellent test–retest reliability for knee extensor/flexor strength, with performance comparable to the MicroFET II in this sample of young, healthy adults. Averaging 2 or 3 repetitions improved precision. These findings suggest that PhysioCode may represent an option for strength assessment, warranting further evaluation in target clinical populations.