Foot Strike Angle Prediction and Pattern Classification Using LoadsolTM Wearable Sensors: A Comparison of Machine Learning Techniques

Abstract

The foot strike pattern performed during running is an important variable for runners, performance practitioners, and industry specialists. Versatile, wearable sensors may provide foot strike information while encouraging the collection of diverse information during ecological running. The purpose of the current study was to predict foot strike angle and classify foot strike pattern from LoadsolTM wearable pressure insoles using three machine learning techniques (multiple linear regression―MR, conditional inference tree―TREE, and random forest―FRST). Model performance was assessed using three-dimensional kinematics as a ground-truth measure. The prediction-model accuracy was similar for the regression, inference tree, and random forest models (RMSE: MR = 5.16°, TREE = 4.85°, FRST = 3.65°; MAPE: MR = 0.32°, TREE = 0.45°, FRST = 0.33°), though the regression and random forest models boasted lower maximum precision (13.75° and 14.3°, respectively) than the inference tree (19.02°). The classification performance was above 90% for all models (MR = 90.4%, TREE = 93.9%, and FRST = 94.1%). There was an increased tendency to misclassify mid foot strike patterns in all models, which may be improved with the inclusion of more mid foot steps during model training. Ultimately, wearable pressure insoles in combination with simple machine learning techniques can be used to predict and classify a runner’s foot strike with sufficient accuracy.

Estimation of Fine-Grained Foot Strike Patterns with Wearable Smartwatch Devices

“…The experimental results revealed that the FS patterns were accurately recognized, similar to those reported previously [44]. Contrary to previous studies [44,45], that used pressure-measuring wearable insoles, another recent study utilized an accelerometer inside a running shoe to recognize two different landing styles (e.g., RF and FF strikes) while running [46]. The study found that a cross-correlation measure between acceleration signals from different axes could be used as a feature to recognize the foot strike patterns and the study envisioned the use of acceleration signals for foot strike classification while running.…”

“…Similarly, another recent work proposed a foot-striking recognition system based on a pressure-measuring sensor mounted inside a running shoe [45]. The experimental results revealed that the FS patterns were accurately recognized, similar to those reported previously [44]. Contrary to previous studies [44,45], that used pressure-measuring wearable insoles, another recent study utilized an accelerometer inside a running shoe to recognize two different landing styles (e.g., RF and FF strikes) while running [46].…”

“…For example, a recent study proposed a wearable system that explored pressure-sensitive insoles incorporated in the normal running shoes to predict foot-strike patterns while running. Experimental results demonstrated a high recognition rate (over 90%), although precise recognition of the MF strike pattern was found to be more challenging than those of FF and RF strikes [44]. Similarly, another recent work proposed a foot-striking recognition system based on a pressure-measuring sensor mounted inside a running shoe [45].…”

“…Although measuring pressures [44,45] and inertial motions [46] of the foot or leg can directly provide information regarding foot strike patterns, we hypothesized that such information can also be observed in body parts other than the feet or legs, such as the hands, since the human body parts are connected organically; thus, different loading patterns while striking on the ground would result in significant differences in whole-body motions. From this perspective, we utilized a wearable smartwatch as a sensing device to observe and recognize the hand motion affected by foot strike patterns.…”

Deep Learning for Walking Behaviour Detection in Elderly People Using Smart Footwear

“…Thanks to the fundamentals related to information theory, the miniaturisation of sensors and the improvement of data storage and transmission systems have been possible and is one of the reasons for the success of monitoring and pattern detection through IoT devices and sensors, particularly in the integration of fabrics and textiles (“smart fabrics/wearable”) [ 27 , 28 , 29 , 30 ]. Particularly, the data retrieved by sensors can be used to monitor the elderly in real time and predict their behaviour, preventing potential health problems, while providing them with independence and facilitating them urban living.…”