Classifying Toe Walking Gait Patterns Among Children Diagnosed With Idiopathic Toe Walking Using Wearable Sensors and Machine Learning Algorithms

Abstract

Idiopathic toe walking (ITW) is a gait abnormality in which children toe touch at initial contact and demonstrate limited or no heel contact throughout the gait cycle. Toe walking results in poor balance, increased risk of falling, and developmental delays among children. Identifying toe walking steps during walking can facilitate targeted intervention among children diagnosed with ITW. With recent advances in wearable sensing, communication technologies, and machine learning, new avenues of managing toe walking behavior among children are feasible. In this study, we investigate the capabilities of Machine Learning (ML) algorithms in identifying initial foot contact (heel strike versus toe strike) utilizing wearable body sensors. Thirty-six children (Age 9.4±2.8 years) diagnosed with ITW participated in this study. Six ML algorithms, consisting of Support Vector Machines (SVM), decision tree (DT), random forest (RF), K-nearest neighbors (KNN), Multi-layer Perceptron (MLP), and Gaussian process (GP), could successfully classify initial contact walking patterns among ITW. We found that a simple KNN algorithm resulted in the highest accuracy of 92.92% and F1-score of 93.20% to differentiate toe walking gait versus best heel strike when using all four body sensors. We also found that toe walking resulted in higher variability in the sacral vertical accelerations among children diagnosed with ITW. Accurate quantification of toe walking steps in clinical applications is critical for assessing rehabilitation progress and designing new interventions for children diagnosed with ITW.

The Application of Wearable Sensors and Machine Learning Algorithms in Rehabilitation Training: A Systematic Review

“…Among them, the four movements of shoulder abduction, shoulder flexion, wall sliding, and wall pressing were performed in a sitting position, the purpose of which was to reduce the compensatory movements of the other muscles during the rehabilitation process. Soangra et al [ 60 ] focused on children’s idiopathic toe walking (ITW), reducing the size of the sensor and wearing it directly on the upper body. Not only did this not limit the walking rehabilitation movement, but it also helped parents monitor the child’s walking status in real time and presented abnormal gait occurrence.…”

Classifying Unstable and Stable Walking Patterns Using Electroencephalography Signals and Machine Learning Algorithms

“…The overall data segmentation into gait cycles is summarized in Figure 4. The EEG data segmentation into gait cycles and procedures was inspired by similar signal analysis approaches [17,18]. EEG Feature extraction during gait cycles: One common approach for analyzing EEG signals is to extract features that can be used to classify different states or activities of the brain during different walking conditions.…”

“…The overall data segmentation into gait cycles is summarized in Figure 4. The EEG data segmentation into gait cycles and procedures was inspired by similar signal analysis approaches [17,18].…”

Dense & Attention Convolutional Neural Networks for Toe Walking Recognition

“…The severity of ITW can vary from landing on the middle foot during the standing phase to loading only on the metatarsal head [2], and the prevalence of toe walking is evaluated to affect up to 5% of normal children [3]. Persistent ITW without treatment may cause an increased risk of falling or tripping [4], leg pain [5], injured muscles and motor coordination [6], and organic anomalies [7]. Early identifying toe walking in clinical diagnosis can facilitate timely intervention and treatment.…”

“…This depends entirely on the experience of the specialist and has a certain degree of subjectivity. Besides, rigorous laboratory-based gait analysis protocol requires special laboratory apparatus like an instrumented walkway, infrared camera-based motion capture system, or treadmill with an integrated force plate [4]. Such laboratory equipment is costly and restrictive, requiring specialized personnel to manipulate and analyze gait data.…”