Search for: All records

Traditional gait event detection methods for heel strike and toe-off utilize thresholding with ground reaction force (GRF) or kinematic data, while recent methods tend to use neural networks. However, when subjects’ walking behaviors are significantly altered by an assistive walking device, these detection methods tend to fail. Therefore, this paper introduces a new long short-term memory (LSTM)-based model for detecting gait events in subjects walking with a pair of custom ankle exoskeletons. This new model was developed by multiplying the weighted output of two LSTM models, one with GRF data as the input and one with heel marker height as input. The gait events were found using peak detection on the final model output. Compared to other machine learning algorithms, which use roughly 8:1 training-to-testing data ratio, this new model required only a 1:79 training-to-testing data ratio. The algorithm successfully detected over 98% of events within 16ms of manually identified events, which is greater than the 65% to 98% detection rate of previous LSTM algorithms. The high robustness and low training requirements of the model makes it an excellent tool for automated gait event detection for both exoskeleton-assisted and unassisted walking of healthy human subjects. 

Over the course of the physical activity transition, machines have largely replaced skeletal muscle as the source of work for locomotion and other forms of occupational physical activity in industrial environments. To better characterize this transition and its effect on back muscles and the spine, we tested to what extent typical occupational activities of rural subsistence farmers demand higher magnitudes and increased variability of back muscle activity and spinal loading compared to occupational activities of urban office workers in Rwanda, and whether these differences were associated with back muscle endurance, the dominant risk factor for back pain. Using electromyography, inertial measurement units, and OpenSim musculoskeletal modeling, we measured back muscle activity and spinal loading continuously while participants performed occupational activities for one hour. We measured back muscle endurance using electromyography median frequency analysis. During occupational work, subsistence farmers activate their back muscles and load their spines at 390% higher magnitudes and with 193% greater variability than office workers. Partial correlations accounting for body mass show magnitude and variability response variables are positively associated with back muscle endurance (R= 0.39–0.90 [P< 0.001–0.210] andR= 0.54–0.72 [P= 0.007–0.071], respectively). Body mass is negatively correlated with back muscle endurance (R= -0.60,P= 0.031), suggesting higher back muscle endurance may be also partly attributable to having lower body mass. Because higher back muscle endurance is a major factor that prevents back pain, these results reinforce evidence that under-activating back muscles and under-loading spines at work increases vulnerability to back pain and may be an evolutionary mismatch. As sedentary occupations become more common, there is a need to study the extent to which occupational and leisure time physical activities that increase back muscle endurance helps prevent back pain.