Margin of stability (MOS) is considered a measure of mechanical gait stability. Due to broad application of treadmills in gait assessment experiments, we aimed to determine if walking on a treadmill vs. overground would affect MOS during three speed-matched conditions. Eight healthy young participants walked on a treadmill and overground at Slow, Preferred, and Fast speed-matched conditions. The mean and variability (standard deviation) of the MOS in anterior-posterior and mediolateral directions at heel contact were calculated. Anterior-posterior and mediolateral mean MOS values decreased with increased speed for both overground and treadmill; although mediolateral mean MOS was always wider on the treadmill compared to overground. Due to lack of optic flow and different proprioceptive inputs during treadmill walking, subjects may employ strategies to increase their lateral stability on treadmill compared to overground. Anterior-posterior MOS variability increased with speed overground, while it did not change on treadmill, which might be due to the fixed speed of treadmill. Whereas, lateral variability on both treadmill and overground was U-shaped. Walking at preferred speed was less variable (may be interpreted as more stable) laterally, compared to fast and slow speeds. Caution should be given when interpreting MOS between modes and speeds of walking. As sagittal plane walking is functionally unstable, this raises the consideration as to the meaningfulness of using MOS as a global measure of gait stability in this direction.

Peripheral artery disease (PAD) manifests from atherosclerosis, which limits blood flow to the legs and causes changes in muscle structure and function, and in gait performance. PAD is underdiagnosed, which delays treatment and worsens clinical outcomes. To overcome this challenge, the purpose of this study is to develop machine learning (ML) models that distinguish individuals with and without PAD. This is the first step to using ML to identify those with PAD risk early. We built ML models based on previously acquired overground walking biomechanics data from patients with PAD and healthy controls. Gait signatures were characterized using ankle, knee, and hip joint angles, torques, and powers, as well as ground reaction forces (GRF). ML was able to classify those with and without PAD using Neural Networks or Random Forest algorithms with 89% accuracy (0.64 Matthew’s Correlation Coefficient) using all laboratory-based gait variables. Moreover, models using only GRF variables provided up to 87% accuracy (0.64 Matthew’s Correlation Coefficient). These results indicate that ML models can classify those with and without PAD using gait signatures with acceptable performance. Results also show that an ML gait signature model that uses GRF features delivers the most informative data for PAD classification.

A healthy respiratory system has variability from breath-to-breath and patients with COPD (PwCOPD) have abnormal variability in breath cycles. The aim of this study was to determine if interbreath-interval and tidal-volume variability, and airflow regularity change as metabolic demands increase (seated, standing, and walking) in PwCOPD as compared to controls. Sixteen PwCOPD (64.3 ± 7.9 yr, 61.3 ± 44.1% FEV₁%predicted) and 21 controls (60.2 ± 6.8 yr, 97.5 ± 16.8% FEV₁%predicted) sat, stood, and walked at their preferred-pace for five-minutes each while breathing patterns were recorded. The mean, standard deviation, and coefficient of variation of interbreath-intervals and tidal-volume, and the regularity (sample entropy) of airflow were quantified. Results were subjected to ANOVA analysis. Interbreath-interval means were shorter in PwCOPD compared to controls (p = 0.04) and as metabolic demand increased (p < 0.0001), standard deviation was decreased in PwCOPD compared to controls during each condition (p’s < 0.002). Mean tidal-volume did decrease as metabolic demand increased across groups (p < 0.0001). Coefficient of variation findings (p = 0.002) indicated PwCOPD decline in tidal-volume variability from sitting to standing to walking; whereas, controls do not. There was an interaction for airflow (p = 0.02) indicating that although, PwCOPD had a more regular airflow across all conditions, control’s airflow became more irregular as metabolic demand increased. PwCOPD’s airflow was always more regular compared to controls (p = 0.006); although, airflow became more irregular as metabolic demand increased (p < 0.0001). Healthy respiratory systems have variability and irregularity from breath-to-breath decreases with adaptation to demand. PwCOPD have more regular and restricted breathing pattern that may affect their ability to adjust in demanding situations.

Understanding metabolic cost through biomechanical data, including ground reaction forces (GRFs) and joint moments, is vital for health, sports, and rehabilitation. The long stabilization time (2–5 min) of indirect calorimetry poses challenges in prolonged tests. This study investigated using artificial neural networks (ANNs) to predict metabolic costs from the GRF and joint moment time series. Data from 20 participants collected over 270 walking trials, including the GRF and joint moments, formed a detailed dataset. Two ANN models were crafted, net_GRF for the GRF and net_Moment for joint moments, and both underwent training, validation, and testing to validate their predictive accuracy for metabolic cost. Net_GRF (six hidden layers, two input delays) showed significant correlations: 0.963 (training), 0.927 (validation), 0.883 (testing), p < 0.001. Net_Moment (three hidden layers, one input delay) had correlations of 0.920 (training), 0.956 (validation), 0.874 (testing), p < 0.001. The models’ low mean squared errors reflect their precision. Using Partial Dependence Plots, we demonstrated how gait cycle phases affect metabolic cost predictions, pinpointing key phases. Our findings show that the GRF and joint moments data can accurately predict metabolic costs via ANN models, with net_GRF being notably consistent. This emphasizes ANNs’ role in biomechanics as a crucial method for estimating metabolic costs, impacting sports science, rehabilitation, assistive technology development, and fostering personalized advancements.