Using Machine Learning to Identify Feelings of Energy and Fatigue in Single-Task Walking Gait: An Exploratory Study

Abstract

The objective of this study was to use machine learning to identify feelings of energy and fatigue using single-task walking gait. Participants (n = 126) were recruited from a university community and completed a single protocol where current feelings of energy and fatigue were measured using the Profile of Moods Survey–Short Form approximately 2 min prior to participants completing a two-minute walk around a 6 m track wearing APDM mobility monitors. Gait parameters for upper and lower extremity, neck, lumbar and trunk movement were collected. Gradient boosting classifiers were the most accurate classifiers for both feelings of energy (74.3%) and fatigue (74.2%) and Random Forest Regressors were the most accurate regressors for both energy (0.005) and fatigue (0.007). ANCOVA analyses of gait parameters comparing individuals who were high or low energy or fatigue suggest that individuals who are low energy have significantly greater errors in walking gait compared to those who are high energy. Individuals who are high fatigue have more symmetrical gait patterns and have trouble turning when compared to their low fatigue counterparts. Furthermore, these findings support the need to assess energy and fatigue as two distinct unipolar moods as the signals used by the algorithms were unique to each mood.

Identifying Individuals Who Currently Report Feelings of Anxiety Using Walking Gait and Quiet Balance: An Exploratory Study Using Machine Learning

“…Taken together, this suggests that anxious individuals may be trying to find a larger base of support when turning (wider turns), while decreasing the amount of time spent in unstable environments (faster turns, as turns, are less stable than straight walking). These findings are similar to those reported among individuals with gait disorders who try to avoid falling [49], as well as individuals who report feeling low energy [50,51]. A visual representation of the gait differences can be found at https://gaitsim.dmanserver.com/ AnxietyYN (accessed on 12 January 2022).…”

Detecting Psychological Interventions Using Bilateral Electromyographic Wearable Sensors

“…The dual-task paradigm for modifying gait and muscle activity pattern focuses on disrupting various motor and cognitive processes, including attention to the environment and neurobiomechanical control of the trunk and limbs during locomotion [ 13 , 14 , 15 ]. These findings [ 13 , 14 , 15 ], combined with the works of various scholars, on the effects of various moods, such as anxiety [ 16 , 17 ], depression [ 18 , 19 , 20 ], anger [ 21 ], and fatigue [ 22 , 23 , 24 , 25 ], suggest that psychological conditions can disrupt motor coordination patterns.…”

Trait Energy and Fatigue Influence Inter-Individual Mood and Neurocognitive Responses during Work Done While Sitting, Standing, and Intermittent Walking: A Randomized-Controlled Crossover Design

“…Further, studies that have examined state level energy and fatigue suggest that dopamine [31], Annexin A1 [32], and peripheral mitochondrial function [33] are associated with changes in feelings of energy, while histamine [34], serotonin [31], and TNF-α [31] are associated with feelings of fatigue. In addition to biological differences, evidence also exists that both trait and state level feelings of energy and fatigue are associated with different aspects of gait and balance, suggesting that these moods may by associated with different aspects of the frontal cortex [35][36][37][38][39]. Recently, Filippi and colleagues [40] provided evidence on the importance of measuring trait, prolonged state, and state level energy and fatigue, as trait may uniquely influence prolonged state and state level energy and fatigue.…”