Physical fatigue is a recurrent problem in running that negatively affects performance and leads to an increased risk of being injured. Identification and management of fatigue helps reducing such negative effects, but is presently commonly based on subjective fatigue measurements. Inertial sensors can record movement data continuously, allowing recording for long durations and extensive amounts of data. Here we aimed to assess if inertial measurement units (IMUs) can be used to distinguish between fatigue levels during an outdoor run with a machine learning classification algorithm trained on IMU-derived biomechanical features, and what is the optimal configuration to do so. Eight runners ran 13 laps of 400 m on an athletic track at a constant speed with 8 IMUs attached to their body (feet, tibias, thighs, pelvis, and sternum). Three segments were extracted from the run: laps 2–4 (no fatigue condition, Rating of Perceived Exertion (RPE) = 6.0 ± 0.0); laps 8–10 (mild fatigue condition, RPE = 11.7 ± 2.0); laps 11–13 (heavy fatigue condition, RPE = 14.2 ± 3.0), run directly after a fatiguing protocol (progressive increase of speed until RPE ≥ 16) that followed lap 10. A random forest classification algorithm was trained with selected features from the 400 m moving average of the IMU-derived accelerations, angular velocities, and joint angles. A leave-one-subject-out cross validation was performed to assess the optimal combination of IMU locations to detect fatigue and selected sensor configurations were considered. The left tibia was the most recurrent sensor location, resulting in accuracies ranging between 0.761 (single left tibia location) and 0.905 (all IMU locations). These findings contribute toward a balanced choice between higher accuracy and lower intrusiveness in the development of IMU-based fatigue detection devices in running.

Physical exercise (PE) is beneficial for both physical and psychological health aspects. However, excessive training can lead to physical fatigue and an increased risk of lower limb injuries. In order to tailor training loads and durations to the needs and capacities of an individual, physical fatigue must be estimated. Different measurement devices and techniques (i.e., ergospirometers, electromyography, and motion capture systems) can be used to identify physical fatigue. The field of biomechanics has succeeded in capturing changes in human movement with optical systems, as well as with accelerometers or inertial measurement units (IMUs), the latter being more user-friendly and adaptable to real-world scenarios due to its wearable nature. There is, however, still a lack of consensus regarding the possibility of using biomechanical parameters measured with accelerometers to identify physical fatigue states in PE. Nowadays, the field of biomechanics is beginning to open towards the possibility of identifying fatigue state using machine learning algorithms. Here, we selected and summarized accelerometer-based articles that either (a) performed analyses of biomechanical parameters that change due to fatigue in the lower limbs or (b) performed fatigue identification based on features including biomechanical parameters. We performed a systematic literature search and analysed 39 articles on running, jumping, walking, stair climbing, and other gym exercises. Peak tibial and sacral acceleration were the most common measured variables and were found to significantly increase with fatigue (respectively, in 6/13 running articles and 2/4 jumping articles). Fatigue classification was performed with an accuracy between 78% and 96% and Pearson’s correlation with an RPE (rate of perceived exertion) between r = 0.79 and r = 0.95. We recommend future effort toward the standardization of fatigue protocols and methods across articles in order to generalize fatigue identification results and increase the use of accelerometers to quantify physical fatigue in PE.

The surge of wearable device technology has enabled out-of-the-lab collection of running gait data for commercial, clinical, and coaching applications. However, low sampling frequencies interfere with measuring peak acceleration magnitudes accurately, and the ability to track relative changes during a prolonged run with lower sampling devices is unknown. The purposes of this study were to compare peak resultant acceleration measured simultaneously at different sampling frequencies and evaluate if different sampling frequencies could track similar relative changes in peak acceleration over a 20-min treadmill run. Seventeen participants ran on a treadmill at a self-selected, “easy” pace for 20-min (mean ± SD = 2.6 ± 0.4 m s ⁻¹ ). Three research-grade, triaxial accelerometers (“HiRes” = 1200 Hz, “MedRes” = 462 Hz, and “LoRes” = 100 Hz) were secured to each of three anatomical locations (tibia, low-back, forehead). Mean peak resultant accelerations from each device during minutes 3 to 18 were compared within each location (linear mixed model, α = 0.050). No significant device by timepoint interaction was observed ( p > 0.999). A significant main effect of sampling frequency at all three locations (HiRes > MedRes > LoRes; p < 0.001) confirmed the underestimation of low sampling frequencies on peak resultant acceleration. However, the significant main effect of time indicated that peak resultant acceleration changed similarly over time between sampling frequencies at the tibia ( p = 0.010) and head ( p = 0.002), but not the low-back ( p = 0.318). Downsampling HiRes to 400 and 100 Hz reduced the underestimation of the resultant peaks within the MedRes and LoRes signals by <7.7% across anatomical locations. This study confirms sampling frequency of wearable devices significantly affects peak resultant acceleration and demonstrates these effects are greater for signals captured at lower sampling frequencies and caudal locations. Despite these effects, this study cautiously supports the use of ≥100 Hz sampling frequencies for within-individual peak resultant acceleration tracking during “easy” prolonged runs for research, clinical, and coaching applications.