Using wearable sensors to classify subject-specific running biomechanical gait patterns based on changes in environmental weather conditions

Running-related overuse injuries can result from a combination of various intrinsic (e.g., gait biomechanics) and extrinsic (e.g., running surface) risk factors. However, it is unknown how changes in environmental weather conditions affect running gait biomechanical patterns since these data cannot be collected in a laboratory setting. Therefore, the purpose of this study was to develop a classification model based on subject-specific changes in biomechanical running patterns across two different environmental weather conditions using data obtained from wearable sensors in real-world environments. Running gait data were recorded during winter and spring sessions, with recorded average air temperatures of -10° C and +6° C, respectively. Classification was performed based on measurements of pelvic drop, ground contact time, braking, vertical oscillation of pelvis, pelvic rotation, and cadence obtained from 66,370 strides (~11,000/runner) from a group of recreational runners. A non-linear and ensemble machine learning algorithm, random forest (RF), was used to classify and compute a heuristic for determining the importance of each variable in the prediction model. To validate the developed subject-specific model, two cross-validation methods (one-against-another and partitioning datasets) were used to obtain experimental mean classification accuracies of 87.18% and 95.42%, respectively, indicating an excellent discriminatory ability of the RF-based model. Additionally, the ranked order of variable importance differed across the individual runners. The results from the RF-based machine-learning algorithm demonstrates that processing gait biomechanical signals from a single wearable sensor can successfully detect changes to an individual's running patterns based on data obtained in real-world environments.

Establishing cut-points for physical activity classification using triaxial accelerometer in middle-aged recreational marathoners

The purpose of this study was to establish GENEA (Gravity Estimator of Normal Everyday Activity) cut-points for discriminating between six relative-intensity activity levels in middle-aged recreational marathoners. Nighty-eight (83 males and 15 females) recreational marathoners, aged 30–45 years, completed a cardiopulmonary exercise test running on a treadmill while wearing a GENEA accelerometer on their non-dominant wrist. The breath-by-breath V̇O₂ data was also collected for criterion measure of physical activity categories (sedentary, light, moderate, vigorous, very vigorous and extremely vigorous). GENEA cut-points for physical activity classification was performed via Receiver Operating Characteristic (ROC) analysis. Spearman’s correlation test was applied to determine the relationship between estimated and measured intensity classifications. Statistical analysis were done for all individuals, and separating samples by sex. The GENEA cut-points established were able to distinguish between all six-relative intensity levels with an excellent classification accuracy (area under the ROC curve (AUC) values between 0.886 and 0.973) for all samples. When samples were separated by sex, AUC values were 0.881–0.973 and 0.924–0.968 for males and females, respectively. The total variance in energy expenditure explained by GENEA accelerometer data was 78.50% for all samples, 78.14% for males, and 83.17% for females. In conclusion, the wrist-worn GENEA accelerometer presents a high capacity of classifying the intensity of physical activity in middle-aged recreational marathoners when examining all samples together, as well as when sample set was separated by sex. This study suggests that the triaxial GENEA accelerometers (worn on the non-dominant wrist) can be used to predict energy expenditure for running activities.

Activity classification using the GENEA: optimum sampling frequency and number of axes

INTRODUCTION

The GENEA shows high accuracy for classification of sedentary, household, walking, and running activities when sampling at 80 Hz on three axes. It is not known whether it is possible to decrease this sampling frequency and/or the number of axes without detriment to classification accuracy. The purpose of this study was to compare the classification rate of activities on the basis of data from a single axis, two axes, and three axes, with sampling rates ranging from 5 to 80 Hz.

METHODS

Sixty participants (age, 49.4 yr (6.5 yr); BMI, 24.6 kg·m (3.4 kg·m)) completed 10-12 semistructured activities in the laboratory and outdoor environment while wearing a GENEA accelerometer on the right wrist. We analyzed data from single axis, dual axes, and three axes at sampling rates of 5, 10, 20, 40, and 80 Hz. Mathematical models based on features extracted from mean, SD, fast Fourier transform, and wavelet decomposition were built, which combined one of the numbers of axes with one of the sampling rates to classify activities into sedentary, household, walking, and running.

RESULTS

Classification accuracy was high irrespective of the number of axes for data collected at 80 Hz (96.93% ± 0.97%), 40 Hz (97.4% ± 0.73%), 20 Hz (96.86% ± 1.12%), and 10 Hz (97.01% ± 1.01%) but dropped for data collected at 5 Hz (94.98% ± 1.36%).

CONCLUSION

Sampling frequencies >10 Hz and/or more than one axis of measurement were not associated with greater classification accuracy. Lower sampling rates and measurement of a single axis would result in a lower data load, longer battery life, and higher efficiency of data processing. Further research should investigate whether a lower sampling rate and a single axis affects classification accuracy when considering a wider range of activities.

Sprint interval running increases insulin sensitivity in young healthy subjects

High intensity cycling training increases oxidative capacity in skeletal muscles and improves insulin sensitivity. The present study compared the effect of eight weeks of sprint interval running (SIT) and continuous running at moderate intensity (CT) on insulin sensitivity and cholesterol profile in young healthy subjects (age 25.2 ± 0.7; VO(2max) 49.3 ± 1.2 ml·kg(-1)·min(-1)). SIT and CT increased maximal oxygen uptake by 5.3 ± 1.8 and 3.8 ± 1.6%, respectively (p < 0.05 for both). Oral glucose tolerance test (OGTT) was performed before and 60 h after the last training session. SIT, but not CT, reduced glucose area under curve and improved HOMA β-cell index (p < 0.05). Insulin area under curve did not decrease significantly in any group. SIT, but not CT, reduced LDL and total cholesterol. In conclusion, sprint interval running improves insulin sensitivity and cholesterol profile in healthy subjects, and sprint interval running may be more effective to improve insulin sensitivity than continuous running at moderate intensity.

Correlation of symmetrical gaits and whole body mechanics: debunking myths in locomotor biodynamics

Independent maturation of gait (Hildebrand) and whole body mechanics (Cavagna et al.) traditions in locomotor analyses has led to conflicting terminology. Re-evaluation of these traditions yields three primary insights. First, walking and running should be recognized by their fundamentally different mechanics. Because duty factor fails to consistently distinguish these mechanics, its use in discriminating walks from runs should be abandoned in preference to parameters that more accurately reflect the movements of the center of mass (COM; phase difference in external mechanical energy or Froude number). Second, "trot" should be reserved as a descriptor of a particular footfall pattern. This and all gait terms lack explicit information about limb compliance and thus COM movements. Third, symmetrical gait definitions should be broadened to reflect the four primary footfall patterns: the lateral-couplet dominated pattern of the pace, the diagonal-couplet dominated pattern of the trot and the more independent sequencing of footfalls of the two singlefoots. Intermediate gaits (perennially confusing and a mouthful to pronounce) are thereby subsumed by these four discrete gaits. Confusion between gait terminologies would be avoided if limb phase were consistently reported.

Demographic characteristics of elite Ethiopian endurance runners

INTRODUCTION

The dominance of East-African athletes in distance running remains largely unexplained; proposed reasons include favorable genetic endowment and optimal environmental conditions.

PURPOSE

To compare the demographics of elite Ethiopian athletes with the general Ethiopian population and assess the validity of reports linking running long distances to school with endurance success.

METHODS

Questionnaires, administered to 114 members (male and female) of the Ethiopian national athletics team and 111 Ethiopian control subjects (C) obtained information on place of birth, language, distance and method of travel to school. Athletes were separated into three groups according to athletic discipline: marathon (M; N = 34); 5,000-10,000 m (5-10 km; N = 42); and other track and field athletes (TF; N = 38). Frequency differences between groups were assessed using contingency chi-square tests.

RESULTS

Regional distributions of marathon athletes differed from controls (P < 0.001) and track and field athletes (P = 0.013), but not the 5- to 10-km athletes (P = 0.21). The 5- to 10-km athletes also differed from controls (P < 0.001). Marathon athletes exhibited excess from the regions of Arsi and Shewa (M: 73%; 5-10 km: 43%; TF: 29%; C: 15%). The language distribution of marathon athletes differed from all groups (P < 0.001), with a predominance of languages of Cushitic origin (M: 75%, 5-10 km: 52%, TF: 46%, C: 30%). A higher proportion of marathon athletes ran to school (M: 68%; 5-10 km: 31%; TF: 16%; C: 24%) and traveled greater distances.

CONCLUSION

Elite endurance athletes are of a distinct environmental background in terms of geographical distribution, ethnicity, and also having generally traveled farther to school, often by running. These findings may reflect both environmental and genetic influences on athletic success in Ethiopian endurance athletes.

Oxygen consumption and energy expenditure of level versus downhill running

OBJECTIVE

The purpose of this study was to assess and compare submaximal oxygen consumption (VO2) and energy expenditure (kJ) while running at 0, -1.8, -3.6, and -5.4% grades for three individually selected running speeds (9.4 + 0.79, 10.3 + 0.74, 11.3 + 0.73 km.h-1).

EXPERIMENTAL DESIGN

Subjects completed the four grade conditions in random order via a modified Latin squares design at three self-selected submaximal running speeds for each condition.

PARTICIPANTS

Thirteen (5 females and 8 males) recreational (< 35 km.wk-1) runners (age: 27.7 +/- 4.3 yrs) volunteered for the study.

MEASURES

Two-way repeated measures ANOVA (Grade x Speed) was used to analyze steady-state VO2 and kJ expenditure. Stepwise linear multiple regression was used to develop an equation for predicting VO2 for running at recreational speeds on moderately negative grades.

RESULTS

VO2 and kJ mean values were significantly different between all speed and % grade comparisons. Compared to level grade, the average reductions in VO2 and kJ expenditure ranged from approximately 9% at a grade of -1.8% to 22% at a grade of -5.4%. The relationship between VO2 and % grade for each running speed was linear.

CONCLUSIONS

For a given speed, running at a modest negative grade can significantly decrease VO2 and kJ expenditure compared to level running. The following regression equation can be used to estimate VO2 (ml.kg-1.min-1) for running at recreational speeds on slight downhills: VO2 = 6.8192 + 0.1313 (speed in m.min-1) + 1.2367 (% grade).

Effects of deep water and treadmill running on oxygen uptake and energy expenditure in seasonally trained cross country runners

OBJECTIVE

The purpose of this study was to physiologically compare submaximal intensity deep water running (DWR) and treadmill running (TMR) exercise in trained athletes.

EXPERIMENTAL DESIGN

Pre-test, post-test, 2 x 2 factoral design.

SETTING

Treadmill exercise tests occurred in the Human Performance Laboratory. DWR trials took place in the deep end of the University pool.

PARTICIPANTS

Seasonally trained college-aged male cross country runners (N = 8). Subjects completed a treadmill maximal oxygen consumption (VO2max) test, followed by a submaximal treadmill and deep water run at heart rates equivalent to 60% and 80% treadmill VO2max.

MEASURES

Oxygen consumption (VO2), ventilation (VEstpd), rates of perceived exertion (RPE), energy expenditure (kcal.min-1), respiratory exchange ratio (RER), fat and carbohydrate oxidation (g.min-1) were measured during two 5 minute steady state stages for both trials.

RESULTS

The trial by intensity interaction for VEstpd was significant, demonstrating greater ventilation during DWR as compared to TMR at 80% VO2max. The main effect of trial demonstrated that significantly higher RER and carbohydrate oxidation, and lower fat oxidation occurred during DWR as compared to TMR. VO2, RPE, and energy expenditure did not differ significantly between trials.

CONCLUSIONS

DWR is a comparable form of submaximal intensity exercise as TMR in well-trained athletes. DWR does, however, maintain unique properties that differs it from TMR. Therefore, the concept of training specificity should be further considered when prescribing DWR and using it as an enhancement tool or substitute for dry land running.

Specificity of training and its relation to the performance of distance runners

Validated six-month retrospective questionnaires were completed by 119 female and 234 male coached distance runners (59% compliance) for a descriptive study of relationships between specificity of training and best performance in a summer season. Runners were aged 26 +/- 10 y (mean +/- SD), specialising in distances from 800 m to the marathon, with seasonal best paces of 82 +/- 7% of sex- and age-group world records. They reported typical weekly durations of interval and strength training, and typical weekly durations and paces of moderate and hard continuous running, for build-up, precompetition, competition, and postcompetition phases of the season. The training programs showed some evidence of specificity, especially for runners preparing for longer events. A potentially beneficial effect of specificity was evident in a significant (p < 0.01) correlation between performance and seasonal mean weekly duration of moderate continuous running for runners specialising in longer distances (r = 0.29). The only other significant correlates of performance were seasonal mean relative training paces of moderate (r = -0.18) and hard (r = -0.42) continuous running, which exemplified detrimental effects of specificity for most runners. Thus, the training of better runners is not characterised strongly by greater specificity.

Exercise-induced muscle damage: effect on circulating leukocyte and lymphocyte subsets

The purpose of this study was to determine the effect of downhill and level running on circulating leukocyte and lymphocyte subsets and T lymphocyte activation. Using a random cross-over design, 10 runners completed two trials of 60 min of level running (0% grade; LR) and downhill running (-10% grade; DHR) at 70% of level VO2max. Blood samples were obtained preexercise and immediately postexercise (POST) and at 1.5, 12, 24, and 48 h of recovery. Creatine kinase activity peaked at 12 h of recovery from DHR and was not significantly altered following LR. The number of total T, CD16+, CD3+CD56+ cells were significantly higher POST DHR compared with LR. Leukocyte and neutrophil counts were significantly higher at 1.5 and 12 h of recovery from DHR compared with LR. The number of activated CD8+ cells (CD25+ CD8+) was significantly higher at 12 h of DHR compared to LR. Total T cells were significantly reduced at various time points during the 48 h of recovery from LR and DHR. In summary, DHR relative to LR resulted in a greater mobilization of lymphocytes (post), neutrophils (1.5-12 h of recovery) and activation of CD8+ cells at 12 h of recovery. In addition, reductions in circulating T lymphocyte subsets occurred following both conditions.

Prevention of running injuries by warm-up, cool-down, and stretching exercises

The purpose of this study was to evaluate the effect of a health education intervention on running injuries. The intervention consisted of information on, and the subsequent performance of, standardized warm-up, cool-down, and stretching exercises. Four hundred twenty-one male recreational runners were matched for age, weekly running distance, and general knowledge of preventing sports injuries. They were randomly split into an intervention and a control group: 167 control and 159 intervention subjects participated throughout the study. During the 16-week study, both groups kept a daily diary on their running distance and time, and reported all injuries. In addition, the intervention group was asked to note compliance with the standardized program. At the end of the study period, knowledge and attitude were again measured. There were 23 injuries in the control group and 26 in the intervention group. Injury incidence for control and intervention subjects was 4.9 and 5.5 running injuries per 1000 hours, respectively. The intervention was not effective in reducing the number of running injuries; it proved significantly effective (P < 0.05) in improving specific knowledge of warm-up and cool-down techniques in the intervention group. This positive change can perhaps be regarded as a first step on the way to a change of behavior, which may eventually lead to a reduction of running injuries.