No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Classification methods can identify external constrains in swimming
Abstract
The purpose of the present study is to examine whether the use of fins is identifiable based on swimmers’ technique and to find out technique-related features that depict fins’ influence. First, a number of features were extracted from kinematic data given by movement sensors attached to swimmers’ bodies during butterfly swimming technique. Then, dimensionality reduction, feature selection and classification methods were applied to the extracted features. Two classification tasks were defined, one for the three classes of long, short and no fins, attaining accuracy up to 70, 62 and 70%, respectively, and the two-class simplified version (long fins, no fins) with accuracy up to 78%. These high accuracy levels were also found statistically significant and suggest that the use of fins influences swimming technique in a recognizable way and that the selected features that depict those differences are swimming type depended.
... Fundamental elements of swimming from a technical point of view can be summarized as: body position in the water, stroke, kicking, breathing, coordination and synchronization. 1 The movements of specific body parts during swimming have been recognized in the procedure of four swimming styles: i) Freestyle is one of the first strokes learnt by young swimmers. To perform it, lie on tummy in the water, by alternating arms forward, whereas legs should be kicking in a flutter movement; ii) Butterfly is the hardest stroke, but offers an excellent workout. ...
... At the beginning of the sports season, corresponding with baseline (T0), the pain level was quite mild (1)(2)(3). Subsequently, it is noticeable that the intensity pain increased to become severe (7-10), especially for Freestyle and Backstroke groups. ...
It is well-known that swimming purposes to increase the tonic-postural control. Beyond its physiological advantages, swimming also offers an exclusive platform to explore the complex interplay between body biomechanics and posture. The specific aim of this study was to investigate the effects of main swimming styles on postural balance in young athletes. Forty-one participants, aged between 11 and 15 years old (M= 13, SD= 1.47), were recruited. The training schedule usually consisted of 2/3 h (2.4 ± 0.46) per day (five to six weekly workouts). Measures included a postural assessment to identify the presence of postural deficits and a baropodometric stabilometry to evaluate the center of pressure. Measurements were performed before T0 (baseline), after 6 months (T1), and at the end, after 12 months (T2). Beforehand, all participants undertook identification of the swimming style and pain intensity level. Results showed that Breaststroke and Butterfly athletes had clear improvements in postural balance compared to Backstroke and Freestyle athletes. In conclusion, our results suggest that a detailed knowledge of the different swimming styles plays a significant role in improving balance and postural stability in young athletes, highlighting the fundamental role of the kinesiology in sports traumatology.
... Lower extremity kinematics were quantified using a pair of wireless Bluetooth, 9-degree-of-freedom IMUs (k-Sens [now K-Move]), Kinvent, Biomechanique, Montpellier, France with 120-Hz sampling frequency. This sensor has been successfully used in other studies [19,20]. Linear acceleration and angular velocity were used for further analysis. ...
We explored the impact of running in the severe intensity domain on running mechanics and muscle oxygenation in competitive runners by investigating the relationship between mechanical deviations from typical stride characteristics and muscle oxygen saturation (SmO2) in the quadriceps muscle. Sixteen youth competitive runners performed an 8‐min exhaustive running test on an outdoor track. Running mechanics were continuously monitored using inertial measurement units. Rectus femoris SmO2 and total hemoglobin (a measure of blood volume) were continuously monitored by near‐infrared spectroscopy. One‐class support vector machine (OCSVM) modeling was employed for subject‐specific analysis of the kinematic data. Statistical analysis included principal component analysis, ANOVA, and correlation analysis. Mechanical deviations from typical stride characteristics increased as the running test progressed. Specifically, the percentage of outliers in the OCSVM model rose gradually from 2.2 ± 0.8% at the start to 43.6 ± 28.2% at the end (p < 0.001, mean ± SD throughout). SmO2 dropped from 74.3 ± 8.4% at baseline to 10.1 ± 6.8% at the end (p < 0.001). A moderate negative correlation (r = −0.61, p = 0.013) was found between the average SmO2 and the percentage of outlier strides during the last 15% of the run. During high‐intensity running, alterations in running biomechanics may occur, linked to decreased quadriceps muscle oxygenation. These parameters highlight the potential of using running kinematics and muscle oxygenation in training to optimize performance and reduce injury risks. Our research contributes to understanding biomechanical and physiological responses to endurance running and emphasizes the importance of individualized monitoring.
... Recently, the study of stroke cycles has been dominated by analyzing a few strokes (Kudo et al. 2019;Gonjo et al. 2020;Gonjo et al. 2021), presenting some methodological limitations. The use of inertial sensors provides a valid solution to overcome this issue (Nikodelis et al. 2013;Averianova et al. 2016;Grigoriou et al. 2019), increasing the capability to capture many consecutive cycles and allowing the implementation of different data treatments like auto-correlation, frequency spectrum and cross-correlation analyses that can capture the repeatability and smoothness of the swimming pattern as well as inter-segmental coupling. ...
Background: Front crawl and backstroke share similar trunk rotating characteristics and tempt coaches to transfer teaching parts from one stroke to the other intuitively. However, the degree of similarity has yet to be determined. The coordination of the pelvis and the 7th cervical vertebrae (C7), during yaw and roll rotation, when sprint swimming front crawl, and backstroke was studied. Methods: Thirty-four swimmers were assessed on their performance in25m-sprint of each stroke. Using inertial sensors, each segment’s time series of angular displacement was calculated. Their amplitudes, mean autocorrelation values, max cross-correlation coefficient, phase lag, and relative power at the main frequency were analyzed. For all comparisons, the p-value was set to 0.05. Results: Pelvis yaw and roll and C7 roll amplitudes were greater at backstroke, C7 yaw was greater at front crawl. Autocorrelations ranged from 0.79 to 0.82 except for the pelvis at front crawl in yaw which was 0.72±0.16. Relative power at the main frequency ranged from 47% to 52% except for the yaw pelvis’ at the front crawl which was lower (32.81±14.09%). Backstroke had larger mean values in all cases and roll had larger mean values than yaw. Cross-correlation between the two segments yielded higher values at roll. At roll direction, the leading segment in the front crawl was the pelvis while in backstroke, it was the C7 which was true in all cases. In all cases, the coupling was slightly deviating from in-phase mode except from backstroke yaw which yield phase lag values of -13.35±1.14% of stroke cycle time. Conclusions: Although both strokes share similar characteristics their intersegmental coupling differs. The findings of the study imply that proper focus should be given to enhance only a positive transfer of learning between the two strokes.
... This could be explained by the influence they have on the body position in displacement, which is one of the most determining aspects in the final speed of the swimmer (Wei et al., 2014). This has been measured with motion simulation systems (Nakashima et al., 2019) and with experimental studies with national level swimmers (Grigoriou et al., 2019). ...
1) Background: Swimming is a common choice to start practicing sports. Besides, swimming can be considered an early specialization sport. Thus, it is important finding adequate methodologies to improve the teaching-learning process of the technical parameters of the swimming styles. It is unknown if the use of fins can affect the athlete's improvements at long-term, despite being a common resource in swimming schools due to its immediate effect on swimming speed; (2) Methods: 37 kids and young recreational swimmers participated in this study. 20 kids swimmers (KS) were between 7 and 9 years old and 17 young swimmers (YS), between 12 and 14 years old. At the same time, both groups were split into two subgroups-one that used fins to train (KSF and YSF) and another one that did not used fins (KSNF and YSNF). The study lasted 8 weeks, with two 1-hour weekly sessions. Weeks 1 and 8 were dedicated to the pre-test and post-test, respectively, where the time of swimming different distances freestyle, backstroke and butterfly was kept. The information was classified based on the groups and a statistical analysis was performed; (3) Results: all groups improved their performance after the intervention. However, no significant differences were found between groups except in KSF and KNSF, where KSF had a greater improvement in the backstroke test (p(ES)= 0.005(0.78)); (4) Conclusions: a 6-week training program improves performance in KS and YS. The continued use of fins does not seem to affect the performance improvement in these age groups, except in backstroke, where it helps improve significantly in KFS. These results can help the coaches of swimming schools to create more homogeneous groups in speed and to organize more easily the training session unifying swimming paces with the use of fins by slower children without the risk of affecting their improvement.
Identification of errors or anomalous values, collectively considered outliers, assists in exploring data or through removing outliers improves statistical analysis. In biomechanics, outlier detection methods have explored the ‘shape’ of the entire cycles, although exploring fewer points using a ‘moving-window’ may be advantageous. Hence, the aim was to develop a moving-window method for detecting trials with outliers in intra-participant time-series data. Outliers were detected through two stages for the strides (mean 38 cycles) from treadmill running. Cycles were removed in stage 1 for one-dimensional (spatial) outliers at each time point using the median absolute deviation, and in stage 2 for two-dimensional (spatial–temporal) outliers using a moving window standard deviation. Significance levels of the t-statistic were used for scaling. Fewer cycles were removed with smaller scaling and smaller window size, requiring more stringent scaling at stage 1 (mean 3.5 cycles removed for 0.0001 scaling) than at stage 2 (mean 2.6 cycles removed for 0.01 scaling with a window size of 1). Settings in the supplied Matlab code should be customised to each data set, and outliers assessed to justify whether to retain or remove those cycles. The method is effective in identifying trials with outliers in intra-participant time series data.
Background: Marginal changes in the execution of competitive sports movements can represent a significant change for performance success. However, such differences may emerge only at certain execution intensities and are not easily detectable through conventional biomechanical techniques. This study aimed to investigate if and how competition standard and progression speed affect race walking kinematics from both a conventional and a coordination variability perspective.
Methods: Fifteen experienced athletes divided into three groups (Elite, International, and National) were studied while race walking on a treadmill at two different speeds (12.0 and 15.5 km/h). Basic gait parameters, the angular displacement of the pelvis and lower limbs, and the variability in continuous relative phase between six different joint couplings were analyzed.
Results: Most of the spatio-temporal, kinematic, and coordination variability measures proved sensitive to the change in speed. Conversely, non-linear dynamics measures highlighted differences between athletes of different competition standard when conventional analytical tools were not able to discriminate between different skill levels. Continuous relative phase variability was higher for National level athletes than International and Elite in two couplings (pelvis obliquity – hip flex/extension and pelvis rotation – ankle dorsi/plantarflexion) and gait phases (early stance for the first coupling, propulsive phase for the second) that are deemed fundamental for correct technique and performance.
Conclusion: Measures of coordination variability showed to be a more sensitive tool for the fine detection of skill-dependent factors in competitive race walking, and showed good potential for being integrated in the assessment and monitoring of sports motor abilities.
Body sensor networks (BSN) have emerged as an active field of research to connect and operate sensors within, on or at close proximity to the human body. BSN have unique roles in health applications, particularly to support real-time decision making and therapeutic treatments. Nevertheless, challenges remain in designing BSN nodes with antennas that operate efficiently around, ingested or implanted inside the human body, as well as new methods to process the heterogeneous and growing amount of data on-node and in a distributed system for optimized performance and power consumption. As the battery operating time and sensor size are two important factors in determining the usability of BSN nodes, ultra-low power transceivers, energy aware network protocol, data compression, on-node processing and energy harvesting techniques are highly demanded to ultimately achieve a self-powered BSN.
Female runners have a two-fold risk of sustaining certain running-related injuries as compared to their male counterparts. Thus, a comprehensive understanding of the sex-related differences in running kinematics is necessary. However, previous studies have either used discrete time point variables and inferential statistics and/or relatively small subject numbers. Therefore, the first purpose of this study was to use a principal component analysis (PCA) method along with a support vector machine (SVM) classifier to examine the differences in running gait kinematics between female and male runners across a large sample of the running population as well as between two age-specific sub-groups. Bilateral 3-dimensional lower extremity gait kinematic data were collected during treadmill running. Data were analysed on the complete sample (n = 483: female 263, male 220), a younger subject group (n = 56), and an older subject group (n = 51). The PC scores were first sorted by the percentage of variance explained and we also employed a novel approach wherein PCs were sorted based on between-gender statistical effect sizes. An SVM was used to determine if the sex and age conditions were separable and classifiable based on the PCA. Forty PCs explained 84.74% of the variance in the data and an SVM classification accuracy of 86.34% was found between female and male runners. Classification accuracies between genders for younger subjects were higher than a subgroup of older runners. The observed interactions between age and gender suggest these factors must be considered together when trying to create homogenous sub-groups for research purposes.
In a biophysical approach to the study of swimming performance (blending biomechanics and bioenergetics), inter-limb coordination is typically considered and analysed to improve propulsion and propelling efficiency. In this approach, 'opposition' or 'continuous' patterns of inter-limb coordination, where continuity between propulsive actions occurs, are promoted in the acquisition of expertise. Indeed a 'continuous' pattern theoretically minimizes intra-cyclic speed variations of the centre of mass. Consequently, it may also minimize the energy cost of locomotion. However, in skilled swimming performance there is a need to strike a delicate balance between inter-limb coordination pattern stability and variability, suggesting the absence of an 'ideal' pattern of coordination toward which all swimmers must converge or seek to imitate. Instead, an ecological dynamics framework advocates that there is an intertwined relationship between the specific intentions, perceptions and actions of individual swimmers, which constrains this relationship between coordination pattern stability and variability. This perspective explains how behaviours emerge from a set of interacting constraints, which each swimmer has to satisfy in order to achieve specific task performance goals and produce particular task outcomes. This overview updates understanding on inter-limb coordination in swimming to analyse the relationship between coordination variability and stability in relation to interacting constraints (related to task, environment and organism) that swimmers may encounter during training and performance.
The complex movement sequences of golf require supporting tools for players and coaches alike. We developed a system that classifies the experience level and trained it with data from an inertial sensor on the club head. Based on 315 golf putts from eleven subjects the system differentiated between experienced and unexperienced players with a classification rate of 86.1%. To improve the classification system and obtain discriminant features we additionally integrated a feature selection step. We compared different selection approaches and concluded that a leave-subject-out feature selection was the appropriate approach to predict the true performance of a live system. The selected features can be fed back to coaches and help them to guide players to a better putting technique.
The purpose of the study was to compare the instantaneous velocity produced by a lifesaver in a 25 m manne-quin carry test, while performing barefoot and with four fin types. Comparisons between the first and second half parts of each test were also accomplished. Complementarily, the fatigue effect over the test distance was analyzed through the assessment of the slope of the velocity over time and of the fatigue index. One licensed male lifesaver performed 5x25m maximal swim test carrying a mannequin (barefoot and with flexible, short, stiff and fiber fin types). A cable speedometer was connected to the mannequin in order to measure instantaneous velocity. Results showed that swimming with barefoot does not allow high mannequin carries performances. Flexible fins were the ones that produce the lower velocity values in the 25m mannequin carry effort. It was observed that fiber fins were the ones who allowed for higher mean velocity irrespectively of the effort part considered. Short and stiff fins use allowed for an increase of velocity during the first half effort part with a consequent reduction of fatigue effects. For the second half effort part, and for the total effort, short fins were the ones which induced lower velocity decay, followed by fiber fins. Short fins and fiber fins revealed the lower fatigue indexes. It was concluded that the selection of the best fin type is dependent of the fin length and could change with the effort distance.
Paddles and fins are used during swim training and practice as tools for improving performance. The use of these equipment can alter physiological and kinematic parameters of swimming. The purpose of this literature review was to present and discuss the effects of paddles and fins on kinematic and physiological variables in front crawl, and provide update on the topic for teachers, researchers, coaches and swimmers. Thirty articles were reviewed. To crawl, paddles can change the averages of stroke length and stroke rate, the average swimming speed, the absolute duration of the stroke phases and the index of coordination. Fins can modify the average stroke rate, the average swimming speed, the kick frequency and deep, and the energy cost. We found no studies that verified the longitudinal effects of the use of paddles and fins on these parameters.
We investigated the effect of a one-time application of elastic constraints on movement-inherent variability during treadmill running. Eleven males ran two 35-min intervals while surface EMG was measured. In one of two 35-min intervals, after 10 min of running without tubes, elastic tubes (between hip and heels) were attached, followed by another 5 min of running without tubes. To assess variability, stride-to-stride iEMG vari-ability was calculated. Significant increases in variability (36 % to 74 %) were observed during tube running, whereas running without tubes after the tube running block showed no significant differences. Results show that elastic tubes affect variability on a muscular level despite the constant environmental conditions and underline the nervous system's adaptability to cope with somehow unpredictable constraints since stride duration was unaltered.
The effect of (a) increasing velocity and (b) added resistance was examined on the stroke (stroke length, stroke rate [SR]), coordination (index of coordination [IdC], propulsive phases), and force (impulse and peaks) parameters of 7 national-level front crawl swimmers (17.14 ± 2.73 years of swimming; 57.67 ± 1.62 seconds in the 100-m freestyle). The additional resistance was provided by a specially designed parachute. Parachute swimming (PA) and free-swimming (F) conditions were compared at 5 velocities per condition. Video footage was used to calculate the stroke and coordination parameters, and sensors allowed the determination of force parameters. The results showed that (a) an increase in velocity (V) led to increases in SR, IdC, propulsive phase duration, and peak propulsive force (p < 0.05), but no significant change in force impulse per cycle, whatever the condition (PA or F); and (b) in PA conditions, significant increases in the IdC, propulsive phase duration, and force impulse and a decrease in SR were recorded at high velocities (p < 0.05). These results indicated that, in the F condition, swimmers adapted to the change in velocity by modifying stroke and coordination rather than force parameters, whereas the PA condition enhanced the continuity of propulsive action and force development. Added resistance, that is, "parachute training," can be used for specific strength training purposes as long as swimming is performed near maximum velocity.
The identification of differences between groups is often important in biomechanics. This paper presents group classification tasks using kinetic and kinematic data from a prospective running injury study. Groups composed of gender, of shod/barefoot running and of runners who developed patellofemoral pain syndrome (PFPS) during the study, and asymptotic runners were classified. The features computed from the biomechanical data were deliberately chosen to be generic. Therefore, they were suited for different biomechanical measurements and classification tasks without adaptation to the input signals. Feature ranking was applied to reveal the relevance of each feature to the classification task. Data from 80 runners were analysed for gender and shod/barefoot classification, while 12 runners were investigated in the injury classification task. Gender groups could be differentiated with 84.7%, shod/barefoot running with 98.3%, and PFPS with 100% classification rate. For the latter group, one single variable could be identified that alone allowed discrimination.
The purpose of the present study was to determine the relative contribution of the aerobic (Aer), anaerobic lactic (AnL) and alactic (AnAl) energy sources during each of the four laps of a 200 m front crawl race. Additionally, energy cost (C) and arm stroke efficiency were also computed. Ten international swimmers performed a 200 m front crawl swim, as well as 50, 100, and 150 m at the 200 m pace. Oxygen consumption was measured during the 200 m swim and blood samples were collected before and after each swim; the C of swimming was calculated as the ratio of E
tot to distance (where E
tot = Aer + AnL + AnAl). Arm stroke efficiency was calculated by kinematic analysis as the speed of center of mass to the ratio of 3D hand speed. For the 200 m the contributions were 65.9% (Aer), 13.6% (AnL), and 20.4% (AnAl) whereas for each lap they were 44.6, 73.2, 83.3 and 66.6% (Aer), 14.1, 5.0, 4.4 and 28.1% (AnL) and 41.3, 21.8, 12.3 and 5.2% (AnAl) for the four laps, respectively. For the 200 m as a whole C was 1.60 kJ m−1 whereas C = 1.71, 1.56, 1.44 and 1.70 kJ m−1 for each consecutive lap, respectively. Arm stroke efficiency ranged from 0.40 to 0.43 and was significantly lower in the last lap as compared to the first (P = 0.002), suggesting the occurrence of fatigue. The decrease in arm stroke efficiency was mirrored by an increase in C as can be expected on theoretical grounds.
The aim of the present study was to quantify the improvements in the economy and efficiency of surface swimming brought about by the use of fins over a range of speeds (v) that could be sustained aerobically. At comparable speeds, the energy cost (C) when swimming with fins was about 40 % lower than when swimming without them; when compared at the same metabolic power, the decrease in C allowed an increase in v of about 0.2 ms(-1). Fins only slightly decrease the amplitude of the kick (by about 10 %) but cause a large reduction (about 40 %) in the kick frequency. The decrease in kick frequency leads to a parallel decrease of the internal work rate ((int), about 75 % at comparable speeds) and of the power wasted to impart kinetic energy to the water ((k), about 40 %). These two components of total power expenditure were calculated from video analysis ((int)) and from measurements of Froude efficiency ((k)). Froude efficiency (eta(F)) was calculated by computing the speed of the bending waves moving along the body in a caudal direction (as proposed for the undulating movements of slender fish); eta(F) was found to be 0.70 when swimming with fins and 0.61 when swimming without them. No difference in the power to overcome frictional forces ((d)) was observed between the two conditions at comparable speeds. Mechanical efficiency [(tot)/(Cv), where (tot)=(k)+(int)+(d)] was found to be about 10 % larger when swimming with fins, i.e. 0.13+/-0.02 with and 0.11+/-0.02 without fins (average for all subjects at comparable speeds).
The purpose of this study was to analyse stroke phases, arm-leg coordination and trunk motion fluctuation during breaststroke in elite male and female 50, 100 and 200 m events at the 9th FINA World Swimming Championships, Fukuoka 2001. Four phases of the arm stroke and three phases of the leg kick as well as phases of simultaneous arm and leg propulsion and recovery were identified from video of swimmers' motions below the surface. The duration of each phase was expressed as a proportion of the whole stroke cycle. Three measures of the arm-leg coordination, percent simultaneous arm-leg recovery time (%SRT), percent arm lag time (%ALT) and percent simultaneous arm-leg propulsion time (%SPT) were calculated. Mean mid-pool swimming hip velocity (V), stroke rate (SR) and stroke length (SL) were also calculated. In addition, the intra-cycle hip velocity of the swimmers was obtained by cinematographic analysis. The SR decreased and SL increased significantly as the event distance increased. For the arm-leg coordination the %ALT, %SPT and %SRT indicated significant differences between event, gender and performance level. In particular, for increasing event distance and for the higher performing swimmer the lower the %SPT and the higher the %SRT. In addition, the range of the intra-cycle hip velocity fluctuation in the lower performing group was greater than the higher performing group. The non-propulsive phase seems to be a key factor for better performance. The breaststroke swimmers must avoid rapid deceleration during the non-propulsive phase by adopting a low resistance posture and stroking technique.
The aim of this study was to investigate how fins with varying physical characteristics affect the energy cost and the efficiency of aquatic locomotion. Experiments were performed on ten college swimmers who were asked to swim the dolphin kick while using a monofin (MF) and to swim the front crawl kick with a small-flexible fin (SF), a large-stiff fin (LS) and without fins (BF, barefoot). The energy expended to cover one unit distance (C) was highest for BF (C=10.6+/-1.8 kJ m(-1) kg(-1) at 0.8 m s(-1)) and decreased by about 50% with LS, 55% with SF and 60% with MF, allowing for an increase in speed (for a given metabolic power) of about 0.4 m s(-1) for MF and of about 0.2 m s(-1) for SF and LS (compared with BF). At any given speed, the fins for which C was lower were those with the lowest kick frequency (KF): KF=1.6+/-0.22 Hz at 0.8 m s(-1) (for BF) and decreased by about 40% for SF, 50% for LS and 60% for MF. The decrease in KF from BF to SF-LS and MF was essentially due to the increasing surface area of the fin which, in turn, was associated with a higher Froude efficiency (eta(F)). eta(F) was calculated by computing the speed of the bending waves moving along the body in a caudal direction (as proposed for the undulating movements of slender fish): it increased from 0.62+/-0.01 in BF to 0.66+/-0.03 in SF and 0.67+/-0.04 in LS reaching the highest values (0.76+/-0.05) with MF. No single fin characteristic can predict a swimmer's performance, rather the better fin (i.e. MF) is the one that is able to reduce most KF at any given speed and hence to produce the greatest distance per kick (d=v/KF). The latter indeed increased from 0.50+/-0.01 m in BF to about 0.90+/-0.05 m in SF and LS and reached values of 1.22+/-0.01 m in MF.
The present study analyzes the changes in acceleration produced by swimmers before and after fatiguing effort. The subjects (n = 15) performed a 25-m crawl series at maximum speed without fatigue, and a second series with fatigue. The data were registered with a synchronized system that consisted of a position transducer (1 kHz) and a video photogrametry (50 Hz). The acceleration (ms(-2)) was obtained by the derivative analysis of the variation of the position with time. The amplitude in the time domain was calculated with the root mean square (RMS); while the peak power (PP), the peak power frequency (PPF) and the spectrum area (SA) were calculated in the frequency domain with Fourier analysis. On the one hand, the results of the temporal domain show that the RMS change percentage between series was 67.5% (p < 0.001). On the other hand, PP, PPF, and SA show significant changes (p < 0.001). PP and SA were reduced by 63.1% and 59.5%, respectively. Our results show that the acceleration analysis of the swimmer with Fourier analysis permits a more precise understanding of which propulsive forces contribute to the swimmer performance before and after fatigue appears.
A portable gait analysis and activity-monitoring system for the evaluation of activities of daily life could facilitate clinical and research studies. This current study developed a small sensor unit comprising an accelerometer and a gyroscope in order to detect shank and foot segment motion and orientation during different walking conditions. The kinematic data obtained in the pre-swing phase were used to classify five walking conditions: stair ascent, stair descent, level ground, upslope and downslope. The kinematic data consisted of anterior-posterior acceleration and angular velocity measured from the shank and foot segments. A machine learning technique known as support vector machine (SVM) was applied to classify the walking conditions. SVM was also compared with other machine learning methods such as artificial neural network (ANN), radial basis function network (RBF) and Bayesian belief network (BBN). The SVM technique was shown to have a higher performance in classification than the other three methods. The results using SVM showed that stair ascent and stair descent could be distinguished from each other and from the other walking conditions with 100% accuracy by using a single sensor unit attached to the shank segment. For classification results in the five walking conditions, performance improved from 78% using the kinematic signals from the shank sensor unit to 84% by adding signals from the foot sensor unit. The SVM technique with the portable kinematic sensor unit could automatically recognize the walking condition for quantitative analysis of the activity pattern.
The identification of between-group differences and changes in gait mechanics are useful for injury prevention. Previous studies suggest the differences in gait biomechanical variables may interact in a complex non-linear fashion rather than a simple linear fashion. A non-linear multivariate analysis technique is therefore necessary to unravel the inherent structure in the gait data. Kernel principal component analysis (KPCA) is an extension of principal component analysis (PCA), the most widely used method in this field, and can provide inside into non-linear relationships of the variables of interest. Despite the growing use of PCA in running gait research, no prior studies have evaluated the utility of KPCA in determining the between-group differences in running gait patterns. Therefore, the objective of this study was to compare the performance of KPCA and PCA in identifying differences between male and female runners and as well between young and older runners in running gait kinematics. Running kinematic data were analysed on a gender group (female, n = 50; male, n = 50) and an age group (young, n = 50; older, n = 50) for discrete variables and waveform data. The following key results were obtained: (1) the performance of the KPCA for the identification of gender and age differences in running gait kinematics increased as compared to using the linear PCA; (2) the Gaussian function often performed better than the polynomial function for these two experiments; (3) there was no consistent optimal value of either the width σ of the Gaussian kernel or the degree d of the polynomial kernel for different data types and experiments. These results suggest that non-linear features extracted by KPCA could be potentially useful in identifying and discriminating between-group differences and changes in running gait patterns, and could provide useful information about the intrinsic non-linear dynamics of running movement.
The purpose of the present study was to investigate the rotational kinematics pattern of the upper trunk and the pelvis and the complexity-variability of their movement, during the sprint butterfly stroke between male and female swimmers with long fins (18–26 cm) and without fins. Two pairs of 3D accelerometers and gyroscopes were used to measure segments' flexion-extension angles. There were no gender differences apart from the record. The amplitude values at the pelvis were significantly larger when swimming without fins while at the C7 they were significantly larger when swimming with them. Autocorrelation coefficients of angles' time histories were higher when swimming with fins for both segments. The power spectrums revealed a dominant frequency representing the stroke period which grew significantly stronger for the fin condition. Correlation Dimension verified a lower dimensionality for the fin condition for the C7 segment movement pattern. Overall fin use seems to offer “strength” to the swimmer’s movement pattern. This does not imply better movement coordination, but rather a trend toward a more stable attractor.
Five classification algorithms namely J48, Naive Bayes, Multi layer Perceptron, IBK and Bayes Net are evaluated using Mc Nemar's test over datasets including both nominal and numeric attributes. It was found that Multi layer Perceptron performed better than the two other classification methods for both nominal and numerical datasets. Furthermore, it was observed that the results of our evaluation concur with Kappa statistic and Root Mean Squared Error, two well-known metrics used for evaluating machine learning algorithms.
A new approach to control charts for industrial processes is suggested, based on the idea that the underlying stochastic process should be specified, its parameters estimated, and the control procedure related to these. Optimum estimators are derived, and practical approximations to these are indicated.
We introduce and explore an approach to estimating statistical significance of classifi-cation accuracy, which is particularly useful in scientific applications of machine learning where high dimensionality of the data and the small number of training examples ren-der most standard convergence bounds too loose to yield a meaningful guarantee of the generalization ability of the classifier. Instead, we estimate statistical significance of the observed classification accuracy, or the likelihood of observing such accuracy by chance due to spurious correlations of the high-dimensional data patterns with the class labels in the given training set. We adopt permutation testing, a non-parametric technique previously developed in classical statistics for hypothesis testing in the generative setting (i.e., com-paring two probability distributions). We demonstrate the method on real examples from neuroimaging studies and DNA microarray analysis and suggest a theoretical analysis of the procedure that relates the asymptotic behavior of the test to the existing convergence bounds.
We report the results of experiments on biological motion demonstrating the presence of critical order parameter fluctuations as the system evolves from one coordinated state to another at a critical control parameter value. This is a key feature of nonequilibrium phase transitions.
The use of swim fins has become popular in various water sport activities. While numerous models of swim fin with various innovative shapes have been subjectively designed, the exact influence of the fin characteristics on swimming performance is still much debated, and remains difficult to quantify. To date, the most common approach for evaluating swim fin propulsion is based on the study of "swimmer-fins" as a global system, where physiological and/or biomechanical responses are considered. However, reproducible swimming technique is difficult (or even impossible) to obtain on human body and may lead to discrepancies in data acquired between trials. In this study, we present and validate a new automat called HERMES which enables an evaluation of various swim fins during an adjustable, standardized and reproducible motion. This test bench reliably and accurately reproduces human fin-swimming motions, and gives resulting dynamic measurements at the ankle joint. Seven fins with various geometrical and mechanical characteristics were tested. For each swim fin, ankle force and hydromechanical efficiency (useful mechanical power output divided by mechanical power input delivered by the motors) were calculated. Efficiencies reported in our study were high (close to 70% for some swim fins) over a narrow range of Strouhal number (St) and peaks within the interval 0.2<St<0.4, as shown in previous studies on flying or swimming animals. Therefore, an interesting prospect in this work would be to accurately study the impact of adjustable fin kinematics and material (design and mechanical properties) on the wake structure and on efficiency.
A new training device derived from the MAD-system (system to measure active drag, Hollander et al. (8], providing fixed push off points in the water for swimming, the front crawl is described. The effects of training on this device (called POP from fixed Push Off Point) are determined by comparing the increase in performance of a training group (n = 11) to a control group (n = 11). The control group continued the normal training program. During ten weeks the training group followed the same program, but three times per week sprints performed on the POP were substituted for normal free swimming sprints. Despite the fact that training time and volume were equal, the training group showed a significantly greater improvement in force (from 91 to 94 N, 3.3%), velocity (from 1.75 to 1.81 m.s-1, 3.4%) and power (from 160 to 172 W, 7%) as measured on the MAD-system, and an increase in distance per stroke in free swimming. The training group showed a significant improvement in race times for 50 m (from 27.2 to 26.6 s), 100 m (from 59.3 to 57.4 s) and 200 m (from 129.6 to 127.3 s). It is concluded that the POP is a specific training device especially suitable for increasing maximal power output during swimming.
The walking gait in 20 healthy adolescent girls and 20 adolescent girls with idiopathic scoliosis was compared using time and frequency domain analysis of the ground reaction forces.
To investigate the support phase mechanics of walking gait in health subjects and those with scoliosis.
Previous studies have demonstrated the important role of gait in the etiology of scoliosis, although, to date, there are no detailed reports that examine kinetic gait parameters.
Selected force-time parameters were used to investigate time domain patterns. The frequency content of the vertical, anterior-posterior, and medial-lateral components was used to examine frequency characteristics. Inter- and intrasubject variability and bilateral symmetry also were examined.
The findings indicated no significant difference between the two groups for the time domain variables. However, the frequency content of the group with scoliosis was significantly higher han that of the control group, especially in the medial lateral component, suggesting presence of a balance control misfunction. In addition, substantially higher inter- and intrasubject variability and asymmetry was found within the frequency content of the group with scoliosis compared with the control group in the medial-lateral and anterior-posterior directions.
Patients with scoliosis exhibit balance problems during the stance phase of gait and have significant asymmetry in the frequency characteristics. These findings could be a primary effect that contributes to the medial-lateral deformity of the spine and its initiation and progression.
An accurate and computationally efficient means of classifying surface myoelectric signal patterns has been the subject of considerable research effort in recent years. Effective feature extraction is crucial to reliable classification and, in the quest to improve the accuracy of transient myoelectric signal pattern classification, an ensemble of time-frequency based representations are proposed. It is shown that feature sets based upon the short-time Fourier transform, the wavelet transform, and the wavelet packet transform provide an effective representation for classification, provided that they are subject to an appropriate form of dimensionality reduction.
This paper investigated application of a machine learning approach (Support vector machine, SVM) for the automatic recognition of gait changes due to ageing using three types of gait measures: basic temporal/spatial, kinetic and kinematic. The gaits of 12 young and 12 elderly participants were recorded and analysed using a synchronized PEAK motion analysis system and a force platform during normal walking. Altogether, 24 gait features describing the three types of gait characteristics were extracted for developing gait recognition models and later testing of generalization performance. Test results indicated an overall accuracy of 91.7% by the SVM in its capacity to distinguish the two gait patterns. The classification ability of the SVM was found to be unaffected across six kernel functions (linear, polynomial, radial basis, exponential radial basis, multi-layer perceptron and spline). Gait recognition rate improved when features were selected from different gait data type. A feature selection algorithm demonstrated that as little as three gait features, one selected from each data type, could effectively distinguish the age groups with 100% accuracy. These results demonstrate considerable potential in applying SVMs in gait classification for many applications.
Comparison of gait patterns between healthy and scoliotic patients using time and frequency domain analysis of ground reaction forces
- Giakas