Maximum-Entropy-Rate Selection of Features for Classifying Changes in Knee and Ankle Dynamics During Running

Einicke, G.A.; Sabti, Haider A.; Thiel, David V.; Fernández, Marta

doi:10.1109/jbhi.2017.2711487

Cited by 15 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4 Logistic regression is half of the total attributes, to see if the accuracy of the stacked model is affected. Then apply the same stacked classifier to check if its accuracy performance improves [28][29][30].…”

Section: Resultsmentioning

confidence: 99%

Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer

Chaurasia

Pal

2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

Breast cancer is the second most common cancer in women worldwide. The uncontrolled growth of breast cells is called breast cancer. The treatment of human breast cancer is a very critical process, and sometimes certain indicators may produce negative results. To avoid this misleading outcome situation, a reliable and accurate breast cancer diagnosis system must be available. The machine learning (ML) method is a modern and accurate technique that researchers have recently applied to predict and diagnose breast cancer. In this research article, we developed stack-based ensemble techniques and feature selection methods for the comprehensive performance of the algorithm and comparative analysis of breast cancer datasets with reduced attributes and all attributes. In this article, we first take the SVM, k nearest neighbors, Naive Bayes and perceptron as four ML algorithms as sub-models that have been trained and predicted from, and then combine them to make a new model called blending (stacking). Finally, logistic regression is used to predict the stacked model. It is significant that sub-models produce different results that are not correlated predictions. The stacking technique is best when all the submodels are skillfully combined together. This article uses the five-feature selection technique because it affects the overall performance of the model. Unrelated or moderately related features may adversely affect the behavior of the model. After applying the feature selection method, now we have data set with reduced features as well as all features. We implemented logistic regression on a dataset with all features and a dataset with reduced features. Finally, we see that the dataset with reduced features has got improved accuracy.

show abstract

Section: Resultsmentioning

confidence: 99%

Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer

Chaurasia

Pal

2021

SN COMPUT. SCI.

View full text Add to dashboard Cite

show abstract

“…Sport is becoming increasingly data driven and as a result, supervised machine learning and artificial neural network models have been adopted for athlete monitoring tasks such as automatic sport-specific movement classification to enhance work-load quantification [1]- [6] and athlete-state prediction to identify when athletes are, for example, entering a fatigued state [7], [8].…”

Section: Introductionmentioning

confidence: 99%

One Size Doesn't Fit All: Supervised Machine Learning Classification in Athlete-Monitoring

et al. 2021

Self Cite

View full text Add to dashboard Cite

Athlete movement data is integral for optimizing athlete-performance and can lead to reduced fatigue and in turn can mitigate injury risk. There is a substantial amount of scientific literature which investigates the ability of computervision and inertial sensor technologies to classify sport-specific movements. The coupling of automatic sport action labelling and athlete-monitoring data can significantly enhance athlete work-load monitoring. Two recent systematic reviews of the literature, pertinent to sport-specific movement classification, revealed that the majority of journal articles use athlete-dependent classification model training and evaluation methods. These methods can significantly enhance model classification performance, particularly with movements which have high inter-athlete technique variation. This is because it enables models to learn features distinctive to all athletes during training. This manuscript details the training and evaluation of supervised machine learning models to automatically classify running surface (athletics track, hard sand and soft sand) using features extracted from an upper-back inertial measurement unit sensor. Possible classification performance enhancement is demonstrated by comparing athlete independent and athlete dependent supervised machine learning methods. Using athlete dependent methods significantly increased the classification performance in terms of weighted average precision, recall, F1-score and accuracy (p < 0.05).

show abstract

“…Wearable inertial sensors have been adopted as an inexpensive tool for athlete monitoring [2][3][4] and can be used to analyse gait features such as symmetry, stride, step and stance durations, and differences between walking and running profiles [5][6][7][8]. The majority of reported investigations were conducted on grass, a treadmill or athletics running track [9,10].…”

Section: Introductionmentioning

confidence: 99%

Features Observed Using Multiple Inertial Sensors for Running Track and Hard-Soft Sand Running: A Comparison Study

Worsey

Espinosa

Shepherd

et al. 2020

The 13th Conference of the International Sports Engineering Association

Self Cite

View full text Add to dashboard Cite

Variability in the running surface can cause an athlete to alter their gait. Most literature report running on grass, a treadmill or athletics running tracks using inertial sensors. This study compares the signals obtained by 9 degrees of freedom (DOF) inertial-magnetic sensors incorporating an accelerometer (±16 g), gyroscope (±2000°/s) and magnetometer (±8 gauss). The sensors were placed on the participant’s shank, knee, lower spine and upper spine, and the participants were asked to run on three different surfaces (running track, hard sand and soft sand). The calculated player loads for a 400 m run on each surface type was very similar. The mean and standard deviation values were 577 ± 130, 581 ± 128, 568 ± 124 for soft sand, hard sand and the running track, respectively. This did not correlate with the participant’s self-assessment RPE (Rate of perceived exertion), which demonstrated running on soft sand to be significantly more challenging, yielding a mean and standard deviation of 5.3 ± 2.5 (Hard to Very Hard). Soft sand running had a decreased swing time duration but increased variability (0.44 ± 0.02 s—Swing Time, 6.5 ± 1.1%—CV), hard sand running had the longest swing and intermediate variability duration (0.46 ± 0.02 s—Swing Time, 3.30 ± 2.58 %—CV) and running track running had the medium swing time but lowest variability (0.45 ± 0.02 s, 2.7 ± 0.9%—CV). Gait dominance was not consistent across the surfaces for each participant and remained below a ratio of 0.4. These results provide an insight into how athletes modify their gait mechanics to accommodate different running surfaces.

show abstract

Maximum-Entropy-Rate Selection of Features for Classifying Changes in Knee and Ankle Dynamics During Running

Cited by 15 publications

References 24 publications

Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer

Stacking-Based Ensemble Framework and Feature Selection Technique for the Detection of Breast Cancer

One Size Doesn't Fit All: Supervised Machine Learning Classification in Athlete-Monitoring

Features Observed Using Multiple Inertial Sensors for Running Track and Hard-Soft Sand Running: A Comparison Study

Contact Info

Product

Resources

About