Defining functional groups based on running kinematics using Self-Organizing Maps and Support Vector Machines

Hoerzer, Stefan; Tscharner, Vinzenz von; Jacob, Christian; Nigg, Benno M.

doi:10.1016/j.jbiomech.2015.03.017

Cited by 39 publications

(36 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[3] Several studies in the gait analysis domain have applied variations of this approach and various marker sets have been used across these studies. [1,[3][4][5][6][7][8][9][10][11][12] The marker sets can be differentiated based on two major aspects. Firstly, whether the marker set covers the full body or body parts; full body marker sets were used most frequently.…”

Section: Introductionmentioning

confidence: 99%

“…Firstly, whether the marker set covers the full body or body parts; full body marker sets were used most frequently. [1,[3][4][5][6]11] Body part marker sets include unilateral placement on the lower extremity, [7,10,12] unilateral placement on the upper and lower extremity [9] and bilateral placement on the lower body and thorax. [8] Secondly, marker sets can be differentiated based on whether the markers are placed exclusively on specific anatomical landmarks.…”

Section: Introductionmentioning

confidence: 99%

“…Marker placement exclusively on anatomical landmarks was used for most associated studies and these marker sets were based on Vicon's Plug-In-Gait R model [3,8,11] or on customized models. [6,7,9,10,12] Federolf, Boyer [4] included marker clusters in addition to the markers on anatomical landmarks. [13] As a result of the variety of marker sets in use the number of markers ranges from 12 to 36 in the mentioned studies.…”

Section: Introductionmentioning

confidence: 99%

“…The majority of the above mentioned studies applied Principal Component Analysis (PCA) for feature extraction and dimensionality reduction. Recently, self-organizing maps have also been used [7] but PCA has been favored mainly because it is particularly powerful on high redundancy data such as kinematic gait data. For classification methods, the linear Support Vector Machine (SVM) algorithm has been used in several approaches.…”

Section: Introductionmentioning

confidence: 99%

“…For classification methods, the linear Support Vector Machine (SVM) algorithm has been used in several approaches. [7][8][9][10][11] Other state of the art classifiers such as Ada Boost, [15] Naïve Bayes [14] or Random Forest [16] are capable of solving non-linear classification problems and are commonly considered in machine learning applications.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Comparison of different marker sets for marker trajectory and principal component analysis based classification of simulated gait impairments

Christian

Kluge

Eskofier

et al. 2016

JBEI

View full text Add to dashboard Cite

Objective: Many different marker sets have been used in marker trajectory based gait classification approaches. Little knowledge exists about the effects of specific marker sets on the subsequent statistical modeling. Such analysis is often based on principal component analysis. The aim of this study was to test the effect of marker set choice on marker trajectory and principal component analysis based gait classification. Methods: This study tested the performance of principal component analysis based gait classification models with various marker sets on the basis of simulated gait impairments. Simulated gait impairments were used to enable a high level of control of the gait patterns. Results: Classification accuracies were similar across most tested marker sets. Improved performance could be detected for some marker sets depending on the type of impairment. Conclusion: Several potentially valid marker sets exist for a specific gait classification task even though trends could be found suggesting that optimal marker set choice is dependent on functional aspects of the movement.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Comparison of different marker sets for marker trajectory and principal component analysis based classification of simulated gait impairments

Christian

Kluge

Eskofier

et al. 2016

JBEI

View full text Add to dashboard Cite

show abstract

Self-Organized Maps for the Analysis of the Biomechanical Response of the Knee Joint During Squat-Like Movements in Subjects Without Physical Conditioning

Molano

Jaramillo-Isaza

Orjuela-Cañón

2019

Communications in Computer and Information Science

View full text Add to dashboard Cite

Running-Related Injury Incidence: Does It Correlate with Kinematic Sub-groups of Runners? A Scoping Review

Adamson,

Vandamme,

Prior

et al. 2024

Sports Med

View full text Add to dashboard Cite

Background Historically, kinematic measures have been compared across injured and non-injured groups of runners, failing to take into account variability in kinematic patterns that exist independent of injury, and resulting in false positives. Research led by gait patterns and not pre-defined injury status is called for, to better understand running-related injury (RRI) aetiology and within- and between-group variability. Objectives Synthesise evidence for the existence of distinct kinematic sub-groups across a population of injured and healthy runners and assess between-group variability in kinematics, demographics and injury incidence. Data Sources Electronic database search: PubMed, Web of Science, Cochrane Central Register of Controlled Trials (Wiley), Embase, OVID, Scopus. Eligibility Criteria Original, peer-reviewed, research articles, published from database start to August 2022 and limited to English language were searched for quantitative and mixed-methods full-text studies that clustered injured runners according to kinematic variables. Results Five studies (n = 690) were included in the review. All studies detected the presence of distinct kinematic sub-groups of runners through cluster analysis. Sub-groups were defined by multiple differences in hip, knee and foot kinematics. Sex, step rate and running speed also varied significantly between groups. Random injury dispersal across sub-groups suggests no strong evidence for an association between kinematic sub-groups and injury type or location. Conclusion Sub-groups containing homogeneous gait patterns exist across healthy and injured populations of runners. It is likely that a single injury may be represented by multiple movement patterns, and therefore kinematics may not predict injury risk. Research to better understand the underlying causes of kinematic variability, and their associations with RRI, is warranted.

show abstract

Defining functional groups based on running kinematics using Self-Organizing Maps and Support Vector Machines

Cited by 39 publications

References 35 publications

Comparison of different marker sets for marker trajectory and principal component analysis based classification of simulated gait impairments

Comparison of different marker sets for marker trajectory and principal component analysis based classification of simulated gait impairments

Self-Organized Maps for the Analysis of the Biomechanical Response of the Knee Joint During Squat-Like Movements in Subjects Without Physical Conditioning

Running-Related Injury Incidence: Does It Correlate with Kinematic Sub-groups of Runners? A Scoping Review

Contact Info

Product

Resources

About