Optimisation of a Machine Learning Algorithm in Human Locomotion Using Principal Component and Discriminant Function Analyses

Overview

Journal PLoS One

Specialties General Medicine
Science

Date 2017 Sep 9

PMID 28886059

Citations 4

Authors

Maria Bisele

Martin Bencsik

Martin G C Lewis

Cleveland T Barnett

Affiliations

Soon will be listed here.

Abstract

Assessment methods in human locomotion often involve the description of normalised graphical profiles and/or the extraction of discrete variables. Whilst useful, these approaches may not represent the full complexity of gait data. Multivariate statistical methods, such as Principal Component Analysis (PCA) and Discriminant Function Analysis (DFA), have been adopted since they have the potential to overcome these data handling issues. The aim of the current study was to develop and optimise a specific machine learning algorithm for processing human locomotion data. Twenty participants ran at a self-selected speed across a 15m runway in barefoot and shod conditions. Ground reaction forces (BW) and kinematics were measured at 1000 Hz and 100 Hz, respectively from which joint angles (°), joint moments (N.m.kg-1) and joint powers (W.kg-1) for the hip, knee and ankle joints were calculated in all three anatomical planes. Using PCA and DFA, power spectra of the kinematic and kinetic variables were used as a training database for the development of a machine learning algorithm. All possible combinations of 10 out of 20 participants were explored to find the iteration of individuals that would optimise the machine learning algorithm. The results showed that the algorithm was able to successfully predict whether a participant ran shod or barefoot in 93.5% of cases. To the authors' knowledge, this is the first study to optimise the development of a machine learning algorithm.

Citing Articles

Using Lower Limb Wearable Sensors to Identify Gait Modalities: A Machine-Learning-Based Approach.

Hughes L, Bencsik M, Bisele M, Barnett C Sensors (Basel). 2023; 23(22).

PMID: 38005627 PMC: 10675053. DOI: 10.3390/s23229241.

Semi-Automatic Prostate Segmentation From Ultrasound Images Using Machine Learning and Principal Curve Based on Interpretable Mathematical Model Expression.

Peng T, Tang C, Wu Y, Cai J Front Oncol. 2022; 12:878104.

PMID: 35747834 PMC: 9209717. DOI: 10.3389/fonc.2022.878104.

Automatic Classification of Barefoot and Shod Populations Based on the Foot Metrics and Plantar Pressure Patterns.

Xiang L, Gu Y, Mei Q, Wang A, Shim V, Fernandez J Front Bioeng Biotechnol. 2022; 10:843204.

PMID: 35402419 PMC: 8984198. DOI: 10.3389/fbioe.2022.843204.

Machine Learning Analysis for Quantitative Discrimination of Dried Blood Droplets.

Hamadeh L, Imran S, Bencsik M, Sharpe G, Johnson M, Fairhurst D Sci Rep. 2020; 10(1):3313.

PMID: 32094359 PMC: 7040018. DOI: 10.1038/s41598-020-59847-x.

References

Ardestani M, Ferrigno C, Moazen M, Wimmer M . From normal to fast walking: Impact of cadence and stride length on lower extremity joint moments. Gait Posture. 2016; 46:118-25. DOI: 10.1016/j.gaitpost.2016.02.005. View

Kerrigan D, Franz J, Keenan G, Dicharry J, Croce U, Wilder R . The effect of running shoes on lower extremity joint torques. PM R. 2009; 1(12):1058-63. DOI: 10.1016/j.pmrj.2009.09.011. View

Eskofier B, Federolf P, Kugler P, Nigg B . Marker-based classification of young-elderly gait pattern differences via direct PCA feature extraction and SVMs. Comput Methods Biomech Biomed Engin. 2011; 16(4):435-42. DOI: 10.1080/10255842.2011.624515. View

Barton G, Hawken M, Scott M, Schwartz M . Movement deviation profile: a measure of distance from normality using a self-organizing neural network. Hum Mov Sci. 2010; 31(2):284-94. DOI: 10.1016/j.humov.2010.06.003. View

Phinyomark A, Hettinga B, Osis S, Ferber R . Do intermediate- and higher-order principal components contain useful information to detect subtle changes in lower extremity biomechanics during running?. Hum Mov Sci. 2015; 44:91-101. DOI: 10.1016/j.humov.2015.08.018. View

Begg R, Kamruzzaman J . A machine learning approach for automated recognition of movement patterns using basic, kinetic and kinematic gait data. J Biomech. 2005; 38(3):401-8. DOI: 10.1016/j.jbiomech.2004.05.002. View

Schutte L, Narayanan U, Stout J, Selber P, Gage J, Schwartz M . An index for quantifying deviations from normal gait. Gait Posture. 2000; 11(1):25-31. DOI: 10.1016/s0966-6362(99)00047-8. View

Dillmann U, Holzhoffer C, Johann Y, Bechtel S, Graber S, Massing C . Principal Component Analysis of gait in Parkinson's disease: relevance of gait velocity. Gait Posture. 2013; 39(3):882-7. DOI: 10.1016/j.gaitpost.2013.11.021. View

Cretual A, Bervet K, Ballaz L . Gillette Gait Index in adults. Gait Posture. 2010; 32(3):307-10. DOI: 10.1016/j.gaitpost.2010.05.015. View

10.

Baker R, McGinley J, Schwartz M, Beynon S, Rozumalski A, Graham H . The gait profile score and movement analysis profile. Gait Posture. 2009; 30(3):265-9. DOI: 10.1016/j.gaitpost.2009.05.020. View

11.

Ali S, Shah M . Human action recognition in videos using kinematic features and multiple instance learning. IEEE Trans Pattern Anal Mach Intell. 2010; 32(2):288-303. DOI: 10.1109/TPAMI.2008.284. View

12.

Sugavaneswaran L, Umapathy K, Krishnan S . Ambiguity domain-based identification of altered gait pattern in ALS disorder. J Neural Eng. 2012; 9(4):046004. DOI: 10.1088/1741-2560/9/4/046004. View

13.

Leardini A, Biagi F, Merlo A, Belvedere C, Benedetti M . Multi-segment trunk kinematics during locomotion and elementary exercises. Clin Biomech (Bristol). 2011; 26(6):562-71. DOI: 10.1016/j.clinbiomech.2011.01.015. View

14.

von Tscharner V, Enders H, Maurer C . Subspace identification and classification of healthy human gait. PLoS One. 2013; 8(7):e65063. PMC: 3704597. DOI: 10.1371/journal.pone.0065063. View

15.

Divert C, Mornieux G, Baur H, Mayer F, Belli A . Mechanical comparison of barefoot and shod running. Int J Sports Med. 2005; 26(7):593-8. DOI: 10.1055/s-2004-821327. View

16.

Kobsar D, Osis S, Hettinga B, Ferber R . Gait Biomechanics and Patient-Reported Function as Predictors of Response to a Hip Strengthening Exercise Intervention in Patients with Knee Osteoarthritis. PLoS One. 2015; 10(10):e0139923. PMC: 4596804. DOI: 10.1371/journal.pone.0139923. View

17.

Jeon H, Han J, Yi W, Jeon B, Park K . Classification of Parkinson gait and normal gait using Spatial-Temporal Image of Plantar pressure. Annu Int Conf IEEE Eng Med Biol Soc. 2009; 2008:4672-5. DOI: 10.1109/IEMBS.2008.4650255. View

18.

Novacheck T, Trost J, Sohrweide S . Examination of the child with cerebral palsy. Orthop Clin North Am. 2010; 41(4):469-88. DOI: 10.1016/j.ocl.2010.07.001. View

19.

Chau T . A review of analytical techniques for gait data. Part 2: neural network and wavelet methods. Gait Posture. 2001; 13(2):102-20. DOI: 10.1016/s0966-6362(00)00095-3. View

20.

Reid S, Graham R, Costigan P . Differentiation of young and older adult stair climbing gait using principal component analysis. Gait Posture. 2009; 31(2):197-203. DOI: 10.1016/j.gaitpost.2009.10.005. View