Medical & Biological Engineering & Computing

, Volume 55, Issue 4, pp 685–695 | Cite as

Selection of clinical features for pattern recognition applied to gait analysis

Original Article
First Online:
Received:
Accepted:

Abstract

This paper deals with the opportunity of extracting useful information from medical data retrieved directly from a stereophotogrammetric system applied to gait analysis. A feature selection method to exhaustively evaluate all the possible combinations of the gait parameters is presented, in order to find the best subset able to classify among diseased and healthy subjects. This procedure will be used for estimating the performance of widely used classification algorithms, whose performance has been ascertained in many real-world problems with respect to well-known classification benchmarks, both in terms of number of selected features and classification accuracy. Precisely, support vector machine, Naive Bayes and K nearest neighbor classifiers can obtain the lowest classification error, with an accuracy greater than 97 %. For the considered classification problem, the whole set of features will be proved to be redundant and it can be significantly pruned. Namely, groups of 3 or 5 features only are able to preserve high accuracy when the aim is to check the anomaly of a gait. The step length and the swing speed are the most informative features for the gait analysis, but also cadence and stride may add useful information for the movement evaluation.

Keywords

Gait analysis Pattern recognition Feature selection Classification 
This is a preview of subscription content, log in to check access.

Notes

Compliance with ethical standards

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

References

  1. 1.
    Ahlrichs C, Samà A, Lawo M, Cabestany J, Rodríguez-Martín D, Pérez-López C, Sweeney D, Quinlan LR, Laighin GÒ, Counihan T et al (2016) Detecting freezing of gait with a tri-axial accelerometer in Parkinson's disease patients. Med Biol Eng Comput 54(1):223–233CrossRefPubMedGoogle Scholar
  2. 2.
    Almuallim H, Dietterich TG (1994) Learning boolean concepts in the presence of many irrelevant features. Artif Intell 69(1–2):279–305CrossRefGoogle Scholar
  3. 3.
    Altilio R, Liparulo L, Panella M, Proietti A, Paoloni M (2015) Multimedia and gaming technologies for telerehabilitation of motor disabilities [leading edge]. IEEE Technol Soc Mag 34(4):23–30. doi: 10.1109/MTS.2015.2494279 CrossRefGoogle Scholar
  4. 4.
    Altman DG, Bland JM (1994) Diagnostic tests. 1: sensitivity and specificity. BMJ Br Med J 308(6943):1552CrossRefGoogle Scholar
  5. 5.
    Ayachi F, Boudaoud S, Marque C (2014) Evaluation of muscle force classification using shape analysis of the semg probability density function: a simulation study. Med Biol Eng Comput 52(8):673–684CrossRefPubMedGoogle Scholar
  6. 6.
    Begg R, Kamruzzaman J (2005) A machine learning approach for automated recognition of movement patterns using basic, kinetic and kinematic gait data. J Biomech 38(3):401–408CrossRefPubMedGoogle Scholar
  7. 7.
    Bhanu B, Lee S, Ming J (1995) Adaptive image segmentation using a genetic algorithm. IEEE Trans Syst Man Cybern 25(12):1543–1567CrossRefGoogle Scholar
  8. 8.
    Broadhurst D, Goodacre R, Jones A, Rowland JJ, Kell DB (1997) Genetic algorithms as a method for variable selection in multiple linear regression and partial least squares regression, with applications to pyrolysis mass spectrometry. Anal Chim Acta 348(1):71–86CrossRefGoogle Scholar
  9. 9.
    Chester VL, Biden EN, Tingley M (2005) Gait analysis. Biomed Instrum Technol 39(1):64–74PubMedGoogle Scholar
  10. 10.
    Cho CW, Chao WH, Lin SH, Chen YY (2009) A vision-based analysis system for gait recognition in patients with parkinsons disease. Expert Syst Appl 36(3):7033–7039CrossRefGoogle Scholar
  11. 11.
    Chun DN, Yang HS (1996) Robust image segmentation using genetic algorithm with a fuzzy measure. Pattern Recogn 29(7):1195–1211CrossRefGoogle Scholar
  12. 12.
    Congalton RG (1991) A review of assessing the accuracy of classifications of remotely sensed data. Remote Sens Environ 37(1):35–46CrossRefGoogle Scholar
  13. 13.
    Cover TM, Hart PE (1967) Nearest neighbor pattern classification. IEEE Trans Inf Theory 13(1):21–27CrossRefGoogle Scholar
  14. 14.
    Dikovski B, Madjarov G, Gjorgjevikj D (2014) Evaluation of different feature sets for gait recognition using skeletal data from kinect. In: 2014 37th international convention on information and communication technology, electronics and microelectronics (MIPRO). IEEE, pp 1304–1308Google Scholar
  15. 15.
    Ertuğrul ÖF, Kaya Y, Tekin R (2015) A novel approach for semg signal classification with adaptive local binary patterns. Med Biol Eng Comput 1–10Google Scholar
  16. 16.
    Exell T, Freeman C, Meadmore K, Kutlu M, Rogers E, Hughes AM, Hallewell E, Burridge J (2013) Goal orientated stroke rehabilitation utilising electrical stimulation, iterative learning and Microsoft Kinect. In: 2013 IEEE international conference on rehabilitation robotics (ICORR), pp 1–6. doi: 10.1109/ICORR.2013.6650493
  17. 17.
    Fang J, Hunt KJ, Xie L, Yang GY (2015) Modelling of the toe trajectory during normal gait using circle-fit approximation. Med Biol Eng Comput 1–9. doi: 10.1007/s11517-015-1414-4
  18. 18.
    Fawcett T (2006) An introduction to ROC analysis. Pattern Recogn Lett 27(8):861–874CrossRefGoogle Scholar
  19. 19.
    Fisher RA (1938) The statistical utilization of multiple measurements. Ann Eugen 8(4):376–386CrossRefGoogle Scholar
  20. 20.
    Friedman J, Hastie T, Tibshirani R (2001) The elements of statistical learning. Springer series in statistics, vol 1. Springer, BerlinGoogle Scholar
  21. 21.
    Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182Google Scholar
  22. 22.
    Ibrahim S, Chowriappa P, Dua S, Acharya UR, Noronha K, Bhandary S, Mugasa H (2015) Classification of diabetes maculopathy images using data-adaptive neuro-fuzzy inference classifier. Med Biol Eng Comput 53(12):1345–1360CrossRefPubMedGoogle Scholar
  23. 23.
    Joshi CD, Lahiri U, Thakor NV (2013) Classification of gait phases from lower limb emg: application to exoskeleton orthosis. In: Point-of-care healthcare technologies (PHT). IEEE, pp 228–231Google Scholar
  24. 24.
    Joyseeree R, Sabha RA, Mueller H (2014) Applying machine learning to gait analysis data for disease identification. Stud Health Technol Inf 210:850–854Google Scholar
  25. 25.
    Krishnan C, Washabaugh EP, Seetharaman Y (2015) A low cost real-time motion tracking approach using webcam technology. J Biomech 48(3):544–548CrossRefPubMedGoogle Scholar
  26. 26.
    Krzanowski W (2000) Principles of multivariate analysis. Oxford University Press, OxfordGoogle Scholar
  27. 27.
    Kwak N, Choi CH (2002) Input feature selection for classification problems. IEEE Trans Neural Netw 13(1):143–159CrossRefPubMedGoogle Scholar
  28. 28.
    Lauer RT, Smith BT, Coiro D, Betz RR, McCarthy J (2004) Feasibility of gait event detection using intramuscular electromyography in the child with cerebral palsy. Neuromodul Technol Neural Interface 7(3):205–213CrossRefGoogle Scholar
  29. 29.
    Leardi R, Gonzalez AL (1998) Genetic algorithms applied to feature selection in pls regression: how and when to use them. Chemom Intell Lab Syst 41(2):195–207CrossRefGoogle Scholar
  30. 30.
    Lin SH, Chen SW, Lo YC, Lai HY, Yang CH, Chen SY, Chang YJ, Chen CH, Huang WT, Jaw FS et al. (2016) Quantitative measurement of Parkinsonian gait from walking in monocular image sequences using a centroid tracking algorithm. Med Biol Eng Comput 54(2–3):485–496CrossRefPubMedGoogle Scholar
  31. 31.
    Liu H, Motoda H (2012) Feature selection for knowledge discovery and data mining, vol 454. Springer, New YorkGoogle Scholar
  32. 32.
    Maisto M, Panella M, Liparulo L, Proietti A (2013) An accurate algorithm for the identification of fingertips using an RGB-D camera. IEEE J Emerg Sel Top Circuits Syst 3(2):272–283. doi: 10.1109/JETCAS.2013.2256830 CrossRefGoogle Scholar
  33. 33.
    Mangone M, Scettri P, Paoloni M, Procaccianti R, Spadaro A, Santilli V (2011) Pelvis-shoulder coordination during level walking in patients with ankylosing spondylitis. Gait Posture 34(1):1–5CrossRefPubMedGoogle Scholar
  34. 34.
    Mazzone P, Paoloni M, Mangone M, Santilli V, Insola A, Fini M, Scarnati E (2014) Unilateral deep brain stimulation of the pedunculopontine tegmental nucleus in idiopathic parkinsons disease: effects on gait initiation and performance. Gait Posture 40(3):357–362CrossRefPubMedGoogle Scholar
  35. 35.
    Møller MF (1993) A scaled conjugate gradient algorithm for fast supervised learning. Neural Netw 6(4):525–533CrossRefGoogle Scholar
  36. 36.
    Muro-de-la Herran A, Garcia-Zapirain B, Mendez-Zorrilla A (2014) Gait analysis methods: an overview of wearable and non-wearable systems, highlighting clinical applications. Sensors 14(2):3362–3394CrossRefPubMedPubMedCentralGoogle Scholar
  37. 37.
    Narendra PM, Fukunaga K (1977) A branch and bound algorithm for feature subset selection. IEEE Trans Comput 100(9):917–922CrossRefGoogle Scholar
  38. 38.
    O’Malley MJ, Abel MF, Damiano DL, Vaughan CL (1997) Fuzzy clustering of children with cerebral palsy based on temporal-distance gait parameters. IEEE Trans Rehabil Eng 5(4):300–309CrossRefPubMedGoogle Scholar
  39. 39.
    Panella M, Rizzi A, Martinelli G (2003) Refining accuracy of environmental data prediction by MoG neural networks. Neurocomputing 55(3–4):521–549. doi: 10.1016/S0925-2312(03)00392-8 CrossRefGoogle Scholar
  40. 40.
    Panella M (2012) A hierarchical procedure for the synthesis of ANFIS networks. Adv Fuzzy Syst 2012:1–12. doi: 10.1155/2012/491237 CrossRefGoogle Scholar
  41. 41.
    Panella M, Martinelli G (2011) Neural networks with quantum architecture and quantum learning. Int J Circuit Theory Appl 39(1):61–77. doi: 10.1002/cta.619 CrossRefGoogle Scholar
  42. 42.
    Papaleo E, Zollo L, Garcia-Aracil N, Badesa F, Morales R, Mazzoleni S, Sterzi S, Guglielmelli E (2015) Upper-limb kinematic reconstruction during stroke robot-aided therapy. Med Biol Eng Comput 53(9):815–828CrossRefPubMedGoogle Scholar
  43. 43.
    Patterson SL, Forrester LW, Rodgers MM, Ryan AS, Ivey FM, Sorkin JD, Macko RF (2007) Determinants of walking function after stroke: differences by deficit severity. Arch Phys Med Rehabil 88(1):115–119CrossRefPubMedGoogle Scholar
  44. 44.
    Pereira T, Paiva JS, Correia C, Cardoso J (2016) An automatic method for arterial pulse waveform recognition using KNN and SVM classifiers. Med Biol Eng Comput 54(7):1049–1059CrossRefPubMedGoogle Scholar
  45. 45.
    Piramuthu S (2004) Evaluating feature selection methods for learning in data mining applications. Eur J Oper Res 156(2):483–494CrossRefGoogle Scholar
  46. 46.
    Pradhan C, Wuehr M, Akrami F, Neuhaeusser M, Huth S, Brandt T, Jahn K, Schniepp R (2015) Automated classification of neurological disorders of gait using spatio-temporal gait parameters. J Electromyogr Kinesiol 25(2):413–422CrossRefPubMedGoogle Scholar
  47. 47.
    Proietti A, Panella M, Leccese F, Svezia E (2015) Dust detection and analysis in museum environment based on pattern recognition. Measurement 66:62–72. doi: 10.1016/j.measurement.2015.01.019 CrossRefGoogle Scholar
  48. 48.
    Purushotham S, Tripathy B (2012) Evaluation of classifier models using stratified tenfold cross validation techniques. In: Krishna PV, Babu MR, Ariwa E (eds) Global trends in information systems and software applications, Springer, Berlin, Heidelberg, pp 680–690. doi: 10.1007/978-3-642-29216-3_74
  49. 49.
    Rish I (2001) An empirical study of the naive bayes classifier. In: IJCAI 2001 workshop on empirical methods in artificial intelligence, vol 3, no 22. IBM, New York, pp 41–46Google Scholar
  50. 50.
    Rizzi A, Panella M, Mascioli FF, Martinelli G (2000) A recursive algorithm for fuzzy Min-Max networks. In: Proceedings of international joint conference on neural networks (IJCNN 2000), vol 6, pp 541–546. doi: 10.1109/IJCNN.2000.859451
  51. 51.
    Roerdink M, Lamoth CJ, Beek PJ et al (2008) Online gait event detection using a large force platform embedded in a treadmill. J Biomech 41(12):2628–2632CrossRefPubMedGoogle Scholar
  52. 52.
    Rogati M, Yang Y (2002) High-performing feature selection for text classification. In: Proceedings of the eleventh international conference on Information and knowledge management. ACM, pp 659–661Google Scholar
  53. 53.
    Rokach L, Maimon O (2014) Data mining with decision trees: theory and applications. World Scientific, SingaporeCrossRefGoogle Scholar
  54. 54.
    Saeys Y, Inza I, Larrañaga P (2007) A review of feature selection techniques in bioinformatics. Bioinformatics 23(19):2507–2517CrossRefPubMedGoogle Scholar
  55. 55.
    Saeys Y, Abeel T, Van de Peer Y (2008) Robust feature selection using ensemble feature selection techniques. In: Daelemans W, Goethals B, Morik K (eds) Machine learning and knowledge discovery in databases. Springer, Berlin, Heidelberg, pp 313–325. doi: 10.1007/978-3-540-87481-2_21
  56. 56.
    Scardapane S, Wang D, Panella M (2016) A decentralized training algorithm for echo state networks in distributed big data applications. Neural Netw 78:65–74. doi: 10.1016/j.neunet.2015.07.006 CrossRefPubMedGoogle Scholar
  57. 57.
    Schmid A, Duncan PW, Studenski S, Lai SM, Richards L, Perera S, Wu SS (2007) Improvements in speed-based gait classifications are meaningful. Stroke 38(7):2096–2100CrossRefPubMedGoogle Scholar
  58. 58.
    Shirakawa T, Sugiyama N, Sato H, Sakurai K, Sato E (2015) Gait analysis and machine learning classification on healthy subjects in normal walking. Int J Parallel Emerg Distrib Syst 1–10. doi: 10.1080/17445760.2015.1044007 Google Scholar
  59. 59.
    Wang X, Yang J, Teng X, Xia W, Jensen R (2007) Feature selection based on rough sets and particle swarm optimization. Pattern Recogn Lett 28(4):459–471CrossRefGoogle Scholar
  60. 60.
    Wu J, Wang J, Liu L (2007) Feature extraction via KPCA for classification of gait patterns. Hum Mov Sci 26(3):393–411CrossRefPubMedGoogle Scholar
  61. 61.
    Xia Y, Gao Q, Lu Y, Ye Q (2015) A novel approach for analysis of altered gait variability in amyotrophic lateral sclerosis. Med Biol Eng Comput 1–10. doi:  10.1007/s11517-015-1413-5 PubMedGoogle Scholar
  62. 62.
    Yang Y, Pedersen JO (1997) A comparative study on feature selection in text categorization. ICML 97:412–420Google Scholar
  63. 63.
    Yang M, Zheng H, Wang H, McClean S (2009) Feature selection and construction for the discrimination of neurodegenerative diseases based on gait analysis. In: 3rd international conference on pervasive computing technologies for healthcare. Pervasive health 2009. IEEE, pp 1–7Google Scholar
  64. 64.
    Zeng W, Wang C (2015) Classification of neurodegenerative diseases using gait dynamics via deterministic learning. Inf Sci 317:246–258CrossRefGoogle Scholar
  65. 65.
    Zeni JA, Higginson JS (2009) Differences in gait parameters between healthy subjects and persons with moderate and severe knee osteoarthritis: a result of altered walking speed? Clin Biomech 24(4):372–378CrossRefGoogle Scholar
  66. 66.
    Zhang Z, Liparulo L, Panella M, Gu X, Fang Q (2016) A fuzzy kernel motion classifier for autonomous stroke rehabilitation. IEEE J Biomed Health Inf 20(3):893–901. doi: 10.1109/JBHI.2015.2430524 CrossRefGoogle Scholar

Copyright information

© International Federation for Medical and Biological Engineering 2016

Authors and Affiliations

  1. 1.Department of Information Engineering, Electronics and Telecommunications (DIET)University of Rome “La Sapienza”RomeItaly
  2. 2.Biomechanics and Movement Analysis Laboratory, Physical Medicine and RehabilitationUniversity of Rome “La Sapienza”RomeItaly

Personalised recommendations