Advertisement

Medical & Biological Engineering & Computing

, Volume 57, Issue 12, pp 2693–2703 | Cite as

Machine learning approach to predict center of pressure trajectories in a complete gait cycle: a feedforward neural network vs. LSTM network

  • Ahnryul Choi
  • Hyunwoo Jung
  • Ki Young Lee
  • Sangsik Lee
  • Joung Hwan MunEmail author
Original Article
  • 145 Downloads

Abstract

Center of pressure (COP) trajectories of human can maintain regulation of forward progression and stability of lateral sway during walking. The insole pressure system can only detect COP trajectories of each foot during single stance. In this study, we developed artificial neural network models that could present COP trajectories in an integrated coordinate system during a complete gait cycle using pressure information of the insole system. A feed forward artificial neural network (FFANN) and a long short-term memory (LSTM) model were developed. For FFANN, among 198 pressure sensors from Pedar-X insoles, proper input variables were selected using sequential forward selection to reduce input dimension. The LSTM model used all 198 signals as inputs because of its self-learning characteristic. As results of cross-validation, the FFANN model showed correlation coefficients of 0.98–0.99 and 0.93–0.95 in anterior/posterior and medial/lateral directions, respectively. For the LSTM model, correlation coefficients were similar to those of FFANN. However, the relative root mean square error (12.5%) of the FFANN model was higher than that (9.8%) of the LSTM model in medial/lateral direction (p = 0.03). This study can be used for quantitative evaluation of clinical diagnosis and rehabilitation status for patient with various diseases through further training using varied databases.

Graphical abstract

Architectures of neural networks developed in this study (a feed forward artificial neural network; b LSTM network)

Keywords

Center of pressure Gait Neural network LSTM Insole system 

Notes

Funding information

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (grant no. 2017R1D1A3B03033675).

References

  1. 1.
    Choi A, Kang TG, Mun JH (2016) Biomechanical evaluation of dynamic balance control ability during golf swing. J Med Biol Eng 36:430–439CrossRefGoogle Scholar
  2. 2.
    Chesnin KJ, Selby-Silverstein L, Besser MP (2000) Comparison of in-shoe pressure measurement device to a force plate: concurrent validity of center of pressure measurements. Gait Posture 12:128–133PubMedCrossRefGoogle Scholar
  3. 3.
    Chiu MC, We HC, Chang LY (2013) Gait speed and gender effects on center of pressure progression during normal walking. Gait Posture 37:43–48PubMedCrossRefPubMedCentralGoogle Scholar
  4. 4.
    Chisholm AE, Perry SD, Mcllroy WE (2001) Inter-limb centre of pressure symmetry during gait among stroke survivors. Gait Posture 33:238–243CrossRefGoogle Scholar
  5. 5.
    Choi A, Sim T, Mun JH (2016) Improved determination of dynamic balance using the centre of mass and centre of pressure inclination variables in a complete golf swing cycle. J Sports Sci 34:906–914PubMedCrossRefPubMedCentralGoogle Scholar
  6. 6.
    Hsue BJ, Miller F, Su FC (2009) The dynamic balance of the children with cerebral palsy and typical developing during gait. Part I: Spatial relationship between COM and COP trajectories. Gait Posture 29:465–470PubMedCrossRefGoogle Scholar
  7. 7.
    Koldenhoven RM, Feger MA, Fraser JJ, Hertel J (2018) Variability in center of pressure position and muscle activation during walking with chronic ankle instability. J Electromyogr Kinesiol 38:155–161PubMedCrossRefPubMedCentralGoogle Scholar
  8. 8.
    Hallemans A, Verbecque E, Duman R, Cheze L, Van Hamme A, Robert T (2018) Developmental changes in spatial margin of stability in typically developing children relate to the mechanics of gait. Gait Posture 63:33–38PubMedCrossRefGoogle Scholar
  9. 9.
    Huang H, Qiu J, Liu T, Yu Y, Guo Q, Luo D, Ao Y (2017) Similarity of center of pressure progression during walking and jogging of anterior cruciate ligament deficient patients. PLoS One 12:e0169421PubMedPubMedCentralCrossRefGoogle Scholar
  10. 10.
    Klamroth S, Steib S, Babner H, Gobler J, Winkler J, Eskofier B, Klucken J, Pfeifer K (2016) Immediate effects of perturbation treadmill training on gait and postural control in patients with Parkinson’s disease. Gait Posture 50:102–108PubMedCrossRefPubMedCentralGoogle Scholar
  11. 11.
    Huang PY, Lin CF, Juo LC, Liao JC (2011) Foot pressure and center of pressure in athletes with ankle instability during lateral shuffling and running gait. Scandi J Med Sci Sport 21:e461–e467CrossRefGoogle Scholar
  12. 12.
    Haim A, Rozen N, Wolf A (2010) The influence of sagittal center of pressure offset on gait kinematics and kinetics. J Biomech 43:969–977PubMedCrossRefPubMedCentralGoogle Scholar
  13. 13.
    Razak AH, Zayegh A, Begg RK, Wahab Y (2012) Foot plantar pressure measurement system: a review. Sensors 12:9884–9912PubMedCrossRefPubMedCentralGoogle Scholar
  14. 14.
    Choi A, Lee JM, Mun JH (2013) Ground reaction forces predicted by using artificial neural network during asymmetric movements. Int J Precis Eng Manuf 14:475–483CrossRefGoogle Scholar
  15. 15.
    Choi A, Jung H, Mun JH (2019) Single inertial sensor-based neural networks to estimate COM-COP inclination angle during walking. Sensors 19:2974CrossRefGoogle Scholar
  16. 16.
    McPoil TG, Cornwall MW, Yamada W (1995) A comparison of two in-shoe plantar pressure measurement systems. Low Extrem 2:95–103Google Scholar
  17. 17.
    Liedtke C, Fokkenrood SA, Menger JT, van der Kooij H, Veltink PH (2007) Evaluation of instrumented shoes for ambulatory assessment of ground reaction forces. Gait Posture 26:39–47PubMedCrossRefPubMedCentralGoogle Scholar
  18. 18.
    Orlin MN, McPoil TG (2000) Plantar pressure assessment. Phys Ther 80:399–409PubMedCrossRefPubMedCentralGoogle Scholar
  19. 19.
    Varrecchia T, De Marchis C, Rinaldi M, Draicchio F, Serrao M, Schmid M, Conforto S, Ranavolo A (2018) Lifting activity assessment using surface electromyographic features and neural networks. Int J Ind Ergon 66:1–9CrossRefGoogle Scholar
  20. 20.
    Kim TH, Choi A, Heo HM, Kim K, Lee K, Mun JH (2019) Machine learning-based pre-impact fall detection model to discriminate various types of fall. J Biomech Eng 141:081010CrossRefGoogle Scholar
  21. 21.
    Barton G, Lisboa P, Lees A, Attfield S (2007) Gait quality assessment using self-organising artificial neural networks. Gait Posture 25:347–349CrossRefGoogle Scholar
  22. 22.
    De Vriesl WHK, Veeger HEJ, Baten CTM, van der Helm FCT (2016) Can shoulder joint reaction forces be estimated by neural networks? J Biomech 49:73–79CrossRefGoogle Scholar
  23. 23.
    Joo SB, Oh SE, Sim T, Kim H, Choi CH, Koo H, Mun JH (2014) Prediction of gait speed from plantar pressure using artificial neural networks. Expert Syst Appl 41:7398–7405CrossRefGoogle Scholar
  24. 24.
    Mehrizi R, Peng X, Xu X, Zhang S, Li K (2019) A deep neural network-based method for estimation of 3D lifting motions. J Biomech 84:87–93PubMedCrossRefPubMedCentralGoogle Scholar
  25. 25.
    LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521:436–444PubMedPubMedCentralCrossRefGoogle Scholar
  26. 26.
    Hu B, Dixon PC, Jacobs JV, Dennerlein JT, Schiffman JM (2018) Machine learning algorithms based on signals from a single wearable inertial sensor can detect surface- and age-related differences in walking. J Biomech 71:37–42PubMedCrossRefGoogle Scholar
  27. 27.
    Hernandez V, Rezzoug N, Gorce P, Venture G (2018) Wheelchair propulsion: force orientation and amplitude prediction with recurrent neural network. J Biomech 78:166–171PubMedCrossRefPubMedCentralGoogle Scholar
  28. 28.
    Dao TT (2018) From deep learning to transfer learning for the prediction of skeletal muscle forces. Med Biol Eng Comput 57:1049–1058PubMedCrossRefPubMedCentralGoogle Scholar
  29. 29.
    Plotnik M, Bartsch RP, Zeev A, Giladi N, Hausdorff JM (2013) Effects of walking speed on asymmetry and bilateral coordination of gait. Gait Posture 38:864–869PubMedPubMedCentralCrossRefGoogle Scholar
  30. 30.
    Russel SD, Bennett BC, Kerrigan DC, Abel MF (2007) Determinants of gait as applied to children with cerebral palsy. Gait Posture 26:295–300CrossRefGoogle Scholar
  31. 31.
    Arjmand N, Ekrami O, Shirazi-Adl A, Plamondon A, Parnianpour M (2013) Relative performances of artificial neural network and regression mapping tools in evaluation of spinal loads and muscle forces during static lifting. J Biomech 46:1454–1462PubMedCrossRefPubMedCentralGoogle Scholar
  32. 32.
    Choi A, Yun TS, Suh SW, Yang JH, Park H, Lee S, Roh MS, Kang TG, Mun JH (2013) Determination of input variables for the development of a gait asymmetry expert system in patients with idiopathic scoliosis. Int J Precis Eng Manuf 14:811–818CrossRefGoogle Scholar
  33. 33.
    Hof AL, van Bockel RM, Schoppen T, Postema K (2007) Control of lateral balance in walking experimental findings in normal subjects and above-knee amputees. Gait Posture 25:250–258PubMedCrossRefPubMedCentralGoogle Scholar
  34. 34.
    Winter DA, Prince F, Frank JS, Powell C, Zabjek KF (1996) Unified theory regarding A/P and M/L balance in quiet stance. J Neurophysiol 75:2334–2343PubMedCrossRefPubMedCentralGoogle Scholar
  35. 35.
    Bennett BC, Abel MF, Wolovick A, Franklin T, Allaire PE, Kerrigan DC (2005) Center of mass movement and energy transfer during walking in children with cerebral palsy. Arch Phys Med Rehabil 86:2189–2194PubMedCrossRefPubMedCentralGoogle Scholar
  36. 36.
    Schollhorn WI (2004) Applications of artificial neural nets in clinical biomechanics. Clin Biomech 19:876–898CrossRefGoogle Scholar
  37. 37.
    Ardestani MM, Zhang X, Wang L, Lian Q, Liu Y, He J, Li D, Jin Z (2014) Human lower extremity joint moment prediction: a wavelet neural network approach. Expert Syst Appl 41:4422–4433CrossRefGoogle Scholar
  38. 38.
    Ngoh KJH, Gouwanda D, Gopalai AA, Chong YZ (2018) Estimation of vertical ground reaction force during running using neural network model and uniaxial accelerometer. J Biomech 76:269–273PubMedCrossRefPubMedCentralGoogle Scholar
  39. 39.
    Liu MM, Herzog W, Savelberg HHCM (1999) Dynamic muscle force prediction from EMG: an artificial neural network approach. J Electromyogr Kinesiol 9:391–400PubMedCrossRefPubMedCentralGoogle Scholar
  40. 40.
    Nweke HF, The YW, Al-Garadi MA, Alo UR (2018) Deep learning algorithms for human activity recognition using mobile and wearable sensor networks: state of the art and research challenge. Expert Syst Appl 105:233–261CrossRefGoogle Scholar
  41. 41.
    Capela NA, Lemaire ED, Baddour N (2015) Feature selection for wearable smartphone-based human activity recognition with able bodied, elderly, and stroke patients. PLoS One 10:e0124414PubMedPubMedCentralCrossRefGoogle Scholar

Copyright information

© International Federation for Medical and Biological Engineering 2019

Authors and Affiliations

  1. 1.Department of Biomedical Engineering, College of Medical ConvergenceCatholic Kwandong UniversityGangneungRepublic of Korea
  2. 2.Department of Bio-Mechatronic Engineering, College of Biotechnology and BioengineeringSungkyunkwan UniversitySuwonRepublic of Korea

Personalised recommendations