한국전문물리치료학회지 (Physical Therapy Korea)

  • 제30권2호
  • /
  • Pages.102-109
  • /
  • 2023
  • /
  • 1225-8962(pISSN)
  • /
  • 2287-982X(eISSN)
한국전문물리치료학회 (Korean Research Society of Physical Therapy)
권호 표지
DOI QR코드

DOI QR Code

Application of Decision Tree to Classify Fall Risk Using Inertial Measurement Unit Sensor Data and Clinical Measurements

  • Junwoo Park (Injury Prevention and Biomechanics Laboratory, Department of Physical Therapy, Yonsei University) ;
  • Jongwon Choi (Injury Prevention and Biomechanics Laboratory, Department of Physical Therapy, Yonsei University) ;
  • Seyoung Lee (Injury Prevention and Biomechanics Laboratory, Department of Physical Therapy, Yonsei University) ;
  • Kitaek Lim (Injury Prevention and Biomechanics Laboratory, Department of Physical Therapy, Yonsei University) ;
  • Woochol Joseph Choi (Injury Prevention and Biomechanics Laboratory, Department of Physical Therapy, Yonsei University)
  • 투고 : 2023.05.05
  • 심사 : 2023.05.08
  • 발행 : 2023.05.20
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Background: While efforts have been made to differentiate fall risk in older adults using wearable devices and clinical methodologies, technologies are still infancy. We applied a decision tree (DT) algorithm using inertial measurement unit (IMU) sensor data and clinical measurements to generate high performance classification models of fall risk of older adults. Objects: This study aims to develop a classification model of fall risk using IMU data and clinical measurements in older adults. Methods: Twenty-six older adults were assessed and categorized into high and low fall risk groups. IMU sensor data were obtained while walking from each group, and features were extracted to be used for a DT algorithm with the Gini index (DT1) and the Entropy index (DT2), which generated classification models to differentiate high and low fall risk groups. Model's performance was compared and presented with accuracy, sensitivity, and specificity. Results: Accuracy, sensitivity and specificity were 77.8%, 80.0%, and 66.7%, respectively, for DT1; and 72.2%, 91.7%, and 33.3%, respectively, for DT2. Conclusion: Our results suggest that the fall risk classification using IMU sensor data obtained during gait has potentials to be developed for practical use. Different machine learning techniques involving larger data set should be warranted for future research and development.

키워드

과제정보

This work was supported, in part, by the "Brain Korea 21 FOUR Project", the National Research Foundation of Korea (Award number: F21SH8303039) for Department of Physical Therapy in the Graduate School of Yonsei University, and by the "Regional Innovation Strategy (RIS)" through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (MOE) (2022RIS-005).

참고문헌

  1. Tinetti ME. Factors associated with serious injury during falls by ambulatory nursing home residents. J Am Geriatr Soc 1987;35(7):644-8. https://doi.org/10.1111/j.1532-5415.1987.tb04341.x
  2. Karlsson MK, Magnusson H, von Schewelov T, Rosengren BE. Prevention of falls in the elderly--a review. Osteoporos Int 2013;24(3):747-62. https://doi.org/10.1007/s00198-012-2256-7
  3. Fingerhut LA, Warner M. Injury chartbook. Health, United States, 1996-97. National Center for Health Statistics; 1997.
  4. Burns ER, Haddad YK, Parker EM. Primary care providers' discussion of fall prevention approaches with their older adult patients-DocStyles, 2014. Prev Med Rep 2018;9:149-52. https://doi.org/10.1016/j.pmedr.2018.01.016
  5. Robinovitch SN, Feldman F, Yang Y, Schonnop R, Leung PM, Sarraf T, et al. Video capture of the circumstances of falls in elderly people residing in long-term care: an observational study. Lancet 2013;381(9860):47-54. Erratum in: Lancet 2013;381(9860):28.
  6. Overstall PW, Exton-Smith AN, Imms FJ, Johnson AL. Falls in the elderly related to postural imbalance. Br Med J 1977;1(6056):261-4. https://doi.org/10.1136/bmj.1.6056.261
  7. Overstall PW, Johnson AL, Exton-Smith AN. Instability and falls in the elderly. Age Ageing 1978;Suppl:92-6.
  8. Fletcher PC, Hirdes JP. Risk factors for serious falls among community-based seniors: results from the national population health survey. Can J Aging 2002;21(1):103-16. https://doi.org/10.1017/S0714980800000684
  9. Chen LK, Woo J, Assantachai P, Auyeung TW, Chou MY, Iijima K, et al. Asian working group for sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. J Am Med Dir Assoc 2020;21(3):300-7.e2. https://doi.org/10.1016/j.jamda.2019.12.012
  10. Sanchez-Rodriguez D, Marco E, Miralles R, Guillen-Sola A, Vazquez-Ibar O, Escalada F, et al. Does gait speed contribute to sarcopenia case-finding in a postacute rehabilitation setting? Arch Gerontol Geriatr 2015;61(2):176-81. https://doi.org/10.1016/j.archger.2015.05.008
  11. Kim M, Won CW. Sarcopenia is associated with cognitive impairment mainly due to slow gait speed: results from the Korean frailty and aging cohort study (KFACS). Int J Environ Res Public Health 2019;16(9):1491.
  12. Bet P, Castro PC, Ponti MA. Fall detection and fall risk assessment in older person using wearable sensors: a systematic review. Int J Med Inform 2019;130:103946.
  13. Brodie MA, Lord SR, Coppens MJ, Annegarn J, Delbaere K. Eight-week remote monitoring using a freely worn device reveals unstable gait patterns in older fallers. IEEE Trans Biomed Eng 2015;62(11):2588-94. https://doi.org/10.1109/TBME.2015.2433935
  14. Zakaria NA, Kuwae Y, Tamura T, Minato K, Kanaya S. Quantitative analysis of fall risk using TUG test. Comput Methods Biomech Biomed Engin 2015;18(4):426-37. https://doi.org/10.1080/10255842.2013.805211
  15. Rivolta MW, Aktaruzzaman M, Rizzo G, Lafortuna CL, Ferrarin M, Bovi G, et al. Evaluation of the Tinetti score and fall risk assessment via accelerometry-based movement analysis. Artif Intell Med 2019;95:38-47. https://doi.org/10.1016/j.artmed.2018.08.005
  16. Lima CA, Ricci NA, Nogueira EC, Perracini MR. The Berg balance scale as a clinical screening tool to predict fall risk in older adults: a systematic review. Physiotherapy 2018;104(4):383-94. https://doi.org/10.1016/j.physio.2018.02.002
  17. Liu TW, Ng SSM. Assessing the fall risks of communitydwelling stroke survivors using the short-form physiological profile assessment (S-PPA). PLoS One 2019;14(5):e0216769.
  18. Castellini G, Demarchi A, Lanzoni M, Castaldi S. Fall prevention: is the STRATIFY tool the right instrument in Italian Hospital inpatient? A retrospective observational study. BMC Health Serv Res 2017;17(1):656.
  19. Oliver D, Britton M, Seed P, Martin FC, Hopper AH. Development and evaluation of evidence based risk assessment tool (STRATIFY) to predict which elderly inpatients will fall: casecontrol and cohort studies. BMJ 1997;315(7115):1049-53. https://doi.org/10.1136/bmj.315.7115.1049
  20. Lord SR, Menz HB, Tiedemann A. A physiological profile approach to falls risk assessment and prevention. Phys Ther 2003;83(3):237-52. https://doi.org/10.1093/ptj/83.3.237
  21. Senden R, Savelberg HH, Grimm B, Heyligers IC, Meijer K. Accelerometry-based gait analysis, an additional objective approach to screen subjects at risk for falling. Gait Posture 2012;36(2):296-300. https://doi.org/10.1016/j.gaitpost.2012.03.015
  22. Smee DJ, Berry HL, Waddington G, Anson J. Association between berg balance, physiological profile assessment and physical activity, physical function and body composition: a cross-sectional study. J Frailty Aging 2016;5(1):20-6.
  23. Brodie MA, Coppens MJ, Ejupi A, Gschwind YJ, Annegarn J, Schoene D, et al. Comparison between clinical gait and dailylife gait assessments of fall risk in older people. Geriatr Gerontol Int 2017;17(11):2274-82.
  24. Simon SR. Quantification of human motion: gait analysisbenefits and limitations to its application to clinical problems. J Biomech 2004;37(12):1869-80. https://doi.org/10.1016/j.jbiomech.2004.02.047
  25. Brodie MA, Coppens MJ, Lord SR, Lovell NH, Gschwind YJ, Redmond SJ, et al. Wearable pendant device monitoring using new wavelet-based methods shows daily life and laboratory gaits are different. Med Biol Eng Comput 2016;54(4):663-74. https://doi.org/10.1007/s11517-015-1357-9
  26. Tao W, Liu T, Zheng R, Feng H. Gait analysis using wearable sensors. Sensors (Basel) 2012;12(2):2255-83. https://doi.org/10.3390/s120202255
  27. Ponti M, Bet P, Oliveira CL, Castro PC. Better than counting seconds: Identifying fallers among healthy elderly using fusion of accelerometer features and dual-task Timed Up and Go. PLoS One 2017;12(4):e0175559.
  28. Engin M, Demirel A, Engin EZ, Fedakar M. Recent developments and trends in biomedical sensors. Measurement 2005;37(2):173-88. https://doi.org/10.1016/j.measurement.2004.11.002
  29. Bonato P. Wearable sensors/systems and their impact on biomedical engineering. IEEE Eng Med Biol Mag 2003;22(3):18-20. https://doi.org/10.1109/MEMB.2003.1213622
  30. Muro-de-la-Herran A, Garcia-Zapirain B, Mendez-Zorrilla A. Gait analysis methods: an overview of wearable and nonwearable systems, highlighting clinical applications. Sensors (Basel) 2014;14(2):3362-94. https://doi.org/10.3390/s140203362
  31. Hori K, Mao Y, Ono Y, Ora H, Hirobe Y, Sawada H, et al. Inertial measurement unit-based estimation of foot trajectory for clinical gait analysis. Front Physiol 2020;10:1530.
  32. Adachi W, Tsujiuchi N, Koizumi T, Shiojima K, Tsuchiya Y, Inoue Y. Calculation of joint reaction force and joint moments using by wearable walking analysis system. Annu Int Conf IEEE Eng Med Biol Soc 2012;2012:507-10.
  33. Saadeh W, Butt SA, Altaf MAB. A patient-specific single sensor IoT-based wearable fall prediction and detection system. IEEE Trans Neural Syst Rehabil Eng 2019;27(5):995-1003. https://doi.org/10.1109/TNSRE.2019.2911602
  34. Santhiranayagam B, Lai D, Begg R. Support vector machines for young and older gait classification using inertial sensor kinematics at minimum toe clearance. EAI Endorsed Trans Pervasive Health Technol 2015;16(7):e2.
  35. Aziz O, Robinovitch SN. An analysis of the accuracy of wearable sensors for classifying the causes of falls in humans. IEEE Trans Neural Syst Rehabil Eng 2011;19(6):670-6. https://doi.org/10.1109/TNSRE.2011.2162250
  36. Drover D, Howcroft J, Kofman J, Lemaire ED. Faller classification in older adults using wearable sensors based on turn and straight-walking accelerometer-based features. Sensors (Basel) 2017;17(6):1321.
  37. Berg K. Measuring balance in the elderly: development and validation of an instrument. Montreal (QC), McGill University, Doctoral Dissertation. 1992.
  38. Liston RA, Brouwer BJ. Reliability and validity of measures obtained from stroke patients using the Balance Master. Arch Phys Med Rehabil 1996;77(5):425-30. https://doi.org/10.1016/S0003-9993(96)90028-3
  39. Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc 1991;39(2):142-8. https://doi.org/10.1111/j.1532-5415.1991.tb01616.x
  40. Yardley L, Beyer N, Hauer K, Kempen G, Piot-Ziegler C, Todd C. Development and initial validation of the Falls Efficacy ScaleInternational (FES-I). Age Ageing 2005;34(6):614-9. https://doi.org/10.1093/ageing/afi196
  41. Salarian A, Russmann H, Vingerhoets FJ, Dehollain C, Blanc Y, Burkhard PR, et al. Gait assessment in Parkinson's disease: toward an ambulatory system for long-term monitoring. IEEE Trans Biomed Eng 2004;51(8):1434-43. https://doi.org/10.1109/TBME.2004.827933
  42. Bertsimas D, Dunn J. Optimal classification trees. Mach Learn 2017;106(7):1039-82. https://doi.org/10.1007/s10994-017-5633-9
  43. Doupe P, Faghmous J, Basu S. Machine learning for health services researchers. Value Health 2019;22(7):808-15. https://doi.org/10.1016/j.jval.2019.02.012
  44. Trabassi D, Serrao M, Varrecchia T, Ranavolo A, Coppola G, De Icco R, et al. Machine learning approach to support the detection of Parkinson's disease in IMU-based gait analysis. Sensors (Basel) 2022;22(10):3700.
  45. Veronese N, Maggi S. Epidemiology and social costs of hip fracture. Injury 2018;49(8):1458-60. https://doi.org/10.1016/j.injury.2018.04.015
  46. Sucerquia A, Lopez JD, Vargas-Bonilla JF. SisFall: a fall and movement dataset. Sensors (Basel) 2017;17(1):198.
  47. Aziz O, Musngi M, Park EJ, Mori G, Robinovitch SN. A comparison of accuracy of fall detection algorithms (threshold-based vs. machine learning) using waist-mounted tri-axial accelerometer signals from a comprehensive set of falls and non-fall trials. Med Biol Eng Comput 2017;55(1):45-55. https://doi.org/10.1007/s11517-016-1504-y
  48. Aziz O, Klenk J, Schwickert L, Chiari L, Becker C, Park EJ, et al. Validation of accuracy of SVM-based fall detection system using real-world fall and non-fall datasets. PLoS One 2017;12(7):e0180318.
  49. Lim K, Choi WJ. Feature extraction and evaluation for classification models of injurious falls based on surface electromyography. Phys Ther Korea 2021;28(2):123-31.  https://doi.org/10.12674/ptk.2021.28.2.123