Feature Extraction and Evaluation for Classification Models of Injurious Falls Based on Surface Electromyography
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Background: Only 2% of falls in older adults result in serious injuries (i.e., hip fracture). Therefore, it is important to differentiate injurious versus non-injurious falls, which is critical to develop effective interventions for injury prevention. Objects: The purpose of this study was to a. extract the best features of surface electromy-ography (sEMG) for classification of injurious falls, and b. find a best model provided by data mining techniques using the extracted features.Methods: Twenty young adults self-initiated falls and landed sideways. Falling trials were consisted of three initial fall directions (forward, sideways, or backward) and three knee po-sitions at the time of hip impact (the impacting-side knee contacted the other knee (“knee together”) or the mat (“knee on mat”), or neither the other knee nor the mat was contacted by the impacting-side knee (“free knee”). Falls involved “backward initial fall direction” or “free knee” were defined as “injurious falls” as suggested from previous studies. Nine features were extracted from sEMG signals of four hip muscles during a fall, including integral of absolute value (IAV), Wilson amplitude (WAMP), zero crossing (ZC), number of turns (NT), mean of am-plitude (MA), root mean square (RMS), average amplitude change (AAC), difference absolute standard deviation value (DASDV). The decision tree and support vector machine (SVM) were used to classify the injurious falls. Results: For the initial fall direction, accuracy of the best model (SVM with a DASDV) was 48%. For the knee position, accuracy of the best model (SVM with an AAC) was 49%. Further-more, there was no model that has sensitivity and specificity of 80% or greater.Conclusion: Our results suggest that the classification model built upon the sEMG features of the four hip muscles are not effective to classify injurious falls. Future studies should con-sider other data mining techniques with different muscles.
Abstract
Background: Only 2% of falls in older adults result in serious injuries (i.e., hip fracture). Therefore, it is important to differentiate injurious versus non-injurious falls, which is critical to develop effective interventions for injury prevention. Objects: The purpose of this study was to a. extract the best features of surface electromy-ography (sEMG) for classification of injurious falls, and b. find a best model provided by data mining techniques using the extracted features.Methods: Twenty young adults self-initiated falls and landed sideways. Falling trials were consisted of three initial fall directions (forward, sideways, or backward) and three knee po-sitions at the time of hip impact (the impacting-side knee contacted the other knee (“knee together”) or the mat (“knee on mat”), or neither the other knee nor the mat was contacted by the impacting-side knee (“free knee”). Falls involved “backward initial fall direction” or “free knee” were defined as “injurious falls” as suggested from previous studies. Nine features were extracted from sEMG signals of four hip muscles during a fall, including integral of absolute value (IAV), Wilson amplitude (WAMP), zero crossing (ZC), number of turns (NT), mean of am-plitude (MA), root mean square (RMS), average amplitude change (AAC), difference absolute standard deviation value (DASDV). The decision tree and support vector machine (SVM) were used to classify the injurious falls. Results: For the initial fall direction, accuracy of the best model (SVM with a DASDV) was 48%. For the knee position, accuracy of the best model (SVM with an AAC) was 49%. Further-more, there was no model that has sensitivity and specificity of 80% or greater.Conclusion: Our results suggest that the classification model built upon the sEMG features of the four hip muscles are not effective to classify injurious falls. Future studies should con-sider other data mining techniques with different muscles.
참고문헌 (33)
- Grisso JA, "Injuries in an elderly inner-city population", J Am Geriatr Soc, 38(12), pp.1326~1331 (1990)
- Empana JP, "Effect of hip fracture on mortality in elderly women: the EPIDOS prospective study", J Am Geriatr Soc, 52(5), pp.685~690 (2004)
- Wolinsky FD, "The effect of hip fracture on mortality, hospitalization, and functional status: a prospective study", Am J Public Health, 87(3), pp.398~403 (1997)
- Sattin RW, "Falls among older persons: a public health perspective", Annu Rev Public Health, 13, pp.489~508 (1992)
- Choi WJ, "Effects of hip abductor muscle forces and knee boundary conditions on femoral neck stresses during simulated falls", Osteoporos Int, 26(1), pp.291~301 (2015)
- Lim KT, "Effect of fall characteristics on the severity of hip impact during a fall on the ground from standing height", Osteoporos Int, 31(9), pp.1713~1719 (2020)
- Yang Y, "The effect of fall biomechanics on risk for hip fracture in older adults: a cohort study of video-captured falls in longterm care", J Bone Miner Res, 35(10), pp.1914~1922 (2020)
- Aziz O, "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, 55(1), pp.45~55 (2017)
- Juan C, "A framework for daily activity monitoring and fall detection based on surface electromyography and accelerometer signals", IEEE J Biomed Health Inform, 17(1), pp.38~45 (2013)
- Ren L, "Research of fall detection and fall prevention technologies: a systematic review", IEEE Access, 7, pp.77702~77722 (2019)
- Cheng WC, "Triaxial accelerometer-based fall detection method using a self-constructing cascade-AdaBoost-SVM classifier", IEEE J Biomed Health Inform, 17(2), pp.411~419 (2013)
- Santos GL, "Accelerometer-based human fall detection using convolutional neural networks", Sensors (Basel), 19(7), pp.1644~ (2019)
- Roy SH, "A combined sEMG and accelerometer system for monitoring functional activity in stroke", IEEE Trans Neural Syst Rehabil Eng, 17(6), pp.585~594 (2009)
- Adewuyi AA, "An analysis of intrinsic and extrinsic hand muscle EMG for improved pattern recognition control", IEEE Trans Neural Syst Rehabil Eng, 24(4), pp.485~494 (2016)
- Phinyomark A, "Feature reduction and selection for EMG signal classification", Expert Systems with Applications, 39(8), pp.7420~7431 (2012)
- Sutherland DH, "The evolution of clinical gait analysis part l:kinesiological EMG", Gait Posture, 14(1), pp.61~70 (2001)
- Xi X, "Evaluation of feature extraction and recognition for activity monitoring and fall detection based on wearable sEMG sensors", Sensors (Basel), 17(6), pp.1229~ (2017)
- Criswell E, "Cram’s introduction to surface electromyography", pp.21~ (2011), Jones and Bartlett
- Nazmi N, "A review of classification techniques of EMG signals during isotonic and isometric contractions", Sensors (Basel), 16(8), pp.1304~ (2016)
- Englehart K, "Classification of the myoelectric signal using time-frequency based representations", Med Eng Phys, 21(6-7), pp.431~438 (1999)
- Kang Soo Kim, "Comparison of k-nearest neighbor, quadratic discriminant and linear discriminant analysis in classification of electromyogram signals based on the wrist-motion directions", Current Applied Physics, 11(3), pp.740~745 (2011), 한국물리학회
- Zardoshti-Kermani M, "EMG feature evaluation for movement control of upper extremity prostheses", IEEE Trans Rehabil Eng, 3(4), pp.324~333 (1995)
- Hudgins B, "A new strategy for multifunction myoelectric control", IEEE Trans Biomed Eng, 40(1), pp.82~94 (1993)
- Boostani R, "Evaluation of the forearm EMG signal features for the control of a prosthetic hand", Physiol Meas, 24(2), pp.309~319 (2003)
- Park SH, "EMG pattern recognition based on artificial intelligence techniques", IEEE Trans Rehabil Eng, 6(4), pp.400~405 (1998)
- Chang GC, "Real-time implementation of electromyogram pattern recognition as a control command of man-machine interface", Med Eng Phys, 18(7), pp.529~537 (1996)
- Keleş S, "Classification of EMG signals using decision tree methods", Paper presented at: 3rd International Symposium on Sustainable Development (ISSD), pp.354~366 (2012), International Burch University
- Kwang Jun Lee, "The Effect on the Hip Muscle Activation of the Fall Direction and Knee Position During a Fall", 한국전문물리치료학회지, 28(1), pp.84~91 (2021), 한국전문물리치료학회[DOI]
- Wei W, "Weighted feature Gaussian kernel SVM for emotion recognition", Comput Intell Neurosci, 2016, pp.7696035~ (2016)
- Song W, "Design of a flexible wearable smart sEMG recorder integrated gradient boosting decision tree based hand gesture recognition", IEEE Trans Biomed Circuits Syst, 13(6), pp.1563~1574 (2019)
- Trevisan C, "Decision tree for ward admissions of older patients at the emergency department after a fall", Geriatr Gerontol Int, 18(9), pp.1388~1392 (2018)
- Cai WY, "GBDTbased fall detection with comprehensive data from posture sensor and human skeleton extraction", J Healthc Eng, 2020, pp.8887340~ (2020)
- Waris A, "Multiday evaluation of techniques for EMG-based classification of hand motions", IEEE J Biomed Health Inform, 23(4), pp.1526~1534 (2019)