IMU 원신호 기반의 기계학습을 통한 충격전 낙상방향 분류
Classification of Fall Direction Before Impact Using Machine Learning Based on IMU Raw Signals
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
As the elderly population gradually increases, the risk of fatal fall accidents among the elderly is increasing. One wayto cope with a fall accident is to determine the fall direction before impact using a wearable inertial measurement unit (IMU). In this context, a previous study proposed a method of classifying fall directions using a support vector machine with sensorvelocity, acceleration, and tilt angle as input parameters. However, in this method, the IMU signals are processed throughseveral processes, including a Kalman filter and the integration of acceleration, which involves a large amount of computationand error factors. Therefore, this paper proposes a machine learning-based method that classifies the fall direction before impact using IMU raw signals rather than processed data. In this study, we investigated the effects of the followingtwo factors on the classification performance: (1) the usage of processed/raw signals and (2) the selection of machine learning techniques. First, as a result of comparing the processed/raw signals, the difference in sensitivities between the two methods was within 5%, indicating an equivalent level of classification performance. Second, as a result of comparing sixmachine learning techniques, K-nearest neighbor and naive Bayes exhibited excellent performance with a sensitivity of86.0% and 84.1%, respectively.
Abstract
As the elderly population gradually increases, the risk of fatal fall accidents among the elderly is increasing. One wayto cope with a fall accident is to determine the fall direction before impact using a wearable inertial measurement unit (IMU). In this context, a previous study proposed a method of classifying fall directions using a support vector machine with sensorvelocity, acceleration, and tilt angle as input parameters. However, in this method, the IMU signals are processed throughseveral processes, including a Kalman filter and the integration of acceleration, which involves a large amount of computationand error factors. Therefore, this paper proposes a machine learning-based method that classifies the fall direction before impact using IMU raw signals rather than processed data. In this study, we investigated the effects of the followingtwo factors on the classification performance: (1) the usage of processed/raw signals and (2) the selection of machine learning techniques. First, as a result of comparing the processed/raw signals, the difference in sensitivities between the two methods was within 5%, indicating an equivalent level of classification performance. Second, as a result of comparing sixmachine learning techniques, K-nearest neighbor and naive Bayes exhibited excellent performance with a sensitivity of86.0% and 84.1%, respectively.
참고문헌 (17)
- 강기선, "노인의 낙상두려움, 일상생활수행능력과 삶의 질에 관한 연구 -낙상경험유무, 만성질환유무, 거주지역별 비교-", 한국산학기술학회논문지, 18(7), pp.193~199 (2017), 한국산학기술학회[DOI]
- D. Evans, "Elderly fall patients triaged to the trauma bay : age, injury patterns, and mortality risk", Am. J. Emerg. Med, 33(11), pp.1635~1638 (2015)
- 이강일, "장기요양기관 노인의 일상스트레스, 자아탄력성이 삶의 질에 미치는 영향", 한국융합학회논문지, 8(6), pp.325~336 (2017), 한국융합학회[DOI]
- M. N. Nyan, "A wearable system for pre-impact fall detection", J. Biomech, 41(16), pp.3475~3481 (2008)
- T. Prajapati, "A survey paper on wearable sensors based fall detection", Int. J. Comput. Appl, 115(13), pp.15~18 (2015)
- J. K. Lee, "Inertial sensing-based pre-impact detection of falls involving near-fall scenarios", IEEE Trans. Neural Syst. Rehabilitation Eng, 23(2), pp.258~266 (2015)
- 이정근, "수직속도 기반 충격전 낙상 감지에 관한 연구", 센서학회지, 23(4), pp.251~258 (2014), 한국센서학회[DOI]
- S. Troiler-McKinstry, "Fall detection using commodity smart watch and smart phone", Int. Conf. Artif. Intell. : Appl. Innov, 436, pp.70~78 (2014)
- T. R. Mauldin, "SmartFall : A smartwatch-based fall detection system using deep learning", Sensors, 18(10), pp.3363~ (2018)
- G. Shi, "Mobile human airbag system for fall protection using MEMS sensors and embedded SVM classifier", IEEE Sensors J, 9(5), pp.495~503 (2009)
- T. Tamura, "A wearable airbag to prevent fall injuries", IEEE Trans. Inf Technol Biomed, 13(6), pp.910~914 (2009)
- S. Ahn, "Optimization of a pre-impact fall detection algorithm and development of hip protection airbag system", Sens. Mater, 30(8), pp.1743~1752 (2018)
- A. T. Özdemir, "Detecting falls with wearable sensors using machine learning techniques", Sensors, 14(6), pp.10691~10708 (2014)
- M. Ş. Turan, "Classification of fall directions via wearable motion sensors" (2021), Digit. Signal Process
- F. Hossain, "A direction-sensitive fall detection system using single 3D accelerometer and learning classifier", Proc. of Int. Conf. Med. Eng. Health Inform. Technol, pp.1~6 (2016)
- 이정근, "서포트벡터머신을 이용한 충격전 낙상방향 판별", 센서학회지, 24(1), pp.47~53 (2015), 한국센서학회[DOI]
- J. K. Lee, "Estimation of attitude and external acceleration using inertial sensor measurement during various dynamic conditions", IEEE Trans. Instrum. Meas, 61(8), pp.2262~2273 (2012)