Design of Elderly Fall Detection Based
on XGBoost
Min Xiao
1(&)
, Yuanjian Huang
1 , Yuanheng Wang
1
,
and Weidong Gao
2
1 School of Computer Science and Technology, Beijing University of Posts
and Telecommunications, Beijing 100876, China
xiaomincloud@163.com, {huangyj,jahracal}@bupt.edu.cn
2 School of Information and Communication Engineering, Beijing University
of Posts and Telecommunications, Beijing 100876, China
gaoweidong@bupt.edu.cn
Abstract. Aging has become a serious problem facing the whole world. Falling
is the leading cause of injury and death in the elderly. This paper proposes a fall
detection algorithm based on machine learning XGBoost and full-field positioning. Using the data of gyroscope and acceleration sensor, we exploit the
“full-field positioning” to increase the dimension of input data and propose a
method “maximum satisfaction rate” to mark and train the threshold of data. The
experimental results show that this design has obtained high accuracy on falling
detection and perfect balance between sensitivity and specificity.
Keywords: Fall detection Á XGBoost machine learning Á Gyroscope Á
Full-field positioning
1 Introduction
The WHO has pointed out that at present world’s aging trend is becoming more and
more serious. Until 2017, the elderly aged 65 and over accounted for 8.696% of the
world’s population [1]. The physiology of the human body is degraded as we grow
older. And coupled with the complexity of the living environment, the elderly are prone
to accidents such as falls. Moreover, some elderly people are alone at home, and no one
is around when they fall, which causes serious consequences. The 2018 World Health
Organization showed that falls are the second leading cause of accidental or unintentional injury deaths worldwide. Besides, adults older than 65 years of age suffer the
greatest number of fatal falls [2]. Thus, a family using automatic device is in urgent
need to protect the elderly from falling injury.
In recent years, there have been many research results of fall detection in the elderly
at home and abroad: wearable fall recognition alarm system, fall alarm system based on
embedded vision, elderly fall detection based on STM32 system, fall detection belt
based on accelerometer, wearable fall monitoring, and so on. All fall detection products
can be divided into two categories: visual inspection-based methods and sensor-based
methods.
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 46–57, 2020.
https://doi.org/10.1007/978-981-15-0187-6_6
on XGBoost
Min Xiao
1(&)
, Yuanjian Huang
1 , Yuanheng Wang
1
,
and Weidong Gao
2
1 School of Computer Science and Technology, Beijing University of Posts
and Telecommunications, Beijing 100876, China
xiaomincloud@163.com, {huangyj,jahracal}@bupt.edu.cn
2 School of Information and Communication Engineering, Beijing University
of Posts and Telecommunications, Beijing 100876, China
gaoweidong@bupt.edu.cn
Abstract. Aging has become a serious problem facing the whole world. Falling
is the leading cause of injury and death in the elderly. This paper proposes a fall
detection algorithm based on machine learning XGBoost and full-field positioning. Using the data of gyroscope and acceleration sensor, we exploit the
“full-field positioning” to increase the dimension of input data and propose a
method “maximum satisfaction rate” to mark and train the threshold of data. The
experimental results show that this design has obtained high accuracy on falling
detection and perfect balance between sensitivity and specificity.
Keywords: Fall detection Á XGBoost machine learning Á Gyroscope Á
Full-field positioning
1 Introduction
The WHO has pointed out that at present world’s aging trend is becoming more and
more serious. Until 2017, the elderly aged 65 and over accounted for 8.696% of the
world’s population [1]. The physiology of the human body is degraded as we grow
older. And coupled with the complexity of the living environment, the elderly are prone
to accidents such as falls. Moreover, some elderly people are alone at home, and no one
is around when they fall, which causes serious consequences. The 2018 World Health
Organization showed that falls are the second leading cause of accidental or unintentional injury deaths worldwide. Besides, adults older than 65 years of age suffer the
greatest number of fatal falls [2]. Thus, a family using automatic device is in urgent
need to protect the elderly from falling injury.
In recent years, there have been many research results of fall detection in the elderly
at home and abroad: wearable fall recognition alarm system, fall alarm system based on
embedded vision, elderly fall detection based on STM32 system, fall detection belt
based on accelerometer, wearable fall monitoring, and so on. All fall detection products
can be divided into two categories: visual inspection-based methods and sensor-based
methods.
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 46–57, 2020.
https://doi.org/10.1007/978-981-15-0187-6_6