Investigating the Performance of Wearable Motion Sensors on recognizing falls and daily activities via machine learning

Kavuncuoğlu, Erhan; Uzunhisarcikli, Esma; Barshan, Billur; Özdemir, Ahmet Turan

doi:10.1016/j.dsp.2021.103365

Cited by 17 publications

(3 citation statements)

References 85 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 2 MTw sensor placed on the waist and the representation of collected data from the Awinda station in computer software [12]. Machine learning application using hybrid dataset.…”

Section: Declarations Figuresmentioning

confidence: 99%

See 1 more Smart Citation

Achieving High Success in Fall Detection through Cross-Brand Inertial Sensor Utilization of Hybrid Data in Machine Learning

BUZPINAR

2024

Preprint

View full text Add to dashboard Cite

Falls can result in severe injuries and even mortality among individuals of all age groups. Hence, numerous wearable sensor-based fall monitoring systems are being developed to provide assistance. Fall detection and activity tracking have been partially successful using smartwatches, smartphones, and specialized devices. However, a comprehensive solution that combines sensor data from different brands in a single model and performs fall detection with high accuracy and at a satisfactory level has not been encountered. This study aims to bridge this research gap by combining data from two different brands of IMUs (inertial measurement units) that incorporate accelerometers, magnetometers, and gyroscopes, in order to create a hybrid dataset. To achieve accurate predictions on data from both brands, machine learning (ML) models were trained using ML algorithms. The first dataset was obtained from 14 volunteers using a commercially available activity tracking system called Motion Trackers Wireless (MTw). The second dataset was collected from 30 volunteers using a custom-designed Activity Tracking Device (ATD) specifically developed for detecting falls and daily-life activities. In both cases, the sensors from the respective brands were positioned on the waist to capture data related to falls and daily-life activities. The data was organized using a time-series style to reveal relational effect of the sequential falling data. During the modelling, ten different classifiers trained, and classification was performed on unseen data using the data splitting method. The Extra Tree algorithm emerged as the most successful model, achieving an accuracy of 99.54%, precision of 99.18%, recall of 99.79%, and an F-score of 99.49% on the hybrid dataset constructed from the MTw and ATD datasets. This study demonstrates hybrid dataset to create a successful system with high accuracy and low false alarm rates using inertial sensor data from various brands.

show abstract

“…Figure 2 MTw sensor placed on the waist and the representation of collected data from the Awinda station in computer software [12]. Machine learning application using hybrid dataset.…”

Section: Declarations Figuresmentioning

confidence: 99%

“…Some studies only consider signi cant acceleration effects [11]. Efforts to improve system performance include increasing the number and variety of sensors and determining the most effective sensor combinations through experimentation [12].…”

Section: Introductionmentioning

confidence: 99%

Achieving High Success in Fall Detection through Cross-Brand Inertial Sensor Utilization of Hybrid Data in Machine Learning

BUZPINAR

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…PIPA has been around in China for well over 2,000 years. Its complex and vibrant timbre has contributed to the particular attractiveness of pipa music art, which can be heard around the world [2,3]. As a solo instrument, the pipa is capable of creating "a piece of its own" throughout the playing process.…”

Section: Introductionmentioning

confidence: 99%

Design of Tone Recognition System for Pipa Strings Based on Wireless Sensors

Pang¹

2022

Mobile Information Systems

View full text Add to dashboard Cite

Pipa has been widely used as a wooden musical instrument as early as the Han period. It mainly relies on human fingers to pluck the strings to make sounds. The timbre of the pipa also has a very strong national characteristic. Understanding the meaning of words, phrases, and sentences requires tone recognition. For tone identification, the fundamental frequency conveys the most unique information. This paper summarizes the influence of the pipa’s characteristics, technology and skills of the player, and the player’s factors on the timbre of the pipa performance. In addition, given the problems of low accuracy and high data packet loss rate in the existing pipa playing tone recognition methods, this paper proposes a pipa string playing tone recognition algorithm based on wireless sensors. According to the sensor stage and weight-corresponding characteristics, on the premise of minimizing the total mean square error, the optimal weight corresponding to the sensor measurement value is found according to the adaptive mode. The unbiased estimated value of the measured value of each node is obtained through iterative calculation. The Euclidean distance between the measured value and the estimated value of each sensor obtained by the normalization process is used as the adaptive weighted recognition weight. It completes the conversion of the adaptive weighted recognition algorithm structure and the binary recognition result. The experimental results show that the proposed method has good tonal recognition accuracy in the pipa playing environment.

show abstract