ECG-Based Concentration Recognition With Multi-Task Regression

Objective: Recognition of human activities and mental states using wearable sensors and smartphones has attracted considerable attention recently. In particular, prediction of the stress level of a subject using an electrocardiogram sensor has been studied extensively. In this paper, we attempt to p...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on biomedical engineering 2019-01, Vol.66 (1), p.101-110
Main Authors: Kaji, Hirotaka, Iizuka, Hisashi, Sugiyama, Masashi
Format: Article
Language:English
Subjects:
Adaptation models
Adult
Attention - physiology
Brain modeling
brain wave
Brain Waves - physiology
Concentration
Data collection
EKG
electrocardiogram
Electrocardiography
Electrocardiography - methods
Electroencephalography - methods
Equipment Design
Female
Frequency modulation
Heart rate
Human Activities
Human influences
Humans
Machine Learning
Male
Mathematical models
Middle Aged
multi-task learning
Prediction models
Recognition
Rhythm
Signal Processing, Computer-Assisted
Smartphones
Stress
Task analysis
Wearable Electronic Devices
Wireless Technology
Young Adult
Citations: Items that this one cites
Items that cite this one
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Objective: Recognition of human activities and mental states using wearable sensors and smartphones has attracted considerable attention recently. In particular, prediction of the stress level of a subject using an electrocardiogram sensor has been studied extensively. In this paper, we attempt to predict the degree of concentration by using heart-rate features. However, due to strong diversity in individuals and high sampling costs, building an accurate prediction model is still highly challenging. Method: To overcome these difficulties, we propose to use a multitask learning (MTL) technique for effectively sharing information among similar individuals. Result: Through experiments with 18 healthy subjects performing daily office works, such as writing reports, we demonstrate that the proposed method significantly improves the accuracy of concentration prediction in small sample situations. Conclusion: The performance of the MTL method is shown to be stable across different subjects, which is an important advantage over conventional models. Significance: This improvement has significant impact in real-world concentration recognition because the data collection burden of each user can be drastically mitigated.
ISSN:0018-9294
1558-2531
DOI:10.1109/TBME.2018.2830366