A Comparison of Artifact Reduction Methods for Real-Time Analysis of fNIRS Data

Authors:
Takayuki Nozawa;Toshiyuki Kondo
Affiliations:
Institute of Symbiotic Science and Technology, Tokyo University of Agriculture, and Technology, Tokyo, Japan 184-8588;Institute of Symbiotic Science and Technology, Tokyo University of Agriculture, and Technology, Tokyo, Japan 184-8588
Venue:
Proceedings of the Symposium on Human Interface 2009 on Human Interface and the Management of Information. Information and Interaction. Part II: Held as part of HCI International 2009
Year:
2009

Citing 4
Cited 1

A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Independent component analysis for noisy data: MEG data analysis

Neural Networks
Digital Signal Processing (4th Edition)

Digital Signal Processing (4th Edition)
A comparison of methods for multiclass support vector machines

IEEE Transactions on Neural Networks

Effectiveness of Intrinsically Motivated Adaptive Agent for Sustainable Human-Agent Interaction

ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Due to its convenient, low physical restraint, and electric noise tolerant features, functional near-infrared spectroscopy (fNIRS) is expected to be a useful tool in monitoring users' brain activity in HCI. However, fNIRS measurement suffers from various kinds of artifacts, and no standardized method for artifact reduction has been established so far. In this study, we compared high-pass/band-pass filtering, global and local average references, independent component analysis (ICA) based method, and their combinations. Their effectiveness for artifact reduction was evaluated by a cognitive task recognition experiment. The results showed all the methods have artifact reduction capability, but their effectiveness depends on subjects and tasks. This suggests that it can be more practical to try various artifact reduction methods and chose the best one for each task and subject, instead of pursuing a single standardized method.