Composite Common Spatial Pattern for Subject-to-Subject Transfer

Kang, Hyohyeong; Nam, Yunjun; Choi, Seungjin

doi:10.1109/lsp.2009.2022557

Cited by 220 publications

(70 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reliable classification of mental states while taking into account the change in data distribution between sessions and subjects (termed as transfer learning , Samek et al, 2013) has generated a considerable amount of interest among BCI researchers (Kang et al, 2009; Devlaminck et al, 2011; Samek et al, 2013). It allows the classifier to be trained on a fixed set of subjects and test it on a completely different set of subjects.…”

Section: Introductionmentioning

confidence: 99%

“…As a result, determination of proper features and generalization of the classifiers is a must to tackle this issue. In the past (Kang et al, 2009; Lotte and Guan, 2010), researchers have averaged the covariance matrix of different subjects toward creating a generalized covariance matrix to improve the cross-subject estimation. Another approach (Devlaminck et al, 2011) toward transfer learning employed common spatial patterns to construct a common feature space among various subjects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Generic Transferable EEG Decoder for Online Detection of Error Potential in Target Selection

et al. 2017

View full text Add to dashboard Cite

Reliable detection of error from electroencephalography (EEG) signals as feedback while performing a discrete target selection task across sessions and subjects has a huge scope in real-time rehabilitative application of Brain-computer Interfacing (BCI). Error Related Potentials (ErrP) are EEG signals which occur when the participant observes an erroneous feedback from the system. ErrP holds significance in such closed-loop system, as BCI is prone to error and we need an effective method of systematic error detection as feedback for correction. In this paper, we have proposed a novel scheme for online detection of error feedback directly from the EEG signal in a transferable environment (i.e., across sessions and across subjects). For this purpose, we have used a P300-speller dataset available on a BCI competition website. The task involves the subject to select a letter of a word which is followed by a feedback period. The feedback period displays the letter selected and, if the selection is wrong, the subject perceives it by the generation of ErrP signal. Our proposed system is designed to detect ErrP present in the EEG from new independent datasets, not involved in its training. Thus, the decoder is trained using EEG features of 16 subjects for single-trial classification and tested on 10 independent subjects. The decoder designed for this task is an ensemble of linear discriminant analysis, quadratic discriminant analysis, and logistic regression classifier. The performance of the decoder is evaluated using accuracy, F1-score, and Area Under the Curve metric and the results obtained is 73.97, 83.53, and 73.18%, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Generic Transferable EEG Decoder for Online Detection of Error Potential in Target Selection

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Atyabi et al (2013) specifically addressed the question of how to optimize the information from other subjects to the current subject played an important role in enhancing subject transfer. Kang et al (2009) attempted to apply the subject transfer to a subject with fewer training samples to generate a better set of subject-specific features. The authors emphasized that the data of other subjects having data distributions similar to each other alleviated the individual difference.…”

Section: Introductionmentioning

confidence: 99%

Improving EEG-Based Emotion Classification Using Conditional Transfer Learning

Lin

Jung

2017

Front. Hum. Neurosci.

134

View full text Add to dashboard Cite

To overcome the individual differences, an accurate electroencephalogram (EEG)-based emotion-classification system requires a considerable amount of ecological calibration data for each individual, which is labor-intensive and time-consuming. Transfer learning (TL) has drawn increasing attention in the field of EEG signal mining in recent years. The TL leverages existing data collected from other people to build a model for a new individual with little calibration data. However, brute-force transfer to an individual (i.e., blindly leveraged the labeled data from others) may lead to a negative transfer that degrades performance rather than improving it. This study thus proposed a conditional TL (cTL) framework to facilitate a positive transfer (improving subject-specific performance without increasing the labeled data) for each individual. The cTL first assesses an individual’s transferability for positive transfer and then selectively leverages the data from others with comparable feature spaces. The empirical results showed that among 26 individuals, the proposed cTL framework identified 16 and 14 transferable individuals who could benefit from the data from others for emotion valence and arousal classification, respectively. These transferable individuals could then leverage the data from 18 and 12 individuals who had similar EEG signatures to attain maximal TL improvements in valence- and arousal-classification accuracy. The cTL improved the overall classification performance of 26 individuals by ~15% for valence categorization and ~12% for arousal counterpart, as compared to their default performance based solely on the subject-specific data. This study evidently demonstrated the feasibility of the proposed cTL framework for improving an individual’s default emotion-classification performance given a data repository. The cTL framework may shed light on the development of a robust emotion-classification model using fewer labeled subject-specific data toward a real-life affective brain-computer interface (ABCI).

show abstract

“…Various feature extraction techniques have been used in the BCI literature such as common spatial pattern (CSP) [7,8], Fourier transform [9], wavelet transform [10], power spectral density analysis [11], filtering methods [12,13], polynomial coefficients [6], and autoregressive model [14,15]. Amid these techniques, CSP is one of the very widely used feature extraction techniques in motor imagery-based BCI applications.…”

Section: Introductionmentioning

confidence: 99%

“…They used the BCI competition III dataset IVa, which was recorded with 118 electrodes during the imagination of the right hand and foot. Their results showed that it was useful for cases where pretrained CSP features need to be adapted to a subject with a low number of training samples [8]. In another approach, Novi et al applied their method to the dataset IVa from BCI competition III, which recorded five subjects during the imagination of the left hand, right hand, and right foot movements.…”

Section: Introductionmentioning

confidence: 99%

Common spatial pattern-based feature extraction from the best time segment of BCI data

Aydemir

2016

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Abstract:Feature extraction is one of the most crucial stages in the field of brain computer interface (BCI). Because of its ability to directly influence the performance of BCI systems, recent studies have generally investigated how to modify existing methods or develop novel techniques. One of the most successful and well-known methods in BCI applications is the common spatial pattern (CSP). In existing CSP-based methods, the spatial filters were extracted either by using the whole data trial or by dividing the trials into a number of overlapping/nonoverlapping time segments. In this paper, we developed a CSP-based moving window technique to obtain the most distinguishable CSP features and increase the classifier performance by finding the best time segment of electroencephalogram trials. The extracted features were tested by using support vector machines (SVMs). The performance of the classifier was measured in terms of classification accuracy and kappa coefficient ( κ) . The proposed method was successfully applied to the two-dimensional cursor movement imagery data sets, which were acquired from three healthy human subjects in two sessions on different days. The experiments proved that instead of using the whole data length of EEG trials, extracting CSP features from the best time segment provides higher classification accuracy and κ rates.

show abstract

Composite Common Spatial Pattern for Subject-to-Subject Transfer

Cited by 220 publications

References 10 publications

A Generic Transferable EEG Decoder for Online Detection of Error Potential in Target Selection

A Generic Transferable EEG Decoder for Online Detection of Error Potential in Target Selection

Improving EEG-Based Emotion Classification Using Conditional Transfer Learning

Common spatial pattern-based feature extraction from the best time segment of BCI data

Contact Info

Product

Resources

About