• Journal of Computing Science and Engineering
• /
• v.7 no.2
• /
• pp.139-146
• /
• 2013

Recent advances in computer science enabled people with severe motor disabilities to use brain-computer interfaces (BCI) for communication, control, and even to restore their motor disabilities. This paper reviews the most recent works of BCI in stroke rehabilitation with a focus on methodology that reported on data collected from stroke patients and clinical studies that reported on the motor improvements of stroke patients. Both types of studies are important as the former advances the technology of BCI for stroke, and the latter demonstrates the clinical efficacy of BCI in stroke. Finally some challenges are discussed.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.197-206
• /
• 2014

Motor imagery based Brain-computer Interface(BCI), which has recently attracted attention, is the technique for decoding the user's voluntary motor intention using Electroencephalography(EEG). For classifying the motor imagery, event-related desynchronization(ERD), which is the phenomenon of EEG voltage drop at sensorimotor area in ${\mu}$-band(8-13Hz), has been generally used but this method are not free from the performance degradation of the BCI system because EEG has low spatial resolution and shows different ERD-appearing band according to users. Common spatial pattern(CSP) was proposed to solve the low spatial resolution problem but it has a disadvantage of being very sensitive to frequency-band selection. Discriminative filter bank common spatial pattern(DFBCSP) tried to solve the frequency-band selection problem by using the Fisher ratio of the averaged EEG signal power and establishing discriminative filter bank(DFB) which only includes the feature frequency-band. However, we found that DFB might not include the proper filters showing the spatial pattern of ERD. To solve this problem, we apply a band-selection process using CSP feature vectors and linear discriminant analysis to DFBCSP instead of the averaged EEG signal power. The filter selection results and the classification accuracies of the existing and the proposed methods show that the CSP feature is more effective than signal power feature.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1535-1543
• /
• 2015

The brain-computer interface obtains a user's electroencephalogram as a replacement communication unit for the disabled such that the user is able to control machines by simply thinking instead of using hands or feet. In this paper, we propose a feature extraction method based on a non-selected filter by SBCSP to classify motor imagery EEG. First, we divide frequencies (4~40 Hz) into 4-Hz units and apply CSP to each Unit. Second, we obtain the FLD score vector by combining FLD results. Finally, the FLD score vector is projected onto the optimal plane for classification using PCA. We use BCI Competition III dataset IVa, and Extracted features are used as input for LS-SVM. The classification accuracy of the proposed method was evaluated using $10{\times}10$ fold cross-validation. For subjects 'aa', 'al', 'av', 'aw', and 'ay', results were $85.29{\pm}0.93%$, $95.43{\pm}0.57%$, $72.57{\pm}2.37%$, $91.82{\pm}1.38%$, and $93.50{\pm}0.69%$, respectively.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.2029-2030
• /
• 2011

이 논문에서는 최근 활발히 연구되고 있는 BCI 실험에서 센서 위치의 변화에 따른 정확도 감소를 줄이는 방법을 알아본다. 이를 위해 특징추출 방법에서 많이 사용되는 두 가지 방법 (Power Spectrum Density, Phase Lock Value) 을 비교 및 평가 한다. motor imagery BCI 실험 결과 phase정보를 이용하는 Phase Lock Value가 달라지는 센서 위치에 덜 민감하다는 것을 확인할 수 있었다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.373-376
• /
• 2004

최근에 뇌파를 이용하여 컴퓨터와 통신하거나 기기를 제어할 수 있는 이른바 뇌-컴퓨터 인터페이스BCI(Brain-Computer Interface)에 대한 연구가 대두되고 있다. 이러한 BCI 연구의 궁극적 목표는 다양한 정신상태에 따른 뇌파의 특성을 파악하여 컴퓨터나 기기 등을 제어하는 것이다. 본 논문에서는 움직임과 관련 있는 10～12Hz의 mu파 영역에서의 ERD/ERS를 계산하였고, 그 결과 왼쪽과 오른쪽 손의 움직임을 상상할 때에 운동과 관련된 기능이 집중되어 있는 일차운동영역(primary motor area)의 mu파에서 ERD/ERS의 차이가 나타남을 발견하였다 또한, RLS방법을 사용한 Adaptive Autoregressive Model 계수의 특징을 추출을 하였으며, 이를 신경망으로 학습시켜 인식률을 비교하였다.

• Journal of KIISE
• /
• v.44 no.9
• /
• pp.887-892
• /
• 2017

Brain-computer interface (BCI) is a technology that controls computer and transmits intention by measuring and analyzing electroencephalogram (EEG) signals generated in multi-channel during mental work. At this time, optimal EEG channel selection is necessary not only for convenience and speed of BCI but also for improvement in accuracy. The optimal channel is obtained by removing duplicate(redundant) channels or noisy channels. This paper propose a dual filter-based channel selection method to select the optimal EEG channel. The proposed method first removes duplicate channels using Spearman's rank correlation to eliminate redundancy between channels. Then, using F score, the relevance between channels and class labels is obtained, and only the top m channels are then selected. The proposed method can provide good classification accuracy by using features obtained from channels that are associated with class labels and have no duplicates. The proposed channel selection method greatly reduces the number of channels required while improving the average classification accuracy.

• Journal of Biomedical Engineering Research
• /
• v.30 no.5
• /
• pp.428-437
• /
• 2009

The Performance research about EEG-BCI algorithm in BCI-naive subjects is very important for evaluating the applicability to the public. We analyzed the result of the performance evaluation experiment about the EEG-BCI algorithm in BCI-naive subjects on three different aspects. The EEG-BCI algorithm used in this paper is composed of the common spatial pattern(CSP) and the least square linear classifier. CSP is used for obtaining the characteristic of event related desynchronization, and the least square linear classifier classifies the motor imagery EEG data of the left hand or right hand. The performance evaluation experiments about EEG-BCI algorithm is conducted for 40 men and women whose age are 23.87${\pm}$2.47. The performance evaluation about EEG-BCI algorithm in BCI-naive subjects is analyzed in terms of the accuracy, the relation between the information transfer rate and the accuracy, and the performance changes when the different types of cue were used in the training session and testing session. On the result of experiment, BCI-naive group has about 20% subjects whose accuracy exceed 0.7. And this results of the accuracy were not effected significantly by the types of cue. The Information transfer rate is in the inverse proportion to the accuracy. And the accuracy shows the severe deterioration when the motor imagery is less then 2 seconds.

• The Journal of the Korea Contents Association
• /
• v.10 no.6
• /
• pp.27-34
• /
• 2010

This paper presents a methodology for classifying left and right motor imagery using a neural network with weighted fuzzy membership functions (NEWFM) and wavelet-based feature extraction. Wavelet coefficients are extracted from electroencephalogram(EEG) signal by wavelet transforms in the first step. In the second step, sixty numbers of initial features are extracted from wavelet coefficients by the frequency distribution and the amount of variability in frequency distribution. The distributed non-overlap area measurement method selects the minimized number of features by removing the worst input features one by one, and then minimized six numbers of features are selected with the highest performance result. The proposed methodology shows that accuracy rate is 86.43% with six numbers of features.

• Korean Journal of Applied Biomechanics
• /
• v.19 no.4
• /
• pp.681-688
• /
• 2009

To better delineate the changes in corticospinal excitability when older adults are asked to observe and/or imagine actions, 22 right-handed older adults without neurological abnormalities were included in this study. The amplitude and latency of motor evoked potentials (MEPs) by transcranial magnetic stimulation were recorded in the abductor pollicis brevis of the dominant hand during passive observation/imagery/active observation of slow/fast action of abduction of right thumb and also at resting state. Thus, active observation showed better changes than passive, but slow and fast action revealed no difference at all.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.1661-1664
• /
• 2013

본 연구는 만성 뇌졸중 환자 5 명을 대상으로 상지 운동(Affected hand의 주먹 쥐기/펴기운동)시 참가자의 운동의지와 운동 수행의 유무에 따라 차이가 있을 것을 가정하고, 운동 수행 및 운동의지가 존재하는 Active movement와 운동 수행을 하지만 운동의지가 없는 Passive movement, 운동 수행은 없지만 운동의지가 있는 Motor Imagery(MI)의 세가지 task에 따른 뇌파의 연결성을 비교하고자 한다. 이 때 EEG 영역 간의 연결성을 보기 위한 분석 방식 중 하나인 Phase locking value(PLV)를 통해 각 task 간의 차이를 비교 및 분석했다. 운동 수행은 동일하지만 운동의지 유무에 따른 차이는 Passive movement가 전반적으로 뇌 영역간 연결이 감소하고 Active movement가 motor task 시작 후 375ms를 기점으로 급격히 증가함을 보이는 데에서 발견할 수 있었으며, 운동 수행 유무에 따른 차이는 687.5ms 이후 Active movement에 비해 MI에서 뇌 영역 간 연결 수가 확연히 감소하는 데에서 큰 차이를 나타내었다. 이에 따라 본 연구에서는 만성 뇌졸중 환자의 상지운동 시의 motor task에 따른 EEG 영역간의 연결성을 토대로 운동의지 검출이 가능성이 있음을 밝혔다.