T59. EEG artifacts removal using machine learning algorithms and independent component analysis

Kang, Guiyeom; Jin, Songwan; Kim, Dae Keun; Kang, Seung Wan

doi:10.1016/j.clinph.2018.04.060

Cited by 15 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It is worth mentioning that in addition to the explained frequency filtering methods, and subspace approaches, various denoising approaches exists in the filed with a similar objective of removing artifacts from EEG and MEG measurements. This includes Machine Learning [39,40] and Deep Learning [41,42], generic algorithms [43], and algorithms that extend subspace approaches for further improvement of the method [44]. However, the mentioned methods have high computational complexity and often not used in practice; therefore, they are not in the scope of this thesis.…”

Section: Computationally Complex Pre-processing Methodsmentioning

confidence: 99%

Automated Pre-Processing and Automated Post-Processing in EEG/MEG Brain Source Analysis

Sadat-Nejad¹

2021

Preprint

View full text Add to dashboard Cite

Analyzing Electroencephalography (EEG)/Magnetoencephalography (MEG) brain source signals allows for a better understanding and diagnosis of various brain-related activities or injuries. Due to the high complexity of the mentioned measurements and their low spatial resolution, different techniques have been employed to enhance the quality of the obtained results. The objective of this work is to employ state-of-the-art approaches and develop algorithms with higher analysis reliability. As a pre-processing method, subspace denoising and artifact removal approaches are taken into consideration, to provide a method that automates and improves the estimation of the Number of Component (NoC) for artifacts such as Eye Blinking (EB). By using synthetic EEG-like simulation and real MEG data, it is shown that the proposed method is more reliable over the conventional manual method in estimating the NoC. For Independent Component Analysis (ICA)-based approaches, the proposed method in this thesis provides an estimation for the number of components with an accuracy of 98.7%. The thesis is also devoted to improving source localization techniques, which aims to estimate the location of the source within the brain, which elicit time-series measurements. In this context, after obtaining a practical insight into the performance of the popular L2-Regularization based approaches, a post-processing thresholding method is introduced. The proposed method improves the spatial resolution of the L2-Regularization inverse solutions, especially for Standard Low-Resolution Electromagnetic Tomography (sLORETA), which is a well-known and widely used inverse solution. As a part of the proposed method, a novel noise variance estimation is introduced, which combines the kurtosis statistical parameter and data (noise) entropy. This new noise variance estimation technique allows for a superior performance of the proposed method compared to the existing ones. The algorithm is validated on the synthetic EEG data using well-established validation metrics. It is shown that the proposed solution improves the resolution of conventional methods in the process of thresholding/denoising automatically and without loss of any critical information.

show abstract

Section: Computationally Complex Pre-processing Methodsmentioning

confidence: 99%

Automated Pre-Processing and Automated Post-Processing in EEG/MEG Brain Source Analysis

Sadat-Nejad¹

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Metode ICA dan BSS ini akan lebih akurat dengan semakin banyaknya jumlah elektroda. Kelemahan dari metode ini adalah terdapat pengaruh dari subjektivitas dan kompetensi dari peneliti dalam melakukan inspeksi visual dan menentukan apakah komponen tertentu merupakan artefak atau gelombang otak (Kang et al, 2018). Selain menggunakan ICA, metode lain yang digunakan misalnya metode regresi, wavelet Tranform, Empirical-mode Decomposition (EMD), maupun metode hybrid atau campuran (Jiang et al, 2019).…”

Section: Baganunclassified

“…Metode yang dijelaskan pada Bagan 5 merupakan campuran ICA dengan BSS. Selain itu, penggunaan machine learning juga digunakan, misalnya dengan algoritma artificial neural network (ANN; Kang et al, 2018).…”

Section: Baganunclassified

Dasar - Dasar Electroencephalography (EEG) bagi Riset Psikologi

Khakim

Kusrohmaniah

2021

buletinpsikologi

View full text Add to dashboard Cite

Electroencephalography (EEG) merupakan metode untuk merekam aktivitas elektris otak pada permukaan kulit kepala. EEG merekam fluktuasi potensial elektris yang muncul sebagai akibat dari aktivitas sel-sel otak. Seiring dengan kemajuan penelitian dan semakin canggih alat ukur, EEG semakin banyak digunakan dalam penelitian mengenai fungsi kognitif. Artikel ini bertujuan untuk menjelaskan pengantar teoretis mengenai alat EEG serta proses dalam analisis data untuk konteks penelitian dan eksperimen dalam kajian ilmu psikologi maupun ilmu sosial secara umum. Bagian awal mendeskripsikan mengenai dasar neural dan asumsi pengukuran dalam EEG, yang diikuti dengan penjelasan mengenai komponen-komponen alat EEG dan standar pemasangan. Bagian kedua menjelaskan mengenai pemrosesan sinyal yang memberikan contoh berbagai artefak yang merusak kualitas data EEG, serta beberapa metode dalam melakukan koreksi artefak yang umum digunakan. Ekstraksi fitur menjelaskan beberapa contoh metode dalam mengolah data EEG untuk kemudian fitur tersebut diasosiasikan dengan perilaku, proses mental atau aktivitas otak.

show abstract

“…But, the recently developed iSyncBrain platform provides an efficient automated process targeted to various clinical circumstances, even for telemedicine. The platform on the cloud comprised of AI-guided EEG denoising process (Kang et al, 2018) trained through 1800 normative EEG data collected during last 7 years, sex classified healthy normative qEEG database, standardized qEEG feature extraction process from adaptive mixture ICA (AMICA) dipole source information to sLORETA-based ROI connectivity, normative library and group statistics for researchers (Kim et al, 2018;D. Lee et al, 2018;Min et al, 2020), and a series of qEEG discriminant biomarker has been implemented, such as early screening of Alzheimer dementia, prognosis of coma patients, and brain age for development disorder (Han et al, 2021;Shim & Shin, 2020;Thapa et al, 2020).…”

Section: A Possibility Of Qeeg-centered Mental Healthcare Platform As a Mainstream Practice In Mental Healthmentioning

confidence: 99%

Proceedings of the 2021 ISNR Annual Conference (Virtual): Keynote and Plenary Presentations

Research¹

2021

View full text Add to dashboard Cite

show abstract

T59. EEG artifacts removal using machine learning algorithms and independent component analysis

Cited by 15 publications

References 0 publications

Automated Pre-Processing and Automated Post-Processing in EEG/MEG Brain Source Analysis

Automated Pre-Processing and Automated Post-Processing in EEG/MEG Brain Source Analysis

Dasar - Dasar Electroencephalography (EEG) bagi Riset Psikologi

Proceedings of the 2021 ISNR Annual Conference (Virtual): Keynote and Plenary Presentations

Contact Info

Product

Resources

About