Jedním z mnoha problémů elektroencefalografie (EEG) je analýza aktivity mozkové činnosti z měřených dat, která mohou být zkreslena různými poruchami-artefakty. V tomto článku jsme se zaměřili na ověření vlastností dvou metod - analýzy nezávislých komponent (Independent Component Analysis, ICA) a analýzy hlavních komponent (Principal Component Analysis, PCA), které mohou tyto artefakty eliminovat. Metody jsme aplikovali na reálná EEG data, která byla kontaminována amplitudovým a síťovým artefaktem. Cílem bylo zjistit, do jaké míry jsou tyto metody schopny znehodnocené signály rekonstruovat do diagnosticky vyuŽitelné podoby. Výsledky jsme u obou metod vzájemně porovnali, včetně ověření shody s názorem lékaře.
Extraction of the meaningful brain activity informationfrom measured signals distorted by various artifacts is a practical problem in electroencephalography (EEG). Eye movements, muscle activity and mechanical and electrical displacements in the measuring apparatus represent typical artefacts. Several methods were developed for removing these artifacts. Two of them - Independent Component Analysis (ICA) and Principal Component Analysis (PCA) are discussed in this paper. Both ICA and PCA are useful in signal description, optimal feature extraction, and data compression. We would like to show that ICA and PCA could as well effectively separate and remove contamination from a wide variety of artifactual sources in EEG records. Results obtained using ICA with those from PCA are compared.
- MeSH
- Artifacts MeSH
- Electroencephalography MeSH
- Factor Analysis, Statistical methods MeSH
- Research Support as Topic MeSH
- Publication type
- Review MeSH
Adaptive and learning systems for sygnal processing, communications, and control
481 s.
We have developed a method focusing on ECG signal de-noising using Independent component analysis (ICA). This approach combines JADE source separation and binary decision tree for identification and subsequent ECG noise removal. In order to to test the efficiency of this method comparison to standard filtering a wavelet- based de-noising method was used. Freely data available at Physionet medical data storage were evaluated. Evaluation criteria was root mean square error (RMSE) between original ECG and filtered data contaminated with artificial noise. Proposed algorithm achieved comparable result in terms of standard noises (power line interference, base line wander, EMG), but noticeably significantly better results were achieved when uncommon noise (electrode cable movement artefact) were compared.
Na kazuistické studii nemocného s obsedantně-kompulzivní poruchou (OCD) je demonstrováno diagnostické a terapeutické využití informace elektrické aktivity mozku (EEG) rozložené metodou analýzy nezávislých komponent (ICA). Klidový EEG záznam pacienta byl analyzován pomocí sLORETA (standardized low-resolution electromagnetic tomography) a pomocí analýzy nezávislých komponent v softwaru Independent Component Neurofeedback (ICoN, Nova Tech EEG, Inc). Srovnání s sLORETA normativní databází (Nova Tech EEG, Inc) ukázalo, že pacient má zvýšené množství absolutní theta aktivity lokalizované zejména v předním cingulu a orbitofrontální kůře, jejichž úloha v patofyziologii OCD byla v minulosti popsána. Abnormní theta vlny byly patrné i při vizuálním hodnocení EEG záznamu. Analýza nezávislých komponent ukázala, že jejich hlavní zdroj je lokalizován v afektivní části předního cingula a mediální orbitofrontální kůře. Je racionální předpokládat, že neurofeedback zaměřený na snižování aktivity tohoto zdroje v pásmu theta by mohl vést k normalizaci dysfunkční neuronální sítě a ke zlepšení klinických symptomů.
We demonstrate the potential diagnostic and therapeutic use of electrical brain activity information decomposed via independent component analysis (ICA ) in an obsessive-compulsive patient. The resting EEG was analyzed by sLORETA (standardized low-resolution electromagnetic tomography) and by the ICA using the Independent Component Neurofeedback software (ICoN, Nova Tech EEG, Inc). The sLORETA normative database comparison (Nova Tech EEG, Inc) revealed increase of absolute power in the theta frequency band, especially in the anterior cingulate and orbitofrontal gyrus whose involvement in OCD pathophysiology has been previously reported. Abnormal theta waves were also detectable by visual EEG inspection. The ICA identified their main source, localized in the affective part of the anterior cingulate and in the medial orbitofrontal cortex. With respect to our findings we hypothesize that neurofeedback aimed at decreasing
- MeSH
- Adult MeSH
- Electroencephalography methods utilization MeSH
- Financing, Organized MeSH
- Image Interpretation, Computer-Assisted methods utilization MeSH
- Humans MeSH
- Obsessive-Compulsive Disorder diagnosis therapy MeSH
- Theta Rhythm methods utilization MeSH
- Check Tag
- Adult MeSH
- Humans MeSH
- Male MeSH
- Publication type
- Case Reports MeSH
This paper addresses the overlearning problem in the independent component analysis (ICA) used for the removal of muscular artifacts from electroencephalographic (EEG) records. We note that for short EEG records with high number of channels the ICA fails to separate artifact-free EEG and muscular artifacts, which has been previously attributed to the phenomenon called overlearning. We address this problem by projecting an EEG record into several subspaces with a lower dimension, and perform the ICA on each subspace separately. Due to a reduced dimension of the subspaces, the overlearning is suppressed, and muscular artifacts are better separated. Once the muscular artifacts are removed, the signals in the individual subspaces are combined to provide an artifact free EEG record. We show that for short signals and high number of EEG channels our approach outperforms the currently available ICA based algorithms for muscular artifact removal. The proposed technique can efficiently suppress ICA overlearning for short signal segments of high density EEG signals.
- MeSH
- Algorithms * MeSH
- Artifacts * MeSH
- Electroencephalography * methods MeSH
- Electromyography methods MeSH
- Humans MeSH
- Adolescent MeSH
- Young Adult MeSH
- Signal Processing, Computer-Assisted * MeSH
- Overlearning MeSH
- Muscles physiology MeSH
- Check Tag
- Humans MeSH
- Adolescent MeSH
- Young Adult MeSH
- Male MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
- Evaluation Study MeSH
- Research Support, Non-U.S. Gov't MeSH
The independent component analysis (ICA) based methods are among the most prevalent techniques used for non-invasive fetal electrocardiogram (NI-fECG) processing. Often, these methods are combined with other methods, such adaptive algorithms. However, there are many variants of the ICA methods and it is not clear which one is the most suitable for this task. The goal of this study is to test and objectively evaluate 11 variants of ICA methods combined with an adaptive fast transversal filter (FTF) for the purpose of extracting the NI-fECG. The methods were tested on two datasets, Labour dataset and Pregnancy dataset, which contained real records obtained during clinical practice. The efficiency of the methods was evaluated from the perspective of determining the accuracy of detection of QRS complexes through the parameters of accuracy (ACC), sensitivity (SE), positive predictive value (PPV), and harmonic mean between SE and PPV (F1). The best results were achieved with a combination of FastICA and FTF, which yielded mean values of ACC = 83.72%, SE = 92.13%, PPV = 90.16%, and F1 = 91.14%. Time of calculation was also taken into consideration in the methods. Although FastICA was ranked to be the sixth fastest with its mean computation time of 0.452 s, it had the best ratio of performance and speed. The combination of FastICA and adaptive FTF filter turned out to be very promising. In addition, such device would require signals acquired from the abdominal area only; no need to acquire reference signal from the mother's chest.
Different strategies have been developed using Independent Component Analysis (ICA) to automatically de-noise fMRI data, either focusing on removing only certain components (e.g. motion-ICA-AROMA, Pruim et al., 2015a) or using more complex classifiers to remove multiple types of noise components (e.g. FIX, Salimi-Khorshidi et al., 2014 Griffanti et al., 2014). However, denoising data obtained in an acute setting might prove challenging: the presence of multiple noise sources may not allow focused strategies to clean the data enough and the heterogeneity in the data may be so great to critically undermine complex approaches. The purpose of this study was to explore what automated ICA based approach would better cope with these limitations when cleaning fMRI data obtained from acute stroke patients. The performance of a focused classifier (ICA-AROMA) and a complex classifier (FIX) approaches were compared using data obtained from twenty consecutive acute lacunar stroke patients using metrics determining RSN identification, RSN reproducibility, changes in the BOLD variance, differences in the estimation of functional connectivity and loss of temporal degrees of freedom. The use of generic-trained FIX resulted in misclassification of components and significant loss of signal (< 80%), and was not explored further. Both ICA-AROMA and patient-trained FIX based denoising approaches resulted in significantly improved RSN reproducibility (p < 0.001), localized reduction in BOLD variance consistent with noise removal, and significant changes in functional connectivity (p < 0.001). Patient-trained FIX resulted in higher RSN identifiability (p < 0.001) and wider changes both in the BOLD variance and in functional connectivity compared to ICA-AROMA. The success of ICA-AROMA suggests that by focusing on selected components the full automation can deliver meaningful data for analysis even in population with multiple sources of noise. However, the time invested to train FIX with appropriate patient data proved valuable, particularly in improving the signal-to-noise ratio.
- MeSH
- Principal Component Analysis * MeSH
- Stroke diagnostic imaging MeSH
- Oxygen blood MeSH
- Humans MeSH
- Magnetic Resonance Imaging * MeSH
- Neural Pathways diagnostic imaging MeSH
- Image Processing, Computer-Assisted * MeSH
- Reproducibility of Results MeSH
- Aged, 80 and over MeSH
- Aged MeSH
- Check Tag
- Humans MeSH
- Male MeSH
- Aged, 80 and over MeSH
- Aged MeSH
- Female MeSH
- Publication type
- Journal Article MeSH
Electroencephalography (EEG) oscillations reflect the superposition of different cortical sources with potentially different frequencies. Various blind source separation (BSS) approaches have been developed and implemented in order to decompose these oscillations, and a subset of approaches have been developed for decomposition of multi-subject data. Group independent component analysis (Group ICA) is one such approach, revealing spatiospectral maps at the group level with distinct frequency and spatial characteristics. The reproducibility of these distinct maps across subjects and paradigms is relatively unexplored domain, and the topic of the present study. To address this, we conducted separate group ICA decompositions of EEG spatiospectral patterns on data collected during three different paradigms or tasks (resting-state, semantic decision task and visual oddball task). K-means clustering analysis of back-reconstructed individual subject maps demonstrates that fourteen different independent spatiospectral maps are present across the different paradigms/tasks, i.e. they are generally stable.
- MeSH
- Algorithms MeSH
- Principal Component Analysis MeSH
- Electroencephalography methods statistics & numerical data MeSH
- Image Interpretation, Computer-Assisted methods MeSH
- Humans MeSH
- Magnetic Resonance Imaging MeSH
- Brain Mapping methods MeSH
- Young Adult MeSH
- Signal Processing, Computer-Assisted MeSH
- Psychomotor Performance physiology MeSH
- Reproducibility of Results MeSH
- Decision Making physiology MeSH
- Cluster Analysis MeSH
- Visual Perception physiology MeSH
- Check Tag
- Humans MeSH
- Young Adult MeSH
- Male MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
The goal of this paper is to describe a robust artifact removal (RAR) method, an automatic sequential procedure which is capable of removing short-duration, high-amplitude artifacts from long-term neonatal EEG recordings. Such artifacts are mainly caused by movement activity, and have an adverse effect on the automatic processing of long-term sleep recordings. The artifacts are removed sequentially in short-term signals using independent component analysis (ICA) transformation and wavelet denoising. In order to gain robustness of the RAR method, the whole EEG recording is processed multiple times. The resulting tentative reconstructions are then combined. We show results in a data set of signals from ten healthy newborns. Those results prove, both qualitatively and quantitatively, that the RAR method is capable of automatically rejecting the mentioned artifacts without changes in overall signal properties such as the spectrum. The method is shown to perform better than either the wavelet-enhanced ICA or the simple artifact rejection method without the combination procedure.
- MeSH
- Algorithms MeSH
- Principal Component Analysis * MeSH
- Artifacts * MeSH
- Time Factors MeSH
- Electroencephalography methods MeSH
- Humans MeSH
- Infant, Newborn MeSH
- Image Processing, Computer-Assisted methods MeSH
- Signal Processing, Computer-Assisted MeSH
- Reference Standards MeSH
- Sleep physiology MeSH
- Wavelet Analysis * MeSH
- Check Tag
- Humans MeSH
- Infant, Newborn MeSH
- Publication type
- Journal Article MeSH
- Research Support, Non-U.S. Gov't MeSH
Newborn screening (NBS) of inborn errors of metabolism (IEMs) is based on the reference ranges established on a healthy newborn population using quantile statistics of molar concentrations of biomarkers and their ratios. The aim of this paper is to investigate whether multivariate independent component analysis (ICA) is a useful tool for the analysis of NBS data, and also to address the structure of the calculated ICA scores. NBS data were obtained from a routine NBS program performed between 2013 and 2022. ICA was tested on 10,213/150 free-diseased controls and 77/20 patients (9/3 different IEMs) in the discovery/validation phases, respectively. The same model computed during the discovery phase was used in the validation phase to confirm its validity. The plots of ICA scores were constructed, and the results were evaluated based on 5sd levels. Patient samples from 7/3 different diseases were clearly identified as 5sd-outlying from control groups in both phases of the study. Two IEMs containing only one patient each were separated at the 3sd level in the discovery phase. Moreover, in one latent variable, the effect of neonatal birth weight was evident. The results strongly suggest that ICA, together with an interpretation derived from values of the "average member of the score structure", is generally applicable and has the potential to be included in the decision process in the NBS program.
- Publication type
- Journal Article MeSH
