Displaying all 7 publications

Abstract:
Sort:
  1. Amin HU, Malik AS, Kamel N, Hussain M
    Brain Topogr, 2016 Mar;29(2):207-17.
    PMID: 26613724 DOI: 10.1007/s10548-015-0462-2
    Feature extraction and classification for electroencephalogram (EEG) in medical applications is a challenging task. The EEG signals produce a huge amount of redundant data or repeating information. This redundancy causes potential hurdles in EEG analysis. Hence, we propose to use this redundant information of EEG as a feature to discriminate and classify different EEG datasets. In this study, we have proposed a JPEG2000 based approach for computing data redundancy from multi-channels EEG signals and have used the redundancy as a feature for classification of EEG signals by applying support vector machine, multi-layer perceptron and k-nearest neighbors classifiers. The approach is validated on three EEG datasets and achieved high accuracy rate (95-99 %) in the classification. Dataset-1 includes the EEG signals recorded during fluid intelligence test, dataset-2 consists of EEG signals recorded during memory recall test, and dataset-3 has epileptic seizure and non-seizure EEG. The findings demonstrate that the approach has the ability to extract robust feature and classify the EEG signals in various applications including clinical as well as normal EEG patterns.
  2. Feroz FS, Leicht G, Steinmann S, Andreou C, Mulert C
    Brain Topogr, 2017 Jan;30(1):30-45.
    PMID: 27659288 DOI: 10.1007/s10548-016-0521-3
    Growing evidence from neuroimaging studies suggest that emotional and cognitive processes are interrelated. Anatomical key structures in this context are the dorsal and rostral-ventral anterior cingulate cortex (dACC and rvACC). However, up to now, the time course of activations within these regions during emotion-cognition interactions has not been disentangled. In the present study, we used event-related potentials (ERP) and standardized low-resolution electromagnetic tomography (sLORETA) region of interest (ROI) source localization analyses to explore the time course of neural activations within the dACC and rvACC using a modified emotional Stroop paradigm. ERP components related to Stroop conflict (N200, N450 and late negativity) were analyzed. The time course of brain activations in the dACC and rvACC was strikingly different with more pronounced initial responses in the rvACC followed by increased dACC activity mainly at the late negativity window. Moreover, emotional valence modulated the earlier N450 stage within the rvACC region with higher neural activations in the positive compared to the negative and neutral conditions. Emotional arousal modulated the late negativity stage; firstly in the significant arousal × congruence ERP effect and then the significant higher current density in the low arousal condition within the dACC. Using sLORETA source localization, substantial differences in the activation time courses in the dACC and rvACC could be found during the emotional Stroop task. We suggest that during late negativity, within the dACC, emotional arousal modulated the processing of response conflict, reflected in the correlation between the ex-Gaussian µ and the current density in the dACC.
  3. Mumtaz W, Malik AS
    Brain Topogr, 2018 09;31(5):875-885.
    PMID: 29860588 DOI: 10.1007/s10548-018-0651-x
    The choice of an electroencephalogram (EEG) reference has fundamental importance and could be critical during clinical decision-making because an impure EEG reference could falsify the clinical measurements and subsequent inferences. In this research, the suitability of three EEG references was compared while classifying depressed and healthy brains using a machine-learning (ML)-based validation method. In this research, the EEG data of 30 unipolar depressed subjects and 30 age-matched healthy controls were recorded. The EEG data were analyzed in three different EEG references, the link-ear reference (LE), average reference (AR), and reference electrode standardization technique (REST). The EEG-based functional connectivity (FC) was computed. Also, the graph-based measures, such as the distances between nodes, minimum spanning tree, and maximum flow between the nodes for each channel pair, were calculated. An ML scheme provided a mechanism to compare the performances of the extracted features that involved a general framework such as the feature extraction (graph-based theoretic measures), feature selection, classification, and validation. For comparison purposes, the performance metrics such as the classification accuracies, sensitivities, specificities, and F scores were computed. When comparing the three references, the diagnostic accuracy showed better performances during the REST, while the LE and AR showed less discrimination between the two groups. Based on the results, it can be concluded that the choice of appropriate reference is critical during the clinical scenario. The REST reference is recommended for future applications of EEG-based diagnosis of mental illnesses.
  4. Feroz FS, Leicht G, Rauh J, Mulert C
    Brain Topogr, 2019 01;32(1):161-177.
    PMID: 30288663 DOI: 10.1007/s10548-018-0677-0
    This paper aims to investigate the temporal dynamics within the dorsal anterior cingulate cortex (dACC) and the rostral-ventral (rv) ACC during the interaction of emotional valence and arousal with cognitive control in patients with Schizophrenia (SZ). Although cognitive deficits in SZ are highly relevant and emotional disturbances are common, the temporal relationship of brain regions involved in the interaction of emotional and cognitive processing in SZ is yet to be determined. To address this issue, the reaction time (RT), event-related potential (ERP) and temporal dynamics of the dACC and rvACC activity were compared between SZ subjects and healthy controls (HC), using a modified emotional Stroop experiment (with factors namely congruence, arousal and valence). EEG was recorded with 64 channels and source localisation was performed using the sLORETA software package. We observed slower initial increase and lower peaks of time course activity within the dACC and rvACC in the SZ group. In this particular group, the dACC activity during late negativity was negatively correlated with a significantly higher RT in the high arousal conflict condition. In contrast to HC subjects, at the N450 window, there was no significant valence (ERP and rvACC ROI) modulation effect in the SZ subjects. Using high density EEG and source localisation, it was possible to distinguish various disturbances within the dACC and rvACC in patients with SZ, during emotion-cognition processing.
  5. Fiedler P, Pedrosa P, Griebel S, Fonseca C, Vaz F, Supriyanto E, et al.
    Brain Topogr, 2015 Sep;28(5):647-656.
    PMID: 25998854 DOI: 10.1007/s10548-015-0435-5
    Current usage of electroencephalography (EEG) is limited to laboratory environments. Self-application of a multichannel wet EEG caps is practically impossible, since the application of state-of-the-art wet EEG sensors requires trained laboratory staff. We propose a novel EEG cap system with multipin dry electrodes overcoming this problem. We describe the design of a novel 24-pin dry electrode made from polyurethane and coated with Ag/AgCl. A textile cap system holds 97 of these dry electrodes. An EEG study with 20 volunteers compares the 97-channel dry EEG cap with a conventional 128-channel wet EEG cap for resting state EEG, alpha activity, eye blink artifacts and checkerboard pattern reversal visual evoked potentials. All volunteers report a good cap fit and good wearing comfort. Average impedances are below 150 kΩ for 92 out of 97 dry electrodes, enabling recording with standard EEG amplifiers. No significant differences are observed between wet and dry power spectral densities for all EEG bands. No significant differences are observed between the wet and dry global field power time courses of visual evoked potentials. The 2D interpolated topographic maps show significant differences of 3.52 and 0.44% of the map areas for the N75 and N145 VEP components, respectively. For the P100 component, no significant differences are observed. Dry multipin electrodes integrated in a textile EEG cap overcome the principle limitations of wet electrodes, allow rapid application of EEG multichannel caps by non-trained persons, and thus enable new fields of application for multichannel EEG acquisition.
  6. Sanchez Bornot JM, Wong-Lin K, Ahmad AL, Prasad G
    Brain Topogr, 2018 11;31(6):895-916.
    PMID: 29546509 DOI: 10.1007/s10548-018-0640-0
    The brain's functional connectivity (FC) estimated at sensor level from electromagnetic (EEG/MEG) signals can provide quick and useful information towards understanding cognition and brain disorders. Volume conduction (VC) is a fundamental issue in FC analysis due to the effects of instantaneous correlations. FC methods based on the imaginary part of the coherence (iCOH) of any two signals are readily robust to VC effects, but neglecting the real part of the coherence leads to negligible FC when the processes are truly connected but with zero or π-phase (modulus 2π) interaction. We ameliorate this issue by proposing a novel method that implements an envelope of the imaginary coherence (EIC) to approximate the coherence estimate of supposedly active underlying sources. We compare EIC with state-of-the-art FC measures that included lagged coherence, iCOH, phase lag index (PLI) and weighted PLI (wPLI), using bivariate autoregressive and stochastic neural mass models. Additionally, we create realistic simulations where three and five regions were mapped on a template cortical surface and synthetic MEG signals were obtained after computing the electromagnetic leadfield. With this simulation and comparison study, we also demonstrate the feasibility of sensor FC analysis using receiver operating curve analysis whilst varying the signal's noise level. However, these results should be interpreted with caution given the known limitations of the sensor-based FC approach. Overall, we found that EIC and iCOH demonstrate superior results with most accurate FC maps. As they complement each other in different scenarios, that will be important to study normal and diseased brain activity.
  7. Kumar S, Choudhary S, Jain A, Singh K, Ahmadian A, Bajuri MY
    Brain Topogr, 2023 Apr 15.
    PMID: 37061591 DOI: 10.1007/s10548-023-00953-0
    In the field of medical imaging, the classification of brain tumors based on histopathological analysis is a laborious and traditional approach. To address this issue, the use of deep learning techniques, specifically Convolutional Neural Networks (CNNs), has become a popular trend in research and development. Our proposed solution is a novel Convolutional Neural Network that leverages transfer learning to classify brain tumors in MRI images as benign or malignant with high accuracy. We evaluated the performance of our proposed model against several existing pre-trained networks, including Res-Net, Alex-Net, U-Net, and VGG-16. Our results showed a significant improvement in prediction accuracy, precision, recall, and F1-score, respectively, compared to the existing methods. Our proposed method achieved a benign and malignant classification accuracy of 99.30 and 98.40% using improved Res-Net 50. Our proposed system enhances image fusion quality and has the potential to aid in more accurate diagnoses.
Related Terms
Filters
Contact Us

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links