ORCID Profile
0000-0002-7410-5951
Current Organisation
University of Brighton
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Elsevier BV
Date: 04-2022
DOI: 10.1016/J.COMPBIOMED.2022.105242
Abstract: Identifying motor and mental imagery electroencephalography (EEG) signals is imperative to realizing automated, robust brain-computer interface (BCI) systems. In the present study, we proposed a pretrained convolutional neural network (CNN)-based new automated framework feasible for robust BCI systems with small and le s les of motor and mental imagery EEG training data. The framework is explored by investigating the implications of different limiting factors, such as learning rates and optimizers, processed versus unprocessed scalograms, and features derived from untuned pretrained models in small, medium, and large pretrained CNN models. The experiments were performed on three public datasets obtained from BCI Competition III. The datasets were denoised with multiscale principal component analysis, and time-frequency scalograms were obtained by employing a continuous wavelet transform. The scalograms were fed into several variants of ten pretrained models for feature extraction and identification of different EEG tasks. The experimental results showed that ShuffleNet yielded the maximum average classification accuracy of 99.52% using an RMSProp optimizer with a learning rate of 0.000 1. It was observed that low learning rates converge to more optimal performances compared to high learning rates. Moreover, noisy scalograms and features extracted from untuned networks resulted in slightly lower performance than denoised scalograms and tuned networks, respectively. The overall results suggest that pretrained models are robust when identifying EEG signals because of their ability to preserve the time-frequency structure of EEG signals and promising classification outcomes.
Publisher: Springer International Publishing
Date: 2021
Publisher: Springer International Publishing
Date: 2021
Publisher: Elsevier BV
Date: 09-2018
DOI: 10.1016/J.PSCYCHRESNS.2018.05.011
Abstract: This study examined the hypothesis that a fronto-temporal disconnection in the language network underpins formal thought disorder (FTD) in schizophrenia. Forty-nine patients with a schizophrenia spectrum disorder (27 with mild FTD, 22 with severe FTD) and 26 healthy controls (HC) were included. Overall psychopathology and FTD were assessed by the Positive and Negative Syndrome Scale and the Thought, Language, and Communication scale, respectively. White matter (WM) microstructure was analysed using Tract-Based Spatial Statistics. In patients, severity of overall FTD (TLC Sum Score) was predicted by decreased fractional anisotropy (FA) in the right superior longitudinal fasciculus (SLF), and severity of negative FTD (TLC Emptiness subscale) was predicted by increased FA in the left SLF and arcuate fasciculus (AF). Notably, these results were no longer significant after correction for multiple comparisons. Compared with HC, patients showed lower FA in all the investigated language-related WM tracts as well as across the whole WM skeleton. No difference in FA was found between patients with severe and patients with mild FTD. Our results are compatible with earlier studies reporting impairments in widely spread WM tracts including those related to language processing in patients with schizophrenia.
Publisher: Elsevier BV
Date: 06-2021
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 08-2021
Publisher: Elsevier BV
Date: 05-2022
Publisher: Elsevier BV
Date: 11-2021
Publisher: Elsevier
Date: 2022
Publisher: Springer Nature Switzerland
Date: 2022
Publisher: MDPI AG
Date: 25-04-2023
DOI: 10.3390/BIOMEDICINES11051277
Abstract: Vitamin D and its role in the coronavirus-19 disease (COVID-19) pandemic has been controversially discussed, with inconclusive evidence about vitamin D3 (cholecalciferol) supplementation in COVID-19 patients. Vitamin D metabolites play an important role in the initiation of the immune response and can be an easily modifiable risk factor in 25-hydroxyvitamin D3 (25(OH)D3)-deficient patients. This is a multicenter, randomized, placebo-controlled double-blind trial to compare the effect of a single high dose of vitamin D3 followed by treatment as usual (TAU) of daily vitamin D3 daily until discharge versus placebo plus TAU in hospitalized patients with COVID-19 and 25(OH)D3-deficiency on length hospital stay. We included 40 patients per group and did not observe a significant difference in the median length of hospital stay (6 days in both groups, p = 0.920). We adjusted the length of stay for COVID-19 risk factors (β = 0.44 95% CI: −2.17–2.22), and center (β = 0.74 95% CI: −1.25–2.73). The subgroup analysis in patients with severe 25(OH)D3-deficiency ( nmol/L) showed a non-significant reduction in the median length of hospital stay in the intervention group (5.5 vs. 9 days, p = 0.299). The competing risk model with death did not reveal significant differences between the group in the length of stay (HR = 0.96, 95% CI 0.62–1.48, p = 0.850). Serum 25(OH)D3 level increased significantly in the intervention group (mean change in nmol/L intervention: +26.35 vs. control: –2.73, p 0.001). The intervention with 140,000 IU vitamin D3 + TAU did not significantly shorten the length of hospital stay but was effective and safe for the elevation of serum 25(OH)D3 levels.
Publisher: Springer Science and Business Media LLC
Date: 09-2022
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 06-2022
Location: United Kingdom of Great Britain and Northern Ireland
No related grants have been discovered for Dr Muhammad Tariq Sadiq.