Are you Hengsong Sheng?

Claim your profile

Publications (4)4.59 Total impact

  • Hengsong Sheng, Jin Xu, Hongjun Tian
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study a novel method is proposed to automatically remove eye-blink artifacts from EEG signals. EEG signals were first decomposed into independent components (ICs) by independent component analysis (ICA). Using temporal correlation theory, the sum of the correlation values between each IC and the EEG signals at some special electrodes was calculated, respectively. Consequently, all the correlation values were sorted by descending order. The ICs which have significantly bigger correlation than the other ICs were picked to set as eye-blink artifact components, which were then set as zero to reconstruct clean EEG signals. Finally, the proposed method was evaluated on some contaminated EEG signals. The experimental results show that it is able to effectively remove eye-blink artifacts with little distortion of underlying brain signals.
    Green Computing and Communications (GreenCom), 2013 IEEE and Internet of Things (iThings/CPSCom), IEEE International Conference on and IEEE Cyber, Physical and Social Computing; 01/2013
  • Jin Xu, Hengsong Sheng, Wutao Lou, Songzhen Zhao
    [Show abstract] [Hide abstract]
    ABSTRACT: This study investigated differences in event-related potential (ERP) parameters among early vascular dementia (VD) patients, healthy elder controls (ECs), and young controls (YCs). A visual "oddball" color identification task was performed while individuals' electroencephalograms (EEGs) were recorded. Approximate entropy (ApEn), a nonlinear measure, along with P300 latencies and amplitudes were used to analyze ERP data and compare these three groups. The patients with VD showed more complex ERP waveforms and higher ApEn values than did ECs while performing the visual task. It was further found that patients with VD showed reduced P300 amplitudes and increased latencies. The results indicate that patients with VD have fewer attention resources to devote to processing stimuli, lower speed of stimulus classification, and lower synchrony in their cortical activity during the response period. We suggest that ApEn, as a measure of ERP complexity, is a promising marker for early diagnosis of VD.
    Journal of clinical neurophysiology: official publication of the American Electroencephalographic Society 06/2012; 29(3):230-6. · 1.47 Impact Factor
  • Wutao Lou, Jin Xu, Hengsong Sheng, Songzhen Zhao
    [Show abstract] [Hide abstract]
    ABSTRACT: Multichannel EEG recorded in a task condition could contain more information about cognition. However, that has not been widely investigated in the vascular-dementia (VaD)- related studies. The purpose of this study was to explore the differences of brain functional states between VaD patients and normal controls while performing a detection task. Three multichannel linear descriptors, i.e. spatial complexity (Ω), field strength (Σ) and frequency of field changes (Φ), were applied to analyse four frequency bands (delta, theta, alpha and beta) of multichannel event-related EEG signals for 12 VaD patients (mean age ± SD: 69.25 ± 10.56 years ; MMSE score ± SD: 22.58 ± 4.42) and 12 age-matched healthy subjects (mean age ± SD: 67.17 ± 5.97 years ; MMSE score ± SD: 29.08 ± 0.9). The correlations between the three measures and MMSE scores were also analysed. VaD patients showed a significant higher Ω value in the delta (p = 0.013) and theta (p = 0.021) frequency bands, a lower Σ value (p = 0.011) and a higher Φ (p = 0.008) value in the delta frequency band compared with normal controls. The MMSE scores were negatively correlated with the Ω (r = -0.52, p = 0.01) and Φ (r = -0.47, p = 0.02) values in the delta frequency band. The results indicated the VaD patients presented a reduction of synchronization in the slow frequency band during target detection, and suggested more neurons might be activated in VaD patients compared with normal controls. The Ω and Φ measures in the delta frequency band might be used to evaluate the degree of cognitive dysfunction. The multichannel linear descriptors are promising measures to reveal the differences in brain functions between VaD patients and normal subjects, and could potentially be used to evaluate the degree of cognitive dysfunction in VaD patients.
    Clinical neurophysiology: official journal of the International Federation of Clinical Neurophysiology 05/2011; 122(11):2151-6. · 3.12 Impact Factor
  • Hengsong Sheng, Jin Xu, Chongxun Zheng
    [Show abstract] [Hide abstract]
    ABSTRACT: Vascular Dementia (VaD) can affect attention, memory and other cognitive function. For maximum benefits from new treatments, VaD should be diagnosed early. Electroencephalogram (EEG) give information about brain status and activity, so it is potential to use EEG-based method to improve early diagnose. In order to explore the brain activities in Vascular Dementia (VaD), S-transform, a time-frequency representation/distribution method and generalized S-transform are used to analyze EEG signals of VaD patients and normal elderly. EEG signals of VaD patients and normal elderly are collected. The results show EEG signals of VaD patients and normal elderly have different time-frequency presentation, probably reflects features of VaD.
    01/2010;