ABSTRACT
In this paper, we propose a multi-scale mel domain feature map extraction algorithm to solve the problem that the speech recognition rate of dysarthria is difficult to improve. We used the empirical mode decomposition method to decompose speech signals and extracted Fbank features and their first-order differences for each of the three effective components to construct a new feature map, which could capture details in the frequency domain. Secondly, due to the problems of effective feature loss and high computational complexity in the training process of single channel neural network, we proposed a speech recognition network model in this paper. Finally, training and decoding were performed on the public UA-Speech dataset. The experimental results showed that the accuracy of the speech recognition model of this method reached 92.77%. Therefore, the algorithm proposed in this paper can effectively improve the speech recognition rate of dysarthria.
Subject(s)
Humans , Dysarthria/diagnosis , Speech , Speech Perception , Algorithms , Neural Networks, ComputerABSTRACT
Objective: To explore the feasibility of arterial spin labeling (ASL) imaging in assessment of cerebral blood flow (CBF) in patients with internal carotid artery stenosis. Methods: Plain MR, DWI, three dimensional time of flight (3D-TOF) and ASL scanning were performed in 30 patients with unilateral internal carotid artery stenosis. Then the patients were divided into mild stenosis, moderate stenosis and severe stenosis cases according to 3D-TOF imaging. CBF of basal ganglia, centrum semiovale, anterior watershed, frontal lobe and temporal lobe were calculated and compared between the healthy side and the affected side. Results: One side internal carotid artery stenosis was mild in 11 cases, moderate in 9 cases and severe in 10 cases. There was no significant difference of CBF between the healthy side and the affected side of basal ganglia, centrum semiovale, anterior watershed, frontal lobe nor temporal lobe in patients with mild and moderate stenosis (all P> 0.05), while CBF of healthy side of each brain area in patients with severe stenosis were all higher than in the affected side (all P< 0.05). CBF value decreased with the increase of stenosis degree and negatively correlated with the latter (r=-0.966, P< 0.001). Conclusion: ASL can quantitatively and accurately evaluate CBF in patients with internal carotid artery stenosis, which is of great significance for guiding clinical treatment and evaluating curative effect.
ABSTRACT
Objective: To explore the effect of resveratrol (Res) on angiogenesis of human umbilical vein endothelial cells (HUVEC) with the possible mechanisms in vitro. Methods: The HUVECs were cultured in 6 groups.①Control group, HUVEC were cultured with high glucose in DMEM,②Res group, the cell were cultured with Res at different concentrations,③Res with PI3K blocker LY294002 (Res+Blocker 1) group, ④Blocker 1 group, HUVEC were cultured with LY294002 alone, ⑤Res with eNOS blocker L-NAME (Res +Blocker 2) group and ⑥Blocker 2 group. The effect of Res on HUVEC proliferation was detected by CCK-8 kit, the protein expressions of p-Akt, p-eNOS were examined by Westin blot analysis, nitric oxide (NO) level was measured by nitrate reduction method and the endothelial cell migration was assayed by transwell chamber method. Results: ① Compared with Control group, HUVEC proliferation increased in Res (1, 5μmol/L ) group, P0.05.④Compared with Control group, the cell migration and tubing formation were higher in Res (5μmol/L) group, P Conclusion: Res could up-regulate NO level via activating PI3-K/Akt/eNOS signaling and therefore, improving the proliferation, migration and tubing formation of HUVEC in vitro.