Image copy-move forgery detection and localization based on super-BPD segmentation and DCNN.

With the increasing importance of image information, image forgery seriously threatens the security of image content. Copy-move forgery detection (CMFD) is a greater challenge because its abnormality is smaller than other forgeries. To solve the problem that the detection results of the most image CMFD based on convolutional neural networks (CNN) have relatively low accuracy, an image copy-move forgery detection and localization based on super boundary-to-pixel direction (super-BPD) segmentation and deep CNN (DCNN) is proposed: SD-Net. Firstly, the segmentation technology is used to enhance the connection between the same or similar image blocks, improving the detection accuracy. Secondly, DCNN is used to extract image features, replacing conventional hand-crafted features with automatic learning features. The feature pyramid is used to improve the robustness to the scaling attack. Thirdly, the image BPD information is used to optimize the edges of rough detected image and obtain final detected image. The experiments proved that the SD-Net could detect and locate multiple, rotated, and scaling forgery well, especially large-level scaling forgery. Compared with other methods, the SD-Net is more accurately located and robust to various post-processing operations: brightness change, contrast adjustments, color reduction, image blurring, JPEG compression, and noise adding.

Acoustic scene classification based on three-dimensional multi-channel feature-correlated deep learning networks.

As an effective approach to perceive environments, acoustic scene classification (ASC) has received considerable attention in the past few years. Generally, ASC is deemed a challenging task due to subtle differences between various classes of environmental sounds. In this paper, we propose a novel approach to perform accurate classification based on the aggregation of spatial-temporal features extracted from a multi-branch three-dimensional (3D) convolution neural network (CNN) model. The novelties of this paper are as follows. First, we form multiple frequency-domain representations of signals by fully utilizing expert knowledge on acoustics and discrete wavelet transformations (DWT). Secondly, we propose a novel 3D CNN architecture featuring residual connections and squeeze-and-excitation attentions (3D-SE-ResNet) to effectively capture both long-term and short-term correlations inherent in environmental sounds. Thirdly, an auxiliary supervised branch based on the chromatogram of the original signal is incorporated in the proposed architecture to alleviate overfitting risks by providing supplementary information to the model. The performance of the proposed multi-input multi-feature 3D-CNN architecture is numerically evaluated on a typical large-scale dataset in the 2019 IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2019) and is shown to obtain noticeable performance gains over the state-of-the-art methods in the literature.

Channel estimation based on superimposed pilot and weighted averaging.

Channel estimation based on superimposed pilot (SP) is a challenge in orthogonal frequency division multiplexing (OFDM) systems. To reduce the pilot data interference in the SP and estimate the channel state information accurately, a weighted averaging (WA) channel estimation method based on the superimposed pilot is proposed in this paper. At the transmitter, two signals with data symbols and pilot symbols superimposed at different subcarriers are transmitted. At the receiver, the elimination scheme is proposed to remove the pilot data interference. Based on the temporal correlation of wireless channels, the WA method is used to reduce the interference caused by additive white Gaussian noise in the channel. Simulation results verify that the proposed method can be applied to different channel scenarios. It has better normalized mean square error and bit error rate performance than other existing methods, and the superimposed pilot can improve the throughput of wireless communication systems.

Exosomal microRNA-133b-3p from bone marrow mesenchymal stem cells inhibits angiogenesis and oxidative stress via FBN1 repression in diabetic retinopathy.

Diabetic retinopathy (DR) is a common microvascular complication. Many studies have focused on the role of microRNAs (miRNAs) in DR but not specifically on miR-133b-3p. Thus, this study is to unmask the mechanisms of miR-133b-3p in DR. KK/Upj-Ay mice (a spontaneous diabetic nephropathy model of DM, referred to as DR mice) were used in the study, and retinal tissues were collected. Bone marrow mesenchymal stem cells (BMSCs) were isolated and identified. High glucose (HG)-treated mouse retinal microvascular endothelial cells (mRMECs) were transfected or co-cultured with BMSCs-derived exosomes. Then, cell proliferation, migration, apoptosis, angiogenesis, and oxidative stress were observed. MiR-133b-3p and FBN1 expression in tissues and cells was detected. MiR-133b-3p expression was reduced, and FBN1 expression was increased in retinal tissues of DR mice and HG-treated mRMECs. Up-regulating miR-133b-3p or down-regulating FBN1 or BMSCs-derived exosomes impaired oxidative stress, angiogenesis, proliferation, migration, and promoted apoptosis of HG-treated mRMECs. This study has elucidated that exosomal miR-133b-3p from BMSCs suppresses angiogenesis and oxidative stress in DR via FBN1 repression.

CircZNF532 knockdown protects retinal pigment epithelial cells against high glucose-induced apoptosis and pyroptosis by regulating the miR-20b-5p/STAT3 axis.

INTRODUCTION: The loss of retinal pigment epithelial (RPE) cells is associated with the etiology of diabetic retinopathy (DR). This study investigated the effects of circular RNA ZNF532 (circZNF532) on apoptosis and pyroptosis of RPE cells. MATERIALS AND METHODS: Blood samples were collected from patients with DR and healthy volunteers. A human RPE cell line ARPE-19 was induced by high glucose (HG) and assayed for cell viability, apoptosis, and pyroptosis. The binding of miR-20b-5p with circZNF532 and STAT3 was confirmed by a luciferase activity assay. A mouse model of diabetic retinopathy was established. RESULTS: CircZNF532 and STAT3 were upregulated but miR-20b-5p was downregulated in the serum samples of patients with DR and HG-induced ARPE-19 cells. Elevated miR-20b-5p or CircZNF532 knockdown enhanced proliferation but reduced apoptosis and pyroptosis of ARPE-19 cells. CircZNF532 sponged miR-20b-5p and inhibited its expression. STAT3 was verified as a target of miR-20b-5p. MiR-20b-5p modulated ARPE-19 cell viability, apoptosis, and pyroptosis by targeting STAT3. Mice with STZ-induced diabetes showed elevated expressions of circZNF532 and STAT3 but decreased the level of miR-20b-5p compared with the controls. Knockdown of circZNF532 inhibited apoptosis and pyroptosis in mouse retinal tissues. CONCLUSION: CircZNF532 knockdown rescued human RPE cells from HG-induced apoptosis and pyroptosis by regulating STAT3 via miR-20b-5p.

Selective ablation of the ligament of Marshall reduces ischemia and reperfusion-induced ventricular arrhythmias.

Cardiac sympathetic tone overdrive is a key mechanism of arrhythmia. Cardiac sympathetic nerves denervation, such as LSG ablation or renal sympathetic denervation, suppressed both the prevalence of VAs and the incidence of SCD. Accumulating evidence demonstrates the ligament of Marshall (LOM) is a key component of the sympathetic conduit between the left stellate ganglion (LSG) and the ventricles. The present study aimed to investigate the roles of the distal segment of LOM (LOMLSPV) denervation in ischemia and reperfusion (IR)-induced VAs, and compared that LSG denervation. Thirty-three canines were randomly divided into group 1 (IR group, n = 11), group 2 (LOMLSPV Denervation + IR, n = 9), and group 3 (LSG Denervation + IR, n = 13). Hematoxylin-Eosin (HE) and Immunohistochemistry staining revealed that LOMLSPV contained bundles of sympathetic but not parasympathetic nerves. IR increased the cardiac sympathetic tone [serum concentrations of noradrenaline (NE) and epinephrine (E)] and induced the prevalence of VAs [ventricular premature beat (VPB), salvo of VPB, ventricular tachycardia (VT), VT duration (VTD) and ventricular fibrillation (VF)]. Both LOMLSPV denervation and LSG denervation could reduce the cardiac sympathetic tone in Baseline (BS) [heart rate variability (HRV)]. Compared with group 1, LOMLSPV denervation and LSG denervation similarly reduced sympathetic tone [NE (1.39±0.068 ng/ml in group 2, 1.29±0.081 ng/ml in group 3 vs 2.32±0.17 ng/ml in group 1, P<0.05) and E (114.64±9.22 pg/ml in group 2, 112.60±9.69 pg/ml in group 3 vs 166.18±15.78 pg/ml in group 1, P<0.05),] and VAs [VT (0±3.00 in group 2, 0±1.75 in group 3 vs 8.00±11.00 in group 1, P<0.05) and VTD (0 ± 4 s in group 2, 0±0.88s in group 3 vs 10.0 ± 22.00s in group 1, P<0.05)] after 2h reperfusion. These findings indicated LOMLSPV denervation reduced the prevalence of VT by suppressing SNS activity. These effects are comparable to those of LSG denervation. In myocardial IR, the anti-arrhythmic effects of LOMLSPV Denervation may be related to the inhibition of the expression of NE and E.

Selective ablation of the ligament of Marshall attenuates atrial electrical remodeling in a short-term rapid atrial pacing canine model.

INTRODUCTION: Cardiac sympathetic activation facilitates atrial electrical remodeling during atrial fibrillation (AF). Selective ablation of the distal part of the ligament of Marshall (LOMLSPV ) could decrease cardiac sympathetic innervation. This study aimed to investigate the effects of LOMLSPV ablation on atrial electrical remodeling in a short-term rapid atrial pacing (RAP) model. METHODS: In 16 anesthetized dogs, 6 hours of RAP (20 Hz, 2 × threshold) was delivered before LOMLSPV ablation (group 1, N  =  8) or after (group 2, N  =  8). Heart rate variability (HRV), serum norepinephrine (NE), atrial electrophysiological indices were analyzed. Six times of burst pacing (20 Hz, 2 × threshold, lasting for 5 seconds, were performed to induce AF, the number of episodes and the duration of AF were compared. RESULTS: LOMLSPV ablation decreased sympathetic indices of HRV and serum NE. Atrial effective refractory period (ERP) was shortened during RAP in both groups with higher reduction degrees in group 1. In group 1, the shortening of atrial ERP, elevating of ERP dispersion and sum of window of vulnerability (ΣWOV), facilitating of AF induced by RAP were subsequently reversed by LOMLSPV ablation. In group 2, LOMLSPV ablation prolonged atrial ERP, decreased ΣWOV, eliminated AF induction. The subsequent RAP failed to alter these indices. Histological studies showed abundant sympathetic nerve fibers in LOMLSPV . CONCLUSION: LOMLSPV ablation could inhibit atrial electrical remodeling during short-term RAP by reducing the cardiac sympathetic activity. LOMLSPV may be a potential target in AF ablation, especially in patients with highly cardiac sympathetic activation or atrial electrical remodeling.

Ablation of the Ligament of Marshall and Left Stellate Ganglion Similarly Reduces Ventricular Arrhythmias During Acute Myocardial Infarction.

BACKGROUND: Sympathetic denervation exerts protective effects against ventricular arrhythmias (VAs) induced by acute myocardial infarction. The results of a previous study indicated that the distal part of the ligament of Marshall (LOMLSPV) might be a sympathetic conduit between the left stellate ganglion (LSG) and the ventricles. The present study was designed to compare the effects between LSG and LOMLSPV ablation on ischemia-induced VAs. METHODS: Twenty-nine dogs were randomly divided into sham ablation group (group 1, n=9), LOMLSPV ablation group (group 2, n=10), and LSG ablation group (group 3, n=10). Ablation was performed before occlusion of the left anterior coronary artery. Changes in the heart rate variability, serum norepinephrine, ventricular effective refractory period, and blood pressure induced by LSG stimulation were observed, and the occurrence of VAs was recorded. Immunostaining examinations of LOMLSPV were performed in dogs without ablation. RESULTS: In group 2, LOMLSPV ablation evidently attenuated blood pressure elevation induced by LSG stimulation. Both LOMLSPV ablation and LSG ablation similarly prolonged ventricular effective refractory period and reduced the concentration of serum norepinephrine, the sympathetic index of heart rate variability, and the incidence of VAs compared with sham ablation. Abundant sympathetic nerve fibers were observed in LOMLSPV. CONCLUSIONS: LOMLSPV ablation prevented acute myocardial infarction-induced VAs with the same efficiency as LSG ablation, potentially by blocking the sympathetic pathway from the LSG to the heart.

Stimulation of ganglionated plexus attenuates cardiac neural remodeling and heart failure progression in a canine model of acute heart failure post-myocardial infarction.

BACKGROUND: Heart failure (HF) is associated with autonomic dysfunction. Vagus nerve stimulation has been shown to improve cardiac function both in HF patients and animal models of HF. The purpose of this present study is to investigate the effects of ganglionated plexus stimulation (GPS) on HF progression and autonomic remodeling in a canine model of acute HF post-myocardial infarction. METHODS AND RESULTS: Eighteen adult mongrel male dogs were randomized into the control (n=8) and GPS (n=10) groups. All dogs underwent left anterior descending artery ligation followed by 6-hour high-rate (180-220bpm) ventricular pacing to induce acute HF. Transthoracic 2-dimensional echocardiography was performed at different time points. The plasma levels of norepinephrine, B-type natriuretic peptide (BNP) and Ang-II were measured using ELISA kits. C-fos and nerve growth factor (NGF) proteins expressed in the left stellate ganglion as well as GAP43 and TH proteins expressed in the peri-infarct zone were measured using western blot. After 6h of GPS, the left ventricular end-diastolic volume, end-systolic volume and ejection fraction showed no significant differences between the 2 groups, but the interventricular septal thickness at end-systole in the GPS group was significantly higher than that in the control group. The plasma levels of norepinephrine, BNP, Ang-II were increased 1h after myocardial infarction while the increase was attenuated by GPS. The expression of c-fos and NGF proteins in the left stellate ganglion as well as GAP43 and TH proteins in cardiac peri-infarct zone in GPS group were significantly lower than that in control group. CONCLUSIONS: GPS inhibits cardiac sympathetic remodeling and attenuates HF progression in canines with acute HF induced by myocardial infarction and ventricular pacing.