A novel feature-level fusion scheme with multimodal attention CNN for heart sound classification.

BACKGROUND AND OBJECTIVE: Most of the existing machine learning-based heart sound classification methods achieve limited accuracy. Since they primarily depend on single domain feature information and tend to focus equally on each part of the signal rather than employing a selective attention mechanism. In addition, they fail to exploit convolutional neural network (CNN) - based features with an effective fusion strategy. METHODS: In order to overcome these limitations, a novel multimodal attention convolutional neural network (MACNN) with a feature-level fusion strategy, in which Mel-cepstral domain as well as general frequency domain features are incorporated to increase the diversity of the features, is proposed in this paper. In the proposed method, DilationAttenNet is first utilized to construct attention-based CNN feature extractors and then these feature extractors are jointly optimized in MACNN at the feature-level. The attention mechanism aims to suppress irrelevant information and focus on crucial diverse features extracted from the CNN. RESULTS: Extensive experiments are carried out to study the efficacy of the feature level fusion in comparison to that with early fusion. The results show that the proposed MACNN method significantly outperforms the state-of-the-art approaches in terms of accuracy and score for the two publicly available Github and Physionet datasets. CONCLUSION: The findings of our experiments demonstrated the high performance for heart sound classification based on the proposed MACNN, and hence have potential clinical usefulness in the identification of heart diseases. This technique can assist cardiologists and researchers in the design and development of heart sound classification methods.

Site 1 protease aggravates acute kidney injury by promoting tubular epithelial cell ferroptosis through SIRT3-SOD2-mtROS signaling.

Ischemia/reperfusion (I/R)-induced acute kidney injury (AKI) is a common clinical syndrome with high morbidity and mortality. Ferroptosis, a newly discovered form of oxidative cell death, is involved in the pathogenesis of renal I/R injury; however, the underlying mechanism remains to be explored. Here, we reported that site 1 protease (S1P) promotes ischemic kidney injury by regulating ferroptotic cell death of tubular epithelial cells. S1P abundance was measured in hypoxia/reoxygenation (H/R)-treated Boston University mouse proximal tubular (BUMPT) cells and I/R-induced murine kidney tissue. S1P expression in BUMPT cells and kidneys was initially activated by hypoxic stimulation, accompanied by the ferroptotic response. Blocking S1P blunted H/R-induced ferroptotic cell death, which also restored sirtuin 3 (SIRT3) expression and superoxide dismutase 2 (SOD2) activity in BUMPT cells. Next, inhibition of S1P expression restored I/R-suppressed SIRT3 abundance, SOD2 activity and reduced the elevated level of mitochondria reactive oxygen species (mtROS), which attenuated tubular cell ferroptosis and renal I/R injury. In conclusion, S1P promoted renal tubular epithelial cell ferroptosis under I/R status by activating SIRT3-SOD2-mtROS signaling, thereby accelerating kidney injury. Thus, targeting S1P signaling may serve as a promising strategy for I/R kidney injury.

Hepatic resection prolongs overall survival in the selected patients with nasopharyngeal carcinoma liver metastases.

BACKGROUND: The role of surgery in nasopharyngeal carcinoma liver metastases (NCLM) remains elusive, and the current application is limited. We aim to investigate whether hepatic resection (HR) of NCLM improves survival compared with non-hepatic resection (NHR) treatment. METHODS: One hundred and thirty-three patients with NCLM from 2007 to 2017 were divided into two groups. Propensity score matching (PSM) analysis was used to compare the clinical outcomes. RESULTS: After PSM the median overall survival (OS) and the 1, 3 and 5-year OS rates in HR group were 32.60 months, 86.2%, 37.3% and 37.3%, respectively; while for NHR group these values were 19.57 months, 61.5%, 12.9% and 2.9%, respectively (P = 0.008). Multivariate analysis indicated hepatitis B virus infection (P = 0.029) and hepatic resection (P = 0.018) were independent prognostic factors. CONCLUSION: Our study revealed that hepatectomy yields a survival benefit safely compared with systemic treatments, especially for patients with the size of largest metastasis < 5 cm, unilobar distribution of liver tumor and received unanatomical hepatectomy.

Dietary folate drives methionine metabolism to promote cancer development by stabilizing MAT IIA.

Folic acid, served as dietary supplement, is closely linked to one-carbon metabolism and methionine metabolism. Previous clinical evidence indicated that folic acid supplementation displays dual effect on cancer development, promoting or suppressing tumor formation and progression. However, the underlying mechanism remains to be uncovered. Here, we report that high-folate diet significantly promotes cancer development in mice with hepatocellular carcinoma (HCC) induced by DEN/high-fat diet (HFD), simultaneously with increased expression of methionine adenosyltransferase 2A (gene name, MAT2A; protein name, MATIIα), the key enzyme in methionine metabolism, and acceleration of methionine cycle in cancer tissues. In contrast, folate-free diet reduces MATIIα expression and impedes HFD-induced HCC development. Notably, methionine metabolism is dynamically reprogrammed with valosin-containing protein p97/p47 complex-interacting protein (VCIP135) which functions as a deubiquitylating enzyme to bind and stabilize MATIIα in response to folic acid signal. Consistently, upregulation of MATIIα expression is positively correlated with increased VCIP135 protein level in human HCC tissues compared to adjacent tissues. Furthermore, liver-specific knockout of Mat2a remarkably abolishes the advocating effect of folic acid on HFD-induced HCC, demonstrating that the effect of high or free folate-diet on HFD-induced HCC relies on Mat2a. Moreover, folate and multiple intermediate metabolites in one-carbon metabolism are significantly decreased in vivo and in vitro upon Mat2a deletion. Together, folate promotes the integration of methionine and one-carbon metabolism, contributing to HCC development via hijacking MATIIα metabolic pathway. This study provides insight into folate-promoted cancer development, strongly recommending the tailor-made folate supplement guideline for both sub-healthy populations and patients with cancer expressing high level of MATIIα expression.

Comprehensive analysis of N6 -methyladenosine-related long non-coding RNAs for prognosis prediction in liver hepatocellular carcinoma.

BACKGROUND: Liver hepatocellular carcinoma (LIHC) is a lethal cancer. This study aimed to identify the N6 -methyladenosine (m6 A)-targeted long non-coding RNA (lncRNA) related to LIHC prognosis and to develop an m6 A-targeted lncRNA model for prognosis prediction in LIHC. METHODS: The expression matrix of mRNA and lncRNA was obtained, and differentially expressed (DE) mRNAs and lncRNAs between tumor and normal samples were identified. Univariate Cox and pathway enrichment analyses were performed on the m6 A-targeted lncRNAs and the LIHC prognosis-related m6 A-targeted lncRNAs. Prognostic analysis, immune infiltration, and gene DE analyses were performed on LIHC subgroups, which were obtained from unsupervised clustering analysis. Additionally, a multi-factor Cox analysis was used to construct a prognostic risk model based on the lncRNAs from the LASSO Cox model. Univariate and multivariate Cox analyses were used to assess prognostic independence. RESULTS: A total of 5031 significant DEmRNAs and 292 significant DElncRNAs were screened, and 72 LIHC-specific m6 A-targeted binding lncRNAs were screened. Moreover, a total of 29 LIHC prognosis-related m6 A-targeted lncRNAs were obtained and enriched in cytoskeletal, spliceosome, and cell cycle pathways. An 11-m6 A-lncRNA prognostic model was constructed and verified; the top 10 lncRNAs included LINC00152, RP6-65G23.3, RP11-620J15.3, RP11-290F5.1, RP11-147L13.13, RP11-923I11.6, AC092171.4, KB-1460A1.5, LINC00339, and RP11-119D9.1. Additionally, the two LIHC subgroups, Cluster 1 and Cluster 2, showed significant differences in the immune microenvironment, m6 A enzyme genes, and prognosis of LIHC. CONCLUSION: The m6 A-lncRNA prognostic model accurately and effectively predicted the prognostic survival of LIHC. Immune cells, immune checkpoints (ICs), and m6 A enzyme genes could act as novel therapeutic targets for LIHC.

An integrated biological-electrocatalytic process for highly-efficient treatment of coking wastewater.

Coking wastewater is typically refractory, mainly due to its biological toxicity and complex composition. In this study, a novel integrated biological-electrocatalytic process consisting of two three-dimensional electrochemical reactors (3DERs), two biological aerated filters (BAFs), and a three-dimensional biofilm electrode reactor (3DBER) is developed for the advanced treatment of coking wastewater. 73.21% of chemical oxygen demand (COD), 38.02% of ammonium nitrogen (NH4+-N) and 91.46% of nitrate nitrogen (NO3--N) are removed by 3DERs. BAFs mainly convert NH4+-N to NO3--N through microbial nitrification. The 3DBER removes the residual NO3--N by bio-electrochemical denitrification. The integrated system can eliminate 74.72-83.27% of COD, 99.38-99.74% of NH4+-N, and 69.64-99.83% of total nitrogen from coking wastewater during the continuous operation, as well as significantly reducing the toxicity of the wastewater. The superiorities of the integrated 3DERs/BAFs/3DBER system recommend the application of such biological-electrocatalytic technology in the treatment of highly toxic wastewater.

CapsCovNet: A Modified Capsule Network to Diagnose COVID-19 From Multimodal Medical Imaging.

Since the end of 2019, novel coronavirus disease (COVID-19) has brought about a plethora of unforeseen changes to the world as we know it. Despite our ceaseless fight against it, COVID-19 has claimed millions of lives, and the death toll exacerbated due to its extremely contagious and fast-spreading nature. To control the spread of this highly contagious disease, a rapid and accurate diagnosis can play a very crucial part. Motivated by this context, a parallelly concatenated convolutional block-based capsule network is proposed in this article as an efficient tool to diagnose the COVID-19 patients from multimodal medical images. Concatenation of deep convolutional blocks of different filter sizes allows us to integrate discriminative spatial features by simultaneously changing the receptive field and enhances the scalability of the model. Moreover, concatenation of capsule layers strengthens the model to learn more complex representation by presenting the information in a fine to coarser manner. The proposed model is evaluated on three benchmark datasets, in which two of them are chest radiograph datasets and the rest is an ultrasound imaging dataset. The architecture that we have proposed through extensive analysis and reasoning achieved outstanding performance in COVID-19 detection task, which signifies the potentiality of the proposed model.

S-Convnet: A Shallow Convolutional Neural Network Architecture for Neuromuscular Activity Recognition Using Instantaneous High-Density Surface EMG Images.

The recent progress in recognizing low-resolution instantaneous high-density surface electromyography (HD-sEMG) images opens up new avenues for the development of more fluid and natural muscle-computer interfaces. However, the existing approaches employed a very large deep convolutional neural network (ConvNet) architecture and complex training schemes for HD-sEMG image recognition, which requires learning of >5.63 million(M) training parameters only during fine-tuning and pre-trained on a very large-scale labeled HD-sEMG training dataset, as a result, it makes high-end resource-bounded and computationally expensive. To overcome this problem, we propose S-ConvNet models, a simple yet efficient framework for learning instantaneous HD-sEMG images from scratch using random-initialization. Without using any pre-trained models, our proposed S-ConvNet demonstrate very competitive recognition accuracy to the more complex state of the art, while reducing learning parameters to only ≈ 2M and using ≈ 12 × smaller dataset. The experimental results proved that the proposed S-ConvNet is highly effective for learning discriminative features for instantaneous HD-sEMG image recognition, especially in the data and high-end resource-constrained scenarios.

Sleep Apnea Event Detection from Sub-frame Based Feature Variation in EEG Signal Using Deep Convolutional Neural Network.

The topic of automatic detection of sleep apnea which is a respiratory sleep disorder, affecting millions of patients worldwide, is continuously being explored by researchers. Electroencephalogram signal (EEG) represents a promising tool due to its direct correlation to neural activity and ease of extraction. Here, an innovative approach is proposed to automatically detect apnea by incorporating local variations of temporal features for identifying the global feature variations over a broader window. An EEG data frame is divided into smaller sub-frames to effectively extract local feature variation within one larger frame. A fully convolutional neural network (FCNN) is proposed that will take each sub-frame of a single frame individually to extract local features. Following that, a dense classifier consisting of a series of fully connected layers is trained to analyze all the local features extracted from subframes for classifying the entire frame as apnea/non-apnea. Finally, a unique post-processing technique is applied which significantly improves accuracy. Both the EEG frame length and post-processing parameters are varied to find optimal detection conditions. Large-scale experimentation is executed on publicly available data of patients with varying apnea-hypopnea indices for performance evaluation of the suggested method.

The Long Noncoding RNA ANRIL Promotes Cell Apoptosis in Lipopolysaccharide-Induced Acute Kidney Injury Mediated by the TLR4/Nuclear Factor-Kappa B Pathway.

BACKGROUND/AIMS: The purpose of this study is to analyze the expression and biological function of lncRNA ANRIL, microRNA-199a, TLR4, and nuclear factor-kappa B (NF-κB) in acute renal injury (AKI) induced by lipopolysaccharide (LPS). METHODS: The levels of ANRIL and microRNA-199a in mouse cells and kidneys were detected by quantitative-polymerase chain reaction. Western blot analysis was used for the NF-κB pathway protein. MTT assay was used for cell viability. Enzyme-linked immunosorbent assay was used for the secretion of inflammatory factors in mouse kidney tissue. Apoptosis was measured by flow cytometry and Western blotting. The potential binding region between ANRIL and miR-199a was verified by luciferase reporter assay. RESULTS: The upregulation of ANRIL can reduce the expression of microRNA-199a and increases the number of apoptotic cells. The expression levels of ANRIL in LPS-induced AKI mice and LPS-treated HK2 cells were upregulated compared with the control group. Overexpression of ANRIL increased apoptosis and promoted TLR4 (Toll-like receptor 4), NF-κB phosphorylation, and downstream transcription factor production. CONCLUSION: ANRIL/NF-κB pathway in LPS-induced apoptosis provided theoretical guidance for ANRIL in the treatment of AKI.

Grayscale-Thermal Tracking via Inverse Sparse Representation based Collaborative Encoding.

Grayscale-thermal tracking has attracted a great deal of attention due to its capability of fusing two different yet complementary target observations. Existing methods often consider extracting the discriminative target information and exploring the target correlation among different images as two separate issues, ignoring their interdependence. This may cause tracking drifts in challenging video pairs. This paper presents a collaborative encoding model called joint correlation and discriminant analysis based inver-sparse representation (JCDA-InvSR) to jointly encode the target candidates in the grayscale and thermal video sequences. In particular, we develop a multi-objective programming to integrate the feature selection and the multi-view correlation analysis into a unified optimization problem in JCDA-InvSR, which can simultaneously highlight the special characters of the grayscale and thermal targets through alternately optimizing two aspects: the target discrimination within a given image and the target correlation across different images. For robust grayscale-thermal tracking, we also incorporate the prior knowledge of target candidate codes into the SVM based target classifier to overcome the overfitting caused by limited training labels. Extensive experiments on GTOT and RGBT234 datasets illustrate the promising performance of our tracking framework.

Olmesartan Medoxomil, An Angiotensin II-Receptor Blocker, Ameliorates Renal Injury In db/db Mice.

BACKGROUND: Diabetic nephropathy (DN) is a common complication of diabetes mellitus (DM) and also a major cause of end-stage renal disease (ESRD). Olmesartan medoxomil (OM) is an angiotensin II receptor blocker (ARB) and has been shown to exhibit renoprotective effects on a streptozotocin (STZ)-induced diabetic rat model. Yet, whether OM affects DN progression and renal injury in db/db mice, a type 2 diabetic murine model, has not been established. METHODS: Wild-type (n = 15) and db/db mice (n = 15) were treated with control saline or OM via oral gavage. The physiological and biochemical parameters were evaluated and histological examinations of kidney specimens were performed. RESULTS: Compared with saline-treated db/db mice, db/db mice administered with OM showed ameliorated diabetic physiological and biochemical parameters. In addition, OM decreased urinary albumin excretion and plasma creatinine level in db/db mice. Moreover, histologically, OM reduced glomerular hypertrophy and injury, and also ameliorated tubular injury, thus suggesting that OM improves renal function and minimizes renal pathological deterioration in db/db mice. CONCLUSION: Our study reveals a beneficial role of OM in ameliorating DN in db/db mice, which is associated with its renoprotective function.

Salidroside and FG-4592 ameliorate high glucose-induced glomerular endothelial cells injury via HIF upregulation.

Increasing research indicates that hyperglycemia plays a crucial role in the progression of diabetic nephropathy (DN); however, effective treatment for preventing or slowing DN progression are seriously lacking. Although salidroside (SAL) has been demonstrated to have a positive anti-diabetic effect, the cellular mechanisms remain unclear. FG-4592, a novel prolyl hydroxylase inhibitor, was used to regulate HIF-1α and HIF-2α expression. The present study aimed to explore the underlying mechanisms of SAL and FG-4592 on high glucose (HG)-induced rat glomerular endothelial cells (rGECs) injury. HG-cultured rGECs were used to induce a diabetic environment. An MTT assay, RT-qPCR, Western blot, flow cytometry, and immunofluorescent staining were performed to investigate the effects of SAL on HG-induced rGECs injury. FG-4592 and SAL protected rGECs against HG-induced injury by increasing cellular viability and reducing the cell apoptosis rate. SAL and FG-4592 downregulated PHD-2 expression and upregulated HIF-1α and HIF-2α expression. In conclusion, our findings suggest that SAL and FG-4592 ameliorate HG-induced rGEC injury by upregulating HIF expression, indicating that SAL and FG-4592 might be favorable for further DN-treatment.

Sleep Apnea Detection Based on Rician Modeling of Feature Variation in Multiband EEG Signal.

Sleep apnea, a serious sleep disorder affecting a large population, causes disruptions in breathing during sleep. In this paper, an automatic apnea detection scheme is proposed using single lead electroencephalography (EEG) signal to discriminate apnea patients and healthy subjects as well as to deal with the difficult task of classifying apnea and nonapnea events of an apnea patient. A unique multiband subframe based feature extraction scheme is developed to capture the feature variation pattern within a frame of EEG data, which is shown to exhibit significantly different characteristics in apnea and nonapnea frames. Such within-frame feature variation can be better represented by some statistical measures and characteristic probability density functions. It is found that use of Rician model parameters along with some statistical measures can offer very robust feature qualities in terms of standard performance criteria, such as Bhattacharyya distance and geometric separability index. For the purpose of classification, proposed features are used in K Nearest Neighbor classifier. From extensive experimentations and analysis on three different publicly available databases it is found that the proposed method offers superior classification performance in terms of sensitivity, specificity, and accuracy.

Cluster based statistical feature extraction method for automatic bleeding detection in wireless capsule endoscopy video.

Wireless capsule endoscopy (WCE) is capable of demonstrating the entire gastrointestinal tract at an expense of exhaustive reviewing process for detecting bleeding disorders. The main objective is to develop an automatic method for identifying the bleeding frames and zones from WCE video. Different statistical features are extracted from the overlapping spatial blocks of the preprocessed WCE image in a transformed color plane containing green to red pixel ratio. The unique idea of the proposed method is to first perform unsupervised clustering of different blocks for obtaining two clusters and then extract cluster based features (CBFs). Finally, a global feature consisting of the CBFs and differential CBF is used to detect bleeding frame via supervised classification. In order to handle continuous WCE video, a post-processing scheme is introduced utilizing the feature trends in neighboring frames. The CBF along with some morphological operations is employed to identify bleeding zones. Based on extensive experimentation on several WCE videos, it is found that the proposed method offers significantly better performance in comparison to some existing methods in terms of bleeding detection accuracy, sensitivity, specificity and precision in bleeding zone detection. It is found that the bleeding detection performance obtained by using the proposed CBF based global feature is better than the feature extracted from the non-clustered image. The proposed method can reduce the burden of physicians in investigating WCE video to detect bleeding frame and zone with a high level of accuracy.

Dihydropyrimidine dehydrogenase predicts survival and response to interferon-α in hepatocellular carcinoma.

Metastasis and recurrence contribute to poor prognosis of hepatocellular carcinoma (HCC). Recently, we reported that interferon-α (IFN-α) can suppress metastasis of HCC; however, the underlying mechanism has not been fully described. In this study, we demonstrated that expression of dihydropyrimidine dehydrogenase (DPYD), a pyrimidine catabolic enzyme, was dose-dependently downregulated by IFN-α in HCC tissues from nude mice. Notably, DPYD expression was found to be significantly increased in HCC cell lines with higher metastatic potentials compared with their controls. Moreover, upregulation of DPYD in HCC cells could promote in vitro migration, invasion, and in vivo lung metastasis, and inducing changes characteristic of epithelial-mesenchymal transition (EMT). In contrast, knockdown of DPYD inhibited these processes. Mechanistically, DPYD functioned as a positive regulator of EMT in HCC by targeting the p38/NF-κB/Snail1 pathway. Clinically, tissue microarray analysis showed that high DPYD expression was positively associated with aggressive tumor characteristics, including larger tumor size, tumor recurrence, and advanced tumor node metastasis (TNM) stage, and independently correlated with poorer overall survival times after curative resection. HCC patients with low DPYD expression have better response to IFN-α therapy. Taken together, our findings elucidate that IFN-α could downregulate DPYD expression to inhibit EMT and HCC metastasis, and suggest that DPYD might be a potential prognostic biomarker and a therapeutic target for HCC.

MicroRNA-26a suppresses epithelial-mesenchymal transition in human hepatocellular carcinoma by repressing enhancer of zeste homolog 2.

BACKGROUND: Our previous study reported that microRNA-26a (miR-26a) inhibited tumor progression by inhibiting tumor angiogenesis and intratumoral macrophage infiltration in hepatocellular carcinoma (HCC). The direct roles of miR-26a on tumor cell invasion remain poorly understood. In this study, we aim to explore the mechanism of miR-26a in modulating epithelial-mesenchymal transition (EMT) in HCC. METHODS: In vitro cell morphology and cell migration were compared between the hepatoma cell lines HCCLM3 and HepG2, which were established in the previous study. Overexpression and down-regulation of miR-26a were induced in these cell lines, and Western blot and immunofluorescence assays were used to detect the expression of EMT markers. Xenograft nude mouse models were used to observe tumor growth and pulmonary metastasis. Immunohistochemical assays were conducted to study the relationships between miR-26a expression and enhancer of zeste homolog 2 (EZH2) and E-cadherin expression in human HCC samples. RESULTS: Down-regulation of miR-26a in HCCLM3 and HepG2 cells resulted in an EMT-like cell morphology and high motility in vitro and increased in tumor growth and pulmonary metastasis in vivo. Through down-regulation of EZH2 expression and up-regulation of E-cadherin expression, miR-26a inhibited the EMT process in vitro and in vivo. Luciferase reporter assay showed that miR-26a directly interacted with EZH2 messenger RNA (mRNA). Furthermore, the expression of miR-26a was positively correlated with E-cadherin expression and inversely correlated with EZH2 expression in human HCC tissue. CONCLUSIONS: miR-26a inhibited the EMT process in HCC by down-regulating EZH2 expression.

DCT domain feature extraction scheme based on motor unit action potential of EMG signal for neuromuscular disease classification.

A feature extraction scheme based on discrete cosine transform (DCT) of electromyography (EMG) signals is proposed for the classification of normal event and a neuromuscular disease, namely the amyotrophic lateral sclerosis. Instead of employing DCT directly on EMG data, it is employed on the motor unit action potentials (MUAPs) extracted from the EMG signal via a template matching-based decomposition technique. Unlike conventional MUAP-based methods, only one MUAP with maximum dynamic range is selected for DCT-based feature extraction. Magnitude and frequency values of a few high-energy DCT coefficients corresponding to the selected MUAP are used as the desired feature which not only reduces computational burden, but also offers better feature quality with high within-class compactness and between-class separation. For the purpose of classification, the K-nearest neighbourhood classifier is employed. Extensive analysis is performed on clinical EMG database and it is found that the proposed method provides a very satisfactory performance in terms of specificity, sensitivity and overall classification accuracy.