RESUMEN
Objective. The textures and detailed structures in computed tomography (CT) images are highly desirable for clinical diagnosis. This study aims to expand the current body of work on textures and details preserving convolutional neural networks for low-dose CT (LDCT) image denoising task.Approach. This study proposed a novel multi-scale feature aggregation and fusion network (MFAF-net) for LDCT image denoising. Specifically, we proposed a multi-scale residual feature aggregation module to characterize multi-scale structural information in CT images, which captures regional-specific inter-scale variations using learned weights. We further proposed a cross-level feature fusion module to integrate cross-level features, which adaptively weights the contributions of features from encoder to decoder by using a spatial pyramid attention mechanism. Moreover, we proposed a self-supervised multi-level perceptual loss module to generate multi-level auxiliary perceptual supervision for recovery of salient textures and structures of tissues and lesions in CT images, which takes advantage of abundant semantic information at various levels. We introduced parameters for the perceptual loss to adaptively weight the contributions of auxiliary features of different levels and we also introduced an automatic parameter tuning strategy for these parameters.Main results. Extensive experimental studies were performed to validate the effectiveness of the proposed method. Experimental results demonstrate that the proposed method can achieve better performance on both fine textures preservation and noise suppression for CT image denoising task compared with other competitive convolutional neural network (CNN) based methods.Significance. The proposed MFAF-net takes advantage of multi-scale receptive fields, cross-level features integration and self-supervised multi-level perceptual loss, enabling more effective recovering of fine textures and detailed structures of tissues and lesions in CT images.
Asunto(s)
Procesamiento de Imagen Asistido por Computador , Tomografía Computarizada por Rayos X , Tomografía Computarizada por Rayos X/métodos , Procesamiento de Imagen Asistido por Computador/métodos , Humanos , Redes Neurales de la Computación , Dosis de Radiación , Relación Señal-RuidoRESUMEN
Background: Low-dose computed tomography (LDCT) scans can effectively reduce the radiation damage to patients, but this is highly detrimental to CT image quality. Deep convolutional neural networks (CNNs) have shown their potential in improving LDCT image quality. However, the conventional CNN-based approaches rely fundamentally on the convolution operations, which are ineffective for modeling the correlations among nonlocal similar structures and the regionally distinct statistical properties in CT images. This modeling deficiency hampers the denoising performance for CT images derived in this manner. Methods: In this paper, we propose an adaptive global context (AGC) modeling scheme to describe the nonlocal correlations and the regionally distinct statistics in CT images with negligible computation load. We further propose an AGC-based long-short residual encoder-decoder (AGC-LSRED) network for efficient LDCT image noise artifact-suppression tasks. Specifically, stacks of residual AGC attention blocks (RAGCBs) with long and short skip connections are constructed in the AGC-LSRED network, which allows valuable structural and positional information to be bypassed through these identity-based skip connections and thus eases the training of the deep denoising network. For training the AGC-LSRED network, we propose a compound loss that combines the L1 loss, adversarial loss, and self-supervised multi-scale perceptual loss. Results: Quantitative and qualitative experimental studies were performed to verify and validate the effectiveness of the proposed method. The simulation experiments demonstrated the proposed method exhibits the best result in terms of noise suppression [root-mean-square error (RMSE) =9.02; peak signal-to-noise ratio (PSNR) =33.17] and fine structure preservation [structural similarity index (SSIM) =0.925] compared with other competitive CNN-based methods. The experiments on real data illustrated that the proposed method has advantages over other methods in terms of radiologists' subjective assessment scores (averaged scores =4.34). Conclusions: With the use of the AGC modeling scheme to characterize the structural information in CT images and of residual AGC-attention blocks with long and short skip connections to ease the network training, the proposed AGC-LSRED method achieves satisfactory results in preserving fine anatomical structures and suppressing noise in LDCT images.
RESUMEN
It is urgent to identify the development of the Corona Virus Disease 2019 (COVID-19) in countries around the world. Therefore, visualization is particularly important for monitoring the COVID-19. In this paper, we visually analyze the real-time data of COVID-19, to monitor the trend of COVID-19 in the form of charts. At present, the COVID-19 is still spreading. However, in the existing works, the visualization of COVID-19 data has not established a certain connection between the forecast of the epidemic data and the forecast of the epidemic. To better predict the development trend of the COVID-19, we establish a logistic growth model to predict the development of the epidemic by using the same data source in the visualization. However, the logistic growth model only has a single feature. To predict the epidemic situation in an all-round way, we also predict the development trend of the COVID-19 based on the Susceptible Exposed Infected Removed epidemic model with multiple features. We fit the data predicted by the model to the real COVID-19 epidemic data. The simulation results show that the predicted epidemic development trend is consistent with the actual epidemic development trend, and our model performs well in predicting the trend of COVID-19.
RESUMEN
Healthcare uses state-of-the-art technologies (such as wearable devices, blood glucose meters, electrocardiographs), which results in the generation of large amounts of data. Healthcare data is essential in patient management and plays a critical role in transforming healthcare services, medical scheme design, and scientific research. Missing data is a challenging problem in healthcare due to system failure and untimely filing, resulting in inaccurate diagnosis treatment anomalies. Therefore, there is a need to accurately predict and impute missing data as only complete data could provide a scientific and comprehensive basis for patients, doctors, and researchers. However, traditional approaches in this paradigm often neglect the effect of the time factor on forecasting results. This paper proposes a time-aware missing healthcare data prediction approach based on the autoregressive integrated moving average (ARIMA) model. We combine a truncated singular value decomposition (SVD) with the ARIMA model to improve the prediction efficiency of the ARIMA model and remove data redundancy and noise. Through the improved ARIMA model, our proposed approach (named MHDP SVD_ARIMA) can capture underlying pattern of healthcare data changes with time and accurately predict missing data. The experiments conducted on the WISDM dataset show that MHDP SVD_ARIMA approach is effective and efficient in predicting missing healthcare data.
RESUMEN
It is well known that non-small cell lung cancer (NSCLC) is a malignant tumor with high incidence in the world. We aimed to clarify a possible target and identify its precise molecular biological mechanism in NSCLC. NLR family CARD domain containing 5 (NLRC5) is widely expressed in tissues and exerts a vital role in anti-tumor immunity. We determined NLRC5 expression by RT-qPCR and western blot assay. The role of NLRC5 in the development of NSCLC was assessed by a loss-of-function assay. CCK-8, Annexin-V-FITC/PI Apoptosis Detection Kit, Transwell, and wound healing assays were used to determine the cell functions. Drug resistance-related proteins were analyzed by western blot assay. Furthermore, the modulation of NLRC5 on carcinoembryonic antigen-related cell adhesion molecule 1 (CEACAM1) expression and subsequent PI3K/AKT signaling was assessed. In this study, a hyper-expression of NLRC5 was found in NSCLC tissues and cell lines. Knockdown of NLRC5 suppressed cell viability, invasion, and migration, and furthermore promoted cell apoptosis in NSCLC cells. Moreover, under normoxia or hypoxia treatment, the upregulation of NLRC5 was related to carboplatin resistance. NLRC5 silencing increased carboplatin-resistant cell chemosensitivity, as evidenced by the increase in the cell inhibition rate and decrease in drug resistance-related protein expression. Mechanistically, NLRC5 knockdown inhibited the expression of CEACAM1 and subsequently blocked the PI3K/AKT signaling pathway. In conclusion, NLRC5 promotes the malignant biological behaviors of NSCLC cells by activating the PI3K/AKT signaling pathway via the regulation of CEACAM1 expression under normoxia and hypoxia.
Asunto(s)
Carcinoma de Pulmón de Células no Pequeñas , Neoplasias Pulmonares , Humanos , Carcinoma de Pulmón de Células no Pequeñas/patología , Proteínas Proto-Oncogénicas c-akt/metabolismo , Carboplatino/farmacología , Fosfatidilinositol 3-Quinasas/metabolismo , Neoplasias Pulmonares/patología , Antígeno Carcinoembrionario , Molécula 1 de Adhesión Celular , Dominio de Reclutamiento y Activación de Caspasas , Factores de Transcripción , Proliferación Celular/genética , Hipoxia , Línea Celular Tumoral , Movimiento Celular/genéticaRESUMEN
BACKGROUND: Lung cancer affects approximately 9% of women and 17% of men worldwide, and has a mortality rate of 17%. Previously published studies have suggested that oxidative stress expansion can lead to lung cancer. The aim of the current study was to analyze the possible inhibitory pathway of atorvastatin against lung cancer cells in an in vivo model. METHODS: The cytotoxic effects of atorvastatin on lung cancer cell lines H460 and A549 were analyzed, as well as cell cycle arrest and cell morphology. Benzo(a)pyrene (BaP) was used for the induction of lung cancer in experimental rats, and atorvastatin (5, 10, and 20 mg/kg body weight) was used for treatment in a dose-dependent manner. Body weight and lung tumors were calculated at regular intervals. Antioxidants, pro-inflammatory cytokines, phase I and II antioxidant enzymes, polyamine enzymes, and apoptosis markers were determined at end of the experimental study. RESULTS: Cell cycle arrest occurred at the G2/M phase after atorvastatin treatment. Atorvastatin increased cytochrome C expression and caspase activity in a dose-dependent manner, and increased the activity of antioxidative enzymes, such as GPx, SOD, GST, reduced glutathione, and catalase, and reduced the level of nitrate and LPO. It also altered the xanthine oxidase (XO), Lactic Acid Dehydrogenase (LDH), quinone reductase (QR), UDP-glucuronosyltransferase (UDP-GT), adenosine deaminase (ADA), Aryl hydrocarbon hydroxylase (AHH), 5'-nucleotidase, cytochrome P450, cytochrome B5 and NADPH cytochrome C reductase levels. Atorvastatin was found to modulate polyamine enzyme levels, such as histamine, spermine, spermidine, and putrescine, and significantly (P<0.001) reduced the pro-inflammatory cytokine levels, such as tumor necrosis factor-α. Interleukin (IL)-6 and interleukin-1ß (IL-1ß) increased caspase-3 and caspase-9 levels in a dose-dependent manner. CONCLUSIONS: Our findings indicate that atorvastatin can inhibit lung cancer through apoptosis.
RESUMEN
Non-small cell lung cancer (NSCLC) is a common malignancy worldwide. MicroRNA (miR)-217 and sirtuin 1 (SIRT1) have been reported to play significant roles in different types of cancer, such as osteosarcoma and prostate cancer; however, the association between miR-217 and SIRT1 in the cell proliferation, apoptosis and invasion of NSCLC remain unknown. Thus, the present study aimed to investigate the roles of miR-217 and SIRT1 in NSCLC. The expression levels of miR-217 and SIRT1 were detected via reverse transcription-quantitative (RT-q)PCR and western blot analyses. The effect of miR-217 on A549 and H1299 cell proliferation, apoptosis and invasion was assessed via the Cell Counting Kit-8, flow cytometry and Transwell assays, respectively. In addition, the association between SIRT1 and miR-217 was predicted using the TargetScan database, and verified via the dual-luciferase reporter assay, and RT-qPCR and western blot analyses. The results demonstrated that miR-217 expression was significantly downregulated, while SIRT1 expression was significantly upregulated in A549 and H1299 cells compared with the human bronchial epithelial cells. Furthermore, transfection with miR-217 mimic significantly inhibited A549 and H1299 cell proliferation and invasion, and induced A549 and H1299 cell apoptosis. The results of the dual-luciferase reporter assay and western blot analysis confirmed that SIRT1 is a target gene of miR-217. In addition, miR-217 inhibited the activation of AMP-activated protein kinase (AMPK) and mTOR signaling. Taken together, the results of the present study suggest that miR-217 inhibits A549 and H1299 cell proliferation and invasion, and induces A549 and H1299 cell apoptosis by targeting SIRT1 and inactivating the SIRT1-mediated AMPK/mTOR signaling pathway. Thus, miR-217 may be used as a potential therapeutic target for the treatment of patients with NSCLC.
RESUMEN
A least squares support vector machine (LS-SVM) offers performance comparable to that of SVMs for classification and regression. The main limitation of LS-SVM is that it lacks sparsity compared with SVMs, making LS-SVM unsuitable for handling large-scale data due to computation and memory costs. To obtain sparse LS-SVM, several pruning methods based on an iterative strategy were recently proposed but did not consider the quantity constraint on the number of reserved support vectors, as widely used in real-life applications. In this article, a noniterative algorithm is proposed based on the selection of globally representative points (global-representation-based sparse least squares support vector machine, GRS-LSSVM) to improve the performance of sparse LS-SVM. For the first time, we present a model of sparse LS-SVM with a quantity constraint. In solving the optimal solution of the model, we find that using globally representative points to construct the reserved support vector set produces a better solution than other methods. We design an indicator based on point density and point dispersion to evaluate the global representation of points in feature space. Using the indicator, the top globally representative points are selected in one step from all points to construct the reserved support vector set of sparse LS-SVM. After obtaining the set, the decision hyperplane of sparse LS-SVM is directly computed using an algebraic formula. This algorithm only consumes O(N2) in computational complexity and O(N) in memory cost which makes it suitable for large-scale data sets. The experimental results show that the proposed algorithm has higher sparsity, greater stability, and lower computational complexity than the traditional iterative algorithms.
RESUMEN
AIMS: Circular RNAs (circRNAs) acted as key regulators in the development of various human tumors. Our present study aimed to investigate the role and molecular mechanism of circ_0076305 in regulating cisplatin (DDP) resistance of non-small cell lung cancer (NSCLC). MAIN METHODS: Using RT-qPCR, the expressions of circ_0076305 in NSCLC tissues and cells (A549, H1650, A549/DDP, H1650/DDP) were measured. Through loss-of-function and overexpression experiments, the role of circ_0076305 in DDP resistance of NSCLC was verified. Inhibitory rate and IC50 for DDP were detected using MTT method after DDP treatment. Western blotting was performed to evaluate protein levels of P-gp and MRP1. The bindings between circ_0076305 and miR-296-5p, as well as miR-296-5p and STAT3 were validated by bioinformatics, CircRIP, Pearson's correlation analysis and luciferase report vector assays. KEY FINDINGS: Circ_0076305 was upregulated in NSCLC, and more significantly elevated in DDP-resistant NSCLC tissues and cells. Further experiments discovered that circ_0076305 could regulate DDP resistance of NSCLC cells via binding to miR-296-5p. Directly targeted by miR-296-5p, STAT3 hindered the miR-296-5p-induced suppression on DDP resistance. Finally, the expression of circ_0076305 was found to have positive correlation with STAT3, and circ_0076305 was validated to regulate STAT3 via targeting miR-296-5p. SIGNIFICANCE: Our present study illustrated that circ_0076305 regulated STAT3 expression and DDP resistance of NSCLC cells via sponging miR-296-5p. These results suggested knockdown of circ_0076305 might provide an effective approach for NSCLC treatment strategy.
Asunto(s)
Carcinoma de Pulmón de Células no Pequeñas/tratamiento farmacológico , Cisplatino/farmacología , Neoplasias Pulmonares/tratamiento farmacológico , MicroARNs/metabolismo , ARN Circular/metabolismo , Factor de Transcripción STAT3/metabolismo , Células A549 , Anciano , Anciano de 80 o más Años , Carcinoma de Pulmón de Células no Pequeñas/genética , Carcinoma de Pulmón de Células no Pequeñas/metabolismo , Carcinoma de Pulmón de Células no Pequeñas/patología , Línea Celular Tumoral , Resistencia a Antineoplásicos , Femenino , Humanos , Neoplasias Pulmonares/genética , Neoplasias Pulmonares/metabolismo , Neoplasias Pulmonares/patología , Masculino , MicroARNs/genética , Persona de Mediana Edad , ARN Circular/genética , Factor de Transcripción STAT3/genéticaRESUMEN
Cloud computing technology is widely used at present. However, cloud computing servers are far from terminal users, which may lead to high service request delays and low user satisfaction. As a new computing architecture, fog computing is an extension of cloud computing that can effectively solve the aforementioned problems. Resource scheduling is one of the key technologies in fog computing. We propose a resource scheduling method for fog computing in this paper. First, we standardize and normalize the resource attributes. Second, we combine the methods of fuzzy clustering with particle swarm optimization to divide the resources, and the scale of the resource search is reduced. Finally, we propose a new resource scheduling algorithm based on optimized fuzzy clustering. The experimental results show that our method can improve user satisfaction and the efficiency of resource scheduling.
RESUMEN
BACKGROUND: Long non-coding RNA (lncRNA) studies play an important role in the development, invasion, and metastasis of the tumor. The analysis and screening of the differential expression of lncRNAs in cancer and corresponding paracancerous tissues provides new clues for finding new cancer diagnostic indicators and improving the treatment. Predicting lncRNA-protein interactions is very important in the analysis of lncRNAs. This article proposes an Ant-Colony-Clustering-Based Bipartite Network (ACCBN) method and predicts lncRNA-protein interactions. The ACCBN method combines ant colony clustering and bipartite network inference to predict lncRNA-protein interactions. RESULTS: A five-fold cross-validation method was used in the experimental test. The results show that the values of the evaluation indicators of ACCBN on the test set are significantly better after comparing the predictive ability of ACCBN with RWR, ProCF, LPIHN, and LPBNI method. CONCLUSIONS: With the continuous development of biology, besides the research on the cellular process, the research on the interaction function between proteins becomes a new key topic of biology. The studies on protein-protein interactions had important implications for bioinformatics, clinical medicine, and pharmacology. However, there are many kinds of proteins, and their functions of interactions are complicated. Moreover, the experimental methods require time to be confirmed because it is difficult to estimate. Therefore, a viable solution is to predict protein-protein interactions efficiently with computers. The ACCBN method has a good effect on the prediction of protein-protein interactions in terms of sensitivity, precision, accuracy, and F1-score.