Abstract This study aimed to analyze the ability of extracting automatically generated features using deep structured algorithms in lung nodule CT image diagnosis, and compare its performance with ...traditional computer aided diagnosis (CADx) systems using hand-crafted features. All of the 1018 cases were acquired from Lung Image Database Consortium (LIDC) public lung cancer database. The nodules were segmented according to four radiologists’ markings, and 134668 samples were generated by rotating every slice of nodule images. Three multichannel ROI based deep structured algorithms were designed and implemented in this study: convolutional neural network (CNN), deep belief network (DBN), and stacked denoising autoencoder (SDAE). For the comparison purpose, we also implemented a CADx system using hand-crafted features including density features, texture features and morphological features. The performance of every scheme was evaluated by using a 10-fold cross-validation method and an assessment index of the area under the receiver operating characteristic curve (AUC). The observed highest area under the curve (AUC) was 0.899±0.018 achieved by CNN, which was significantly higher than traditional CADx with the AUC= 0.848±0.026. The results from DBN was also slightly higher than CADx, while SDAE was slightly lower. By visualizing the automatic generated features, we found some meaningful detectors like curvy stroke detectors from deep structured schemes. The study results showed the deep structured algorithms with automatically generated features can achieve desirable performance in lung nodule diagnosis. With well-tuned parameters and large enough dataset, the deep learning algorithms can have better performance than current popular CADx. We believe the deep learning algorithms with similar data preprocessing procedure can be used in other medical image analysis areas as well.
Magnetotelluric measurements reveal the presence of high conductivity anomalies (up to ∼1 S/m) in both the forearc and backarc regions of subduction zones as well as the continental middle–lower ...crust. Such anomalies are commonly interpreted as a consequence of aqueous fluid released from the dehydration of hydrous minerals. Amphibole is an important constituent of the continental mid-crust and a major hydrous phase in subduction zones, such that its dehydration at high temperature has been suggested to provide a significant source of aqueous fluid. We performed electrical conductivity measurements of a natural Fe-bearing amphibole at 623–1173 K and 0.5–2.0 GPa using a multi-anvil apparatus and an impedance spectroscopy. Our results show that pressure has a very weak effect on conductivity compared with temperature. An abrupt variation of the impedance semicircular arc followed by a remarkable increase of electrical conductivity is observed at temperature of 843±20 K. However, the enhancement in conductivity is not attributed to conductive aqueous fluid but rather to amphibole oxidation–dehydrogenation, as confirmed by infrared spectroscopy and optical microscopy observations. A slight decrease in activation enthalpy from ∼0.80 eV to ∼0.70 eV suggests that the conduction mechanism does not change before and after dehydrogenation, and small polaron conduction (electron holes hopping between Fe2+ and Fe3+) is considered to dominate the conductivity of amphibole over the entire temperature range. Although amphibole dehydrogenation at high temperature cannot serve as a principal source of aqueous fluid, the enhanced electrical conductivity of amphibole after dehydrogenation is sufficient to account for the high conductivity anomalies observed in slab–mantle wedge interfaces and the continental lowermost mid-crust, particularly in local regions with high heat flow.
•Variations of impedance and conductivity at 843 K mark dehydrogenation occurrence.•Small polaron conduction dominates electrical conductivity of amphibole.•Dehydrogenated amphibole can explain conductivity anomaly in subduction zone.•Dehydrogenated amphibole interprets conductivity anomaly in continental mid-crust.
The enhanced electrical conductivity of amphibole after dehydrogenation can explain the high conductivity anomalies in the slab–mantle wedge interface at depth of >70 km as well as lowermost mid-crust of stable continental.
Deep learning techniques have been extensively used in computerized pulmonary nodule analysis in recent years. Many reported studies still utilized hybrid methods for diagnosis, in which ...convolutional neural networks (CNNs) are used only as one part of the pipeline, and the whole system still needs either traditional image processing modules or human intervention to obtain final results. In this paper, we introduced a fast and fully-automated end-to-end system that can efficiently segment precise lung nodule contours from raw thoracic CT scans. Our proposed system has four major modules: candidate nodule detection with Faster regional-CNN (R-CNN), candidate merging, false positive (FP) reduction with CNN, and nodule segmentation with customized fully convolutional neural network (FCN). The entire system has no human interaction or database specific design. The average runtime is about 16 s per scan on a standard workstation. The nodule detection accuracy is 91.4% and 94.6% with an average of 1 and 4 false positives (FPs) per scan. The average dice coefficient of nodule segmentation compared to the groundtruth is 0.793.
Abstract
Background
Silicosis is one of the most common occupational pulmonary fibrosis caused by respirable silica-based particle exposure, with no ideal drugs at present. Metformin, a commonly used ...biguanide antidiabetic agent, could activate AMP-activated protein kinase (AMPK) to exert its pharmacological action. Therefore, we sought to investigate the role of metformin in silica-induced lung fibrosis.
Methods
The anti-fibrotic role of metformin was assessed in 50 mg/kg silica-induced lung fibrosis model. Silicon dioxide (SiO
2
)-stimulated lung epithelial cells/macrophages and transforming growth factor-beta 1 (TGF-β1)-induced differentiated lung fibroblasts were used for in vitro models.
Results
At the concentration of 300 mg/kg in the mouse model, metformin significantly reduced lung inflammation and fibrosis in SiO
2
-instilled mice at the early and late fibrotic stages. Besides, metformin (range 2–10 mM) reversed SiO
2
-induced cell toxicity, oxidative stress, and epithelial-mesenchymal transition process in epithelial cells (A549 and HBE), inhibited inflammation response in macrophages (THP-1), and alleviated TGF-β1-stimulated fibroblast activation in lung fibroblasts (MRC-5) via an AMPK-dependent pathway.
Conclusions
In this study, we identified that metformin might be a potential drug for silicosis treatment.
Display omitted
•The effects of temperature, water and Na contents on electrical conductivity of plagioclase are quantified.•Crustal rocks containing structural water hardly cause the high ...conductivity anomalies in the mid-lower crust.•At least 1 vol% of fluid with seawater salinity is required to explain high conductivity anomalies in mid-lower crust.
Aqueous fluids are regarded as the crucial origin of high conductivity and low velocity anomalies observed by magnetotelluric and seismic surveys in the middle to lower continental crust. Integration of field geophysical methods with experimental conductivity data is essential to accurately constrain the fluid fraction and salinity. Studies of fluid content and salinity in the crust from the experimental or theoretically calculated conductivity of fluid–minerals systems and saline-bearing fluid have demonstrated that crustal constituent minerals play a prominent role in the conductivity of fluid–mineral systems. We performed systematic measurements on electrical conductivity of dry and hydrous plagioclase solid solutions at high temperature and pressure. A positive dependence of electrical conductivity on Na contents is observed in dry samples, and discontinuous variation in conductivity is associated with the degree of Al/Si order–disorder. For hydrous samples, trace amounts of water considerably enhance the electrical conductivity. The dominant charge carriers in dry and hydrous plagioclase are Na ions and hydrogen, respectively, both of which migrate via the Frenkel mechanism. If there are no free fluids present in minerals, the crustal major constituent minerals containing structure water cannot account for the high-conductivity anomalies in the middle to lower crust based on our model of the rock conductivity–depth profiles in the continental crust. The calculated bulk conductivities of fluid–rock systems further suggest that at least 1 vol% of fluid with seawater salinity (3.5 wt%) is necessary in the middle to lower crust to interpret the high conductivity anomalies.
Abstract
Recently, the method of estimating magnetic field through monochromatic images by deep learning has been proposed, demonstrating good morphological similarity but somewhat poor magnetic ...polarity consistency relative to real observation. In this paper, we propose to estimate magnetic field from H
α
images by using a conditional generative adversarial network (cGAN) as the basic framework. The H
α
images from the Global Oscillation Network Group are used as the inputs and the line-of-sight magnetograms of the Helioseismic Magnetic Imager (HMI) are used as the targets. First, we train a cGAN model (Model A) with shuffling training data. However, the estimated magnetic polarities are not very consistent with real observations. Second, to improve the accuracy of estimated magnetic polarities, we train a cGAN model (Model B) with the chronological H
α
and HMI images, which can implicitly exploit the magnetic polarity constraint of time-series observation to generate more accurate magnetic polarities. We compare the generated magnetograms with the target HMI magnetograms to evaluate the two models. It can be observed that Model B has better magnetic polarity consistency than Model A. To quantitatively measure this consistency, we propose a new metric called pixel-to-pixel polarity accuracy (PPA). With respect to PPA, Model B is superior to Model A. This work gives us an insight that the time-series constraint can be implicitly exploited through organizing training data chronologically, and this conclusion also can be applied to other similar tasks related to time-series data.
Positron emission tomography (PET)-computed tomography (CT) images have been widely used in clinical practice for radiotherapy treatment planning of the radiotherapy. Many existing segmentation ...approaches only work for a single imaging modality, which suffer from the low spatial resolution in PET or low contrast in CT. In this work, we propose a novel method for the co-segmentation of the tumor in both PET and CT images, which makes use of advantages from each modality: the functionality information from PET and the anatomical structure information from CT. The approach formulates the segmentation problem as a minimization problem of a Markov random field model, which encodes the information from both modalities. The optimization is solved using a graph-cut based method. Two sub-graphs are constructed for the segmentation of the PET and the CT images, respectively. To achieve consistent results in two modalities, an adaptive context cost is enforced by adding context arcs between the two sub-graphs. An optimal solution can be obtained by solving a single maximum flow problem, which leads to simultaneous segmentation of the tumor volumes in both modalities. The proposed algorithm was validated in robust delineation of lung tumors on 23 PET-CT datasets and two head-and-neck cancer subjects. Both qualitative and quantitative results show significant improvement compared to the graph cut methods solely using PET or CT.
Dysregulation of non‐coding RNAs (ncRNAs) has been proved to play pivotal roles in epithelial‐mesenchymal transition (EMT) and fibrosis. We have previously demonstrated the crucial function of long ...non‐coding RNA (lncRNA) ATB in silica‐induced pulmonary fibrosis‐related EMT progression. However, the underlying molecular mechanism has not been fully elucidated. Here, we verified miR‐29b‐2‐5p and miR‐34c‐3p as two vital downstream targets of lncRNA‐ATB. As opposed to lncRNA‐ATB, a significant reduction of both miR‐29b‐2‐5p and miR‐34c‐3p was observed in lung epithelial cells treated with TGF‐β1 and a murine silicosis model. Overexpression miR‐29b‐2‐5p or miR‐34c‐3p inhibited EMT process and abrogated the pro‐fibrotic effects of lncRNA‐ATB in vitro. Further, the ectopic expression of miR‐29b‐2‐5p and miR‐34c‐3p with chemotherapy attenuated silica‐induced pulmonary fibrosis in vivo. Mechanistically, TGF‐β1‐induced lncRNA‐ATB accelerated EMT as a sponge of miR‐29b‐2‐5p and miR‐34c‐3p and shared miRNA response elements with MEKK2 and NOTCH2, thus relieving these two molecules from miRNA‐mediated translational repression. Interestingly, the co‐transfection of miR‐29b‐2‐5p and miR‐34c‐3p showed a synergistic suppression effect on EMT in vitro. Furthermore, the co‐expression of these two miRNAs by using adeno‐associated virus (AAV) better alleviated silica‐induced fibrogenesis than single miRNA. Approaches aiming at lncRNA‐ATB and its downstream effectors may represent new effective therapeutic strategies in pulmonary fibrosis.
N
-methyladenosine (m
A) is the most common and abundant internal modification of RNA. Its critical functions in multiple physiological and pathological processes have been reported. However, the ...role of m
A in silica-induced pulmonary fibrosis has not been fully elucidated. AlkB homolog 5 (ALKBH5), a well-known m
A demethylase, is upregulated in the silica-induced mouse pulmonary fibrosis model. Here, we sought to investigate the function of ALKBH5 in pulmonary fibrosis triggered by silica inhalation.
We performed studies with fibroblast cell lines and silica-induced mouse pulmonary fibrosis models. The expression of ALKBH5, miR-320a-3p, and forkhead box protein M1 (FOXM1) was determined by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. RNA immunoprecipitation (RIP) assays and m
A RNA immunoprecipitation assays (MeRIP), western bolt, immunofluorescence assays, and 5-ethynyl-2'-deoxyuridine (EdU) fluorescence staining were performed to explore the roles of ALKBH5, miR-320a-3p, and FOXM1 in fibroblast activation.
ALKBH5 expression was increased in silica-inhaled mouse lung tissues and transforming growth factor (TGF)-β1-stimulated fibroblasts. Moreover, ALKBH5 knockdown exerted antifibrotic effects in vitro. Simultaneously, downregulation of ALKBH5 elevated miR-320a-3p but decreased pri-miR-320a-3p. Mechanically, ALKBH5 demethylated pri-miR-320a-3p, thus blocking the microprocessor protein DGCR8 from interacting with pri-miR-320a-3p and leading to mature process blockage of pri-miR-320a-3p. We further demonstrated that miR-320a-3p could regulate fibrosis by targeting FOXM1 messenger RNA (mRNA) 3'-untranslated region (UTR). Notably, our study also verified that ALKBH5 could also directly regulate FOXM1 in an m
A-dependent manner.
Our findings suggest that ALKBH5 promotes silica-induced lung fibrosis via the miR-320a-3p/FOXM1 axis or targeting FOXM1 directly. Approaches aimed at ALKBH5 may be efficacious in treating lung fibrosis.
The continually growing human population creates a concomitantly increasing demand for nutritious crops with high yields. Advances in high throughput sequencing technologies have revealed the genetic ...architecture of major crops. This includes extensive information enabling comprehensive genetic markers for breeding selection, new gene discoveries, and novel gene regulatory strategies for crop editing. RNA structure is an important type of genetic feature, essential for post-transcriptional regulation of gene expression. Here, we summarize recent advances in genome-wide RNA structure studies in crops and review the associated RNA structure-mediated regulation of gene expression. We also discuss the functional importance of those single nucleotide variations that induce large RNA structure disparities. Lastly, we discuss the potential role of RNA structure in crop molecular breeding.