Search

COVID-19 is an ongoing pandemic disease. To make more accurate diagnosis on COVID-19 than existing approaches, this paper proposed a novel method combining DenseNet and optimization of transfer learning setting (OTLS) strategy. Preprocessing was used to enhance, crop, and resize the collected chest CT images. Data augmentation method was used to increase the size of training set. A composite learning factor (CLF) was employed which assigned different learning factor to three types of layers: frozen layers, middle layers, and new layers. Meanwhile, the OTLS strategy was proposed. Finally, precomputation method was utilized to reduce RAM storage and accelerate the algorithm. We observed that optimization setting "201-IV" can achieve the best performance by proposed OTLS strategy. The sensitivity, specificity, precision, and accuracy of our proposed method were 96.35 ± 1.07, 96.25 ± 1.16, 96.29 ± 1.11, and 96.30 ± 0.56, respectively. The proposed DenseNet-OTLS method achieved better performances than state-of-the-art approaches in diagnosing COVID-19. [ABSTRACT FROM AUTHOR]

Minimally invasive thermal therapy is a successful alternative treatment to surgery in solid tumors with high complete ablation rates, however, tumor recurrence remains a concern. Central memory CD8+ T cells (TCM) play important roles in protection from chronic infection and cancer. Here we find, by single-cell RNA analysis of human breast cancer samples, that although the memory phenotype of peripheral CD8+ T cells increases slightly after microwave ablation (MWA), the metabolism of peripheral CD8+ T cells remains unfavorable for memory phenotype. In mouse models, glycolysis inhibition by 2-deoxy-D-glucose (2DG) in combination with MWA results in long-term anti-tumor effect via enhancing differentiation of tumor-specific CD44hiCD62L+CD8+ TCM cells. Enhancement of CD8+ TCM cell differentiation determined by Stat-1, is dependent on the tumor-draining lymph nodes (TDLN) but takes place in peripheral blood, with metabolic remodeling of CD8+ T cells lasting the entire course of the the combination therapy. Importantly, in-vitro glycolysis inhibition in peripheral CD8+ T cells of patients with breast or liver tumors having been treated with MWA thrice leads to their differentiation into CD8+ TCM cells. Our work thus offers a potential strategy to avoid tumor recurrence following MWA therapy and lays down the proof-of-principle for future clinical trials. Central memory CD8+ T cells (Tcm) are important for lasting anti-tumour immunity and their differentiation is determined by underpinning transcriptional and metabolic regulation. Authors here show that although microwave ablation (MWA) cancer therapy induces T cells to acquire some of the gene expression features of Tcms, but the full metabolic remodeling required for functionality is only achieved if glycolysis is inhibited simultaneously. [ABSTRACT FROM AUTHOR]

The soil ecosystem plays a significant role in the dynamics of microplastics. Qualitative and quantitative analyses of microplastics in six farmland plots from the cities of Nanjing, Suzhou, and Xuzhou were conducted by using laser infrared imaging system (LDIR). The results showed that the abundance of microplastics in farmland soils in Jiangsu Province ranged from 667–9333 items/kg, and the difference between the abundance of microplastics in soils from open-field cultivation and greenhouse farming was not significant. The vertical distribution characteristics showed that the abundance of microplastic decreased significantly with the increase of soil depth. However, there was no significant difference in particle size between top soil (0–5 cm) and deep soil (10–20 cm). Fragments were the most common microplastic form in soil samples (94.6%), with polyvinyl chloride (PVC) being the main polymer type, followed by polyethylene terephthalate (PET). Compared with previous studies conducted in China, the abundance of microplastics in farmland soils in Jiangsu Province was at moderate level but was notably higher than other places within the Yangtze River Delta region. The conclusions drawn from this paper provided important reference data for future assessment of microplastic pollution in agricultural fields. Furthermore, they establish a fundamental groundwork for understanding the migration patterns of MPs in soil environments. [ABSTRACT FROM AUTHOR]

The application of microalloying technology has been emerging as an effective method to enhance the strength-ductility synergy of high-strength hot-stamped steels, although the impact of microalloying elements on the hot ductility of hot-stamped steels remains an open question. In this work, we systematically studied the hot ductility behavior of V microalloyed hot-stamped steels, covering a broad temperature range from 650 °C to 1300 °C. It is found that a hot brittle range exists at testing temperatures ranging from 650 °C to 900 °C, with the hot ductility gradually decreasing between 900 °C and 650 °C. An increase in microalloying elements is not beneficial to hot ductility in this temperature range. Detailed microstructural analysis and thermodynamic calculation show that the deterioration of hot ductility from 900 °C to 650 °C is closely related to the formation of proeutectoid ferrite and nano-scaled precipitation. [ABSTRACT FROM AUTHOR]

Neoadjuvant chemotherapy (NCT) is the standard treatment for patients with locally advanced breast cancer (LABC). The predictive value of heterogeneous circulating tumor cells (CTCs) in NCT response has not been determined. All patients were staged as LABC, and blood samples were collected at the time of biopsy, and after the first and eighth NCT courses. Patients were divided into High responders (High-R) and Low responders (Low-R) according to Miller–Payne system and changes in Ki-67 levels after NCT treatment. A novel SE-i·FISH strategy was applied to detect CTCs. Heterogeneities were successfully analyzed in patients undergoing NCT. Total CTCs increased continuously and were higher in Low-R group, while in High-R group, CTCs increased slightly during NCT before returning to baseline levels. Triploid and tetraploid chromosome 8 increased in Low-R but not High-R group. The number of small CTCs in Low-R group increased significantly until the last sample, however, remained constant in High-R group. The patients with more CTCs had shorter PFS and OS than those with less CTCs after the eighth course of NCT. Total CTCs following NCT could predict patients' responses. More detailed characterizations of CTC blood profiles may improve predictive capacity and treatments of LABC. [ABSTRACT FROM AUTHOR]

To resolve the problems of low detection accuracy, slow detection speed, and high missed detection rate of traditional warp-knitted fabrics, this study researches and proposes an improved YOLOv5 algorithm for automatic detection of warp-knitted fabric defects, utilizing YOLOv5's fast detection speed and high accuracy. First, a multi-head self-attention mechanism module with an improved activation function is proposed to enhance the model's attention to the defect area of the fabric, improve the detection accuracy of warp-knitted fabric defects and reduce the missed detection rate. Second, a hybrid atrous space pyramid module is added to the backbone extraction network to enhance the receptive field, capture global feature details, and improve the model's recognition and location accuracy of warp-knitted fabric defects. Finally, the transposed convolution is used as an upsampling layer to improve the feature fusion network. The feature extraction layer can better combine fine-grained details with highly abstract information, enhance the accuracy of feature fusion, and then improve the detection accuracy of the model. Experimental results show that using the self-built warp-knitted fabric dataset, the mean average precision of the improved YOLOv5 is 91.3%, the precision rate is 89.7%, and the recall rate is 79.9%, which is 7.9%, 15.6% and 4.1% higher than the original YOLOv5 algorithm, respectively. The improved YOLOv5 defect detection algorithm has a higher accuracy, faster speed, and better robustness, which is helpful for the development and application of a warp-knitted fabric automatic inspection system. [ABSTRACT FROM AUTHOR]

Deep Auto-Encoders (DAE) have been widely used in dimensionality reduction and feature extraction of brain functional networks. However, the features in aggregation matrices of functional networks obtained by DAE dimensionality reduction might lose part of time-varying information, so that DAE cannot learn the distribution of original features well. To solve these problems, we extracted and analyzed brain functional statuses for early Mild Cognitive Impairment (eMCI) based on Variational Auto-Encoder (VAE). The high dimensions of features in aggregation matrices of dynamic functional networks were reduced by VAE to obtain the corresponding hidden variable matrices. Gaussian Mixture Model (GMM) was used to cluster the features in the hidden variable matrices to form several Common Functional Networks (CFNs) representing different functional statuses. We analyzed the similarities and differences of functional statuses between eMCI subjects and normal subjects in different sub-segments, as well as the switching of functional statuses in the entire time series. The experimental results show that there are similarities between the most frequent functional statuses of the two types of subjects and differences between the least frequent functional statuses. The proposed method can more significantly reveal the similarity and difference of functional statuses between eMCI subjects and normal subjects than the comparable methods, and the switching rule of functional statuses can help better understand the dynamic characteristics of brain functional networks for eMCI patients. [ABSTRACT FROM AUTHOR]

Cerebral microbleed (CMB) is a serious public health concern. It is associated with dementia, which can be detected with brain magnetic resonance image (MRI). CMBs often appear as tiny round dots on MRIs, and they can be spotted anywhere over brain. Therefore, manual inspection is tedious and lengthy, and the results are often short in reproducible. In this paper, a novel automatic CMB diagnosis method was proposed based on deep learning and optimization algorithms, which used the brain MRI as the input and output the diagnosis results as CMB and non-CMB. Firstly, sliding window processing was employed to generate the dataset from brain MRIs. Then, a pre-trained VGG was employed to obtain the image features from the dataset. Finally, an ELM was trained by Gaussian-map bat algorithm (GBA) for identification. Results showed that the proposed method VGG-ELM-GBA provided better generalization performance than several state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Diffusion tensor imaging (DTI) is a non-invasive magnetic resonance imaging technique and a special type of magnetic resonance imaging, which has been widely used to study the diffusion process in the brain. The signal-to-noise ratio of DTI data is relatively low, the shape and direction of the noisy tensor data are destroyed. This limits the development of DTI in clinical applications. In order to remove the Rician noise and preserve the diffusion tensor geometry of DTI, we propose a DTI denoising algorithm based on Riemann nonlocal similarity. Firstly, DTI tensor is mapped to the Riemannian manifold to preserve the structural properties of the tensor. The Riemann similarity measure is used to search for non-local similar blocks to form similar patch groups. Then the Gaussian mixture model is used to learn the prior distribution of patch groups. Finally, the noisy patch group is denoised by Bayesian inference and the denoised patch group is reconstructed to obtain the final denoised image. The denoising experiments of real and simulated DTI data are carried out to verify the feasibility and effectiveness of the proposed algorithm. The experimental results show that our algorithm not only effectively removes the Rician noise in the DTI image, but also preserves the nonlinear structure of the DTI image. Comparing to the existing denoising algorithms, our algorithm has better improvement of the principal diffusion direction, lower absolute error of fractional anisotropy and higher peak signal-to-noise ratio. [ABSTRACT FROM AUTHOR]

Module can not only affect the integration of network functions, but also contribute to understand the characteristics of local connections in the network. However, the connection between nodes in the network changes with the passage of time, and the module structure changes accordingly. To overcome this drawback, we propose a method of applying weighted clustering ensemble to partition multilayer brain functional networks into modules. Firstly, k-means clustering is adopted to carry out base clustering for a certain layer of functional network for several times, and each clustering is corresponding to a subordinate matrix and a similarity matrix. Then clustering validity index is used to assess each partitioning and the assessed values are taken as the weights of similarity matrix. Finally, the weighted similarity matrix is partitioned by means of fuzzy C-means clustering, and the results are evaluated by modularity function to obtain the optimal partitioned modules. Experimental results show that the effect of module partitioning resulting from weighted clustering ensemble is better than that of other comparable methods. The proposed framework could be promising to analyze the differences between corresponding modules of patients with Alzheimer's disease and normal people, so as to better understanding some dynamical pathological changes in brain connectome of patients. [ABSTRACT FROM AUTHOR]

Copyright of Rare Metals is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background: Cancer-related fatigue is a widely prevalent global public health concern with serious consequences. Increasing evidence suggests the effectiveness of exercise intervention in treating cancer-related fatigue, but there is a lack of a summary of relevant literature on the same to help reach a clear consensus. Objective: To summarize evidence regarding the efficacy of exercise interventions to reduce cancer fatigue, as determined in systematic reviews (SRs) and/or meta-analyses (MAs). Method: From inception to September 2022, PubMed (1948–2022), Embase (1974–2022), Cochrane Library (1993–2022), CINAHL (1937–2022), Web of Science (1997–2022), China Knowledge Resource Integrated Database (1999–2022), Wanfang Database (1993–2022), and Chinese Biomedical Database (1994–2022) were searched for inclusion to the study. Two reviewers independently extracted the data from the included articles. AMSTAR II was to evaluate the methodological quality of the reviews. Results: A total of 46 systematic reviews were assessed for data on exercise intervention in reducing cancer-related fatigue among cancer patients. In addition, some studies have reported adverse events during the exercise intervention period. The quality of the included systematic review was found to be low or critically low. Conclusions: The present systematic review of systematic reviews supports exercise intervention for reducing cancer-related fatigue. Further higher-quality studies are warranted to improve the level of evidence for exercise interventions for application in the treatment of cancer-related fatigue. [ABSTRACT FROM AUTHOR]

Adequately choosing the physicochemical characteristics of adsorbent is crucial in improving its adsorption performance. This work investigated the effect of particle size of magnetic resins on adsorption behaviors of tetracycline (TC) and their antifouling performance. Smaller particle size resin Q150 (10–30 μm) shows notably faster TC adsorption kinetics when compared resins with hundreds of microns (Q100 and Q1). Simulated by Weber-Morris equation, the film diffusion time of Q150 was only 20 min, 2–25 times faster than that of other resins. At this adsorption time, Q150 can reach more than 80% of the maximum adsorption, and the ring-like fluorescence images indicate that the molecules are accumulated on the external surface. Q150 also shows better reusability and antifouling performance over Q100 and Q1. After 20 adsorption–desorption cycles, the adsorption capacity of Q150 at 20 min only decreases 9.7%. The presence of tannic acid also only slightly decreases the adsorption capacity. The faster adsorption kinetics and the superior antifouling performance of Q150 make it a promising adsorbent in future use. [ABSTRACT FROM AUTHOR]

Rice and maize are two of the world's most productive crops. Their production is often severely threatened by devastating pathogens. As a key negative immune regulator, CONSTITUTIVE EXPRESSER OF PATHOGENESIS-RELATED GENES 5 (CPR5) has been characterized well in dicots but not in monocots. Here we took advantage of Arabidopsis cpr5 mutants to explore the functional conservation of its counterparts from the monocots rice and maize. A single copy of the CPR5 gene is present in the Arabidopsis genome, while two copies of it are identified in the genomes of both rice and maize. Like the Arabidopsis CPR5 protein, all those from rice and maize have an RNA recognition motif in the N-terminal region and five transmembrane domains in the C-terminal region. The 35S promoter-driven coding DNA sequences of four CPR5 genes from rice and maize were introduced into an Arabidopsis cpr5 mutant. The results showed that cpr5-induced immunity and programmed cell death were fully suppessed by these genes, indicating that the function of CPR5 proteins are conserved between dicots and monocots. Intriguingly, we showed that mutations of the CPR5 gene on chromosome I but not the other on chromosome II confer resistance to rice yellow mottle virus in African rice, suggesting that these two copies of rice CPR5 proteins have a distinctive role in immunity. These findings will provide a basis for application of CPR5 to improve disease resistance in monocot crops. [ABSTRACT FROM AUTHOR]

Extreme learning machine (ELM) is a training algorithm for single hidden layer feedforward neural network (SLFN), which converges much faster than traditional methods and yields promising performance. In this paper, we hope to present a comprehensive review on ELM. Firstly, we will focus on the theoretical analysis including universal approximation theory and generalization. Then, the various improvements are listed, which help ELM works better in terms of stability, efficiency, and accuracy. Because of its outstanding performance, ELM has been successfully applied in many real-time learning tasks for classification, clustering, and regression. Besides, we report the applications of ELM in medical imaging: MRI, CT, and mammogram. The controversies of ELM were also discussed in this paper. We aim to report these advances and find some future perspectives. [ABSTRACT FROM AUTHOR]

CNN is inspired from Primary Visual (V1) neurons. It is a typical deep learning technique and can help teach machine how to see and identify objects. In the most recent decade, deep learning develops rapidly and has been well used in various fields of expertise such as computer vision and natural language processing. As the representative algorithm of deep learning, Convolution Neural Network (CNN) has been regarded as a breakthrough of historic significance in image processing and visual recognition tasks since the astonishing results achieved on ImageNet Large Scale Visual Recognition Competition (ILSVRC) Unlike methods based on handcrafted features, CNN models can build high-level features from low-level ones in a data-driven fashion and have displayed great potential in medical image analysis among the aspects of segmentation of histological images identification, lesion detection, tissue classification, etc. This paper provides a review on CNN from the perspectives of its basic mechanism introduction, structure, typical architecture and main application in medical image analysis through analyzing over 100 references from Google Scholar, PubMed, Web of Science and various sources published from 1958 to 2020. [ABSTRACT FROM AUTHOR]

The concentrations of six priority phthalate esters (PAEs) in 700 soil samples and 110 sediment samples from an area in China containing plastic solid waste (PSW) recycling sites were determined. The total concentrations of the six PAEs in soil and sediment were not detected – 274 and not detected – 597 mg kg−1, respectively, and the mean concentrations in soil and sediment were 14.4 and 31.7 mg kg−1, respectively. The dominant PAEs were di(2-ethylhexyl) phthalate and di-n-butyl phthalate. PAEs were detected in soil collected from the surface to 0.5 m below ground level around the PSW recycling sites, and the concentrations were markedly higher in these areas than at other polluted sites. PSW recycling is an important source of PAEs to soil and sediment. The di(2-ethylhexyl)phthalate concentrations in soil were higher than the relevant concentrations that pose environmental risks for sensitive land uses and non-sensitive land uses (42 and 121 mg kg−1, respectively), indicating emissions of PAEs from PSW recycling sites may pose environmental risks. The results indicate that PAE pollution at PSW sites needs to be better controlled and managed. [ABSTRACT FROM AUTHOR]

White blood cells (WBCs) are a portion of the immune system which fights against germs. Leukemia is the most common blood cancer which may lead to death. It occurs due to the production of a large number of immature WBCs in the bone marrow that destroy healthy cells. To overcome the severity of this disease, it is necessary to diagnose the shapes of immature cells at an early stage that ultimately reduces the modality rate of the patients. Recently different types of segmentation and classification methods are presented based upon deep-learning (DL) models but still have some limitations. This research aims to propose a modified DL approach for the accurate segmentation of leukocytes and their classification. The proposed technique includes two core steps: preprocessing-based classification and segmentation. In preprocessing, synthetic images are generated using a generative adversarial network (GAN) and normalized by color transformation. The optimal deep features are extracted from each blood smear image using pretrained deep models i.e., DarkNet-53 and ShuffleNet. More informative features are selected by principal component analysis (PCA) and fused serially for classification. The morphological operations based on color thresholding with the deep semantic method are utilized for leukemia segmentation of classified cells. The classification accuracy achieved with ALL-IDB and LISC dataset is 100% and 99.70% for the classification of leukocytes i.e., blast, no blast, basophils, neutrophils, eosinophils, lymphocytes, and monocytes, respectively. Whereas semantic segmentation achieved 99.10% and 98.60% for average and global accuracy, respectively. The proposed method achieved outstanding outcomes as compared to the latest existing research works. [ABSTRACT FROM AUTHOR]

Medical imaging is an essential medical diagnosis system subsequently integrated with artificial intelligence for assistance in clinical diagnosis. The actual medical images acquired during the image capturing procedures generate poor quality images as a result of numerous physical restrictions of the imaging equipment and time constraints. Recently, medical image super-resolution (SR) has emerged as an indispensable research subject in the community of image processing to address such limitations. SR is a classical computer vision operation that attempts to restore a visually sharp high-resolution images from the degraded low-resolution images. In this study, an effective medical super-resolution approach based on weighted least squares optimisation via multiscale convolutional neural networks (CNNs) has been proposed for lesion localisation. The weighted least squares optimisation strategy that particularly is well-suited for progressively coarsening the original images and simultaneously extract multiscale information has been executed. Subsequently, a SR model by training CNNs based on wavelet analysis has been designed by carrying out wavelet decomposition of optimized images for multiscale representations. Then multiple CNNs have been trained separately to approximate the wavelet multiscale representations. The trained multiple convolutional neural networks characterize medical images in many directions and multiscale frequency bands, and thus facilitate image restoration subject to increased number of variations depicted in different dimensions and orientations. Finally, the trained CNNs regress wavelet multiscale representations from a LR medical images, followed by wavelet synthesis that forms a reconstructed HR medical image. The experimental performance indicates that the proposed model SR restoration approach achieve superior SR efficiency over existing comparative methods [ABSTRACT FROM AUTHOR]

Magnetic reconnection is a fundamental plasma process by which magnetic field lines on two sides of the current sheet flow inward to yield an X-line topology. It is responsible for producing energetic electrons in explosive phenomena in space, astrophysical, and laboratorial plasmas. The X-line region is supposed to be the important place for generating energetic electrons. However, how these energetic electrons are generated in such a limited region is still poorly understood. Here, using Magnetospheric multiscale mission data acquired in Earth's magnetotail, we present direct evidence of super-thermal electrons up to 300 keV inside an X-line region, and the electrons display a power-law spectrum with an index of about 8.0. Concurrently, three-dimensional network of dynamic filamentary currents in electron scale is observed and leads to electromagnetic turbulence therein. The observations indicate that the electrons are effectively accelerated while the X-line region evolves into turbulence with a complex filamentary current network. Magnetotail reconnection plays an important role in explosive energy conversion. Here, the authors show direct evidence of super-thermal electrons up to 300 keV within X-line region in Earth's magnetotail, indicating effective electron acceleration due to turbulence. [ABSTRACT FROM AUTHOR]

Cerebral microbleed (CMB) is related to cerebral vascular diseases. In this paper, we propose the use of deep convolutional neural network to implement CMB automatic diagnosis based on brain susceptibility-weighted images (SWIs). First of all, a sliding neighborhood method was employed to get 13,031 samples for training and testing. Then, an 18-layer CMB-Net was designed to classify the samples as CMB or non-CMB. The CMB-Net was trained by RMSprop based on the five-fold cross- validation. The total running time of the five-fold cross-validation was merely 184.79 s, and the average testing accuracy reached 98.39%, which was better than several recently published methods. The results suggested that our CMB-Net was accurate in detecting CMB. [ABSTRACT FROM AUTHOR]

Background: Circulating tumor cells (CTCs) have been shown to be associated with the response to neoadjuvant chemotherapy (NCT) and the prognosis of locally advanced breast cancer (LABC) patients. Our study aimed to investigate whether the change of CTC status during NCT could serve as a supplement to the Response Evaluation Criteria in Solid Tumors (RECIST) in the treatment and evaluation of LABC patients. Methods: 6 ml of blood samples were collected before NCT, after the first cycle of NCT and after the completion of NCT, respectively. According to the change of CTC number during NCT, the patients were divided into "CTC low-response (low-R)" group and "CTC high-response (high-R)" group. Survival data of each group of patients were obtained through long-term follow-up. Results: A total of 35 patients diagnosed with LABC were enrolled. The median follow-up for distant metastasis was 27 months (range 7–36 months). There was no significant difference in distant metastasis-free survival (DMFS) between PR/CR group and PD/SD group (P = 0.0914), while CTC low-R group had a worse DMFS than CTC high-R group (P = 0.0199). In PR/CR subgroup, patients with CTC low-R showed a lower DMFS compared with those with CTC high-R (P = 0.0159). However, in PD/SD subgroup, there was no significant difference in DMFS between CTC low-R and CTC high-R group (P = 0.7521). In terms of assessing response to NCT, CTC change or RECIST classification alone had an AUC of 0.533 (95% CI 0.277–0.790) and 0.700 (95% CI 0.611–0.789), respectively. When combining the two, the AUC slightly increased to 0.713 (95% CI 0.532–0.895). Conclusion: The change of CTC number during NCT has a potential to serve as a supplement to RECIST in the assessment of NCT efficacy and the prognosis of LABC patients. [ABSTRACT FROM AUTHOR]

COVID-19 is a contagious infection that has severe effects on the global economy and our daily life. Accurate diagnosis of COVID-19 is of importance for consultants, patients, and radiologists. In this study, we use the deep learning network AlexNet as the backbone, and enhance it with the following two aspects: 1) adding batch normalization to help accelerate the training, reducing the internal covariance shift; 2) replacing the fully connected layer in AlexNet with three classifiers: SNN, ELM, and RVFL. Therefore, we have three novel models from the deep COVID network (DC-Net) framework, which are named DC-Net-S, DC-Net-E, and DC-Net-R, respectively. After comparison, we find the proposed DC-Net-R achieves an average accuracy of 90.91% on a private dataset (available upon email request) comprising of 296 images while the specificity reaches 96.13%, and has the best performance among all three proposed classifiers. In addition, we show that our DC-Net-R also performs much better than other existing algorithms in the literature. [ABSTRACT FROM AUTHOR]

This document is a retraction note for an article titled "Alcoholism identification via convolutional neural network based on parametric ReLU, dropout, and batch normalization" published in the journal Neural Computing & Applications. The article has been retracted by the Editor-in-Chief and the publisher due to concerns regarding compromised editorial handling and peer review process, inappropriate or irrelevant references, and not being in scope of the journal or guest-edited issue. The authors of the article disagree with this retraction. Springer Nature, the publisher, remains neutral in regards to jurisdictional claims and institutional affiliations. [Extracted from the article]

Secondary pulmonary tuberculosis (SPT) is one of the top ten causes of death from a single infectious agent. To recognize SPT more accurately, this paper proposes a novel artificial intelligence model, which uses Pseudo Zernike moment (PZM) as the feature extractor and deep stacked sparse autoencoder (DSSAE) as the classifier. In addition, 18-way data augmentation is employed to avoid overfitting. This model is abbreviated as PZM-DSSAE. The ten runs of 10-fold cross-validation show this model achieves a sensitivity of 93.33% ± 1.47%, a specificity of 93.13% ± 0.95%, a precision of 93.15% ± 0.89%, an accuracy of 93.23% ± 0.81%, and an F1 score of 93.23% ± 0.83%. The area-under-curve reaches 0.9739. This PZM-DSSAE is superior to 5 state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Computer-aided diagnosis system is becoming a more and more important tool in clinical treatment, which can provide a verification of the doctors' decisions. In this paper, we proposed a novel abnormal brain detection method for magnetic resonance image. Firstly, a pre-trained AlexNet was modified with batch normalization layers and trained on our brain images. Then, the last several layers were replaced with an extreme learning machine. A searching method was proposed to find the best number of layers to be replaced. Finally, the extreme learning machine was optimized by chaotic bat algorithm to obtain better classification performance. Experiment results based on 5 × hold-out validation revealed that our method achieved state-of-the-art performance. [ABSTRACT FROM AUTHOR]

Ductal carcinoma in situ (DCIS) is a pre-cancerous lesion in the ducts of the breast, and early diagnosis is crucial for optimal therapeutic intervention. Thermography imaging is a non-invasive imaging tool that can be utilized for detection of DCIS and although it has high accuracy (~ 88%), it is sensitivity can still be improved. Hence, we aimed to develop an automated artificial intelligence-based system for improved detection of DCIS in thermographs. This study proposed a novel artificial intelligence based system based on convolutional neural network (CNN) termed CNN-BDER on a multisource dataset containing 240 DCIS images and 240 healthy breast images. Based on CNN, batch normalization, dropout, exponential linear unit and rank-based weighted pooling were integrated, along with L-way data augmentation. Ten runs of tenfold cross validation were chosen to report the unbiased performances. Our proposed method achieved a sensitivity of 94.08 ± 1.22%, a specificity of 93.58 ± 1.49 and an accuracy of 93.83 ± 0.96. The proposed method gives superior performance than eight state-of-the-art approaches and manual diagnosis. The trained model could serve as a visual question answering system and improve diagnostic accuracy. [ABSTRACT FROM AUTHOR]

Air samples were collected around industrial parks in Jiangsu, China, to allow the concentrations, profiles, and risk assessment of polybrominated dibenzo-p-dioxins and dibenzofurans (PBDD/Fs), polychlorinated naphthalenes (PCNs), and metals to be investigated. The concentrations of ΣPBDD/Fs and ΣPCNs were 1324.26–2080.98 fg/m3 (11.35–42.57 fg I-TEQ/m3) and 10,404.9–29,322.9 fg/m3 (1.32–7.19 fg I-TEQ/ m3), respectively. The highest concentration of ΣPBDD/Fs and ΣPCNs were observed at site C. PBDD/Fs were mainly dominated by PBDFs. The main contributor to the ΣPBDD/Fs in all samples was 1,2,3,4,6,7,8-HpBDF, which accounted for 25.75%–39.4%. For PCNs, the predominating homologues were tetra-, tri- and penta-CNs, which contributed 30.7%–43.3%, 24.7%–31.0%, and 10.6%–21.6%, respectively. As for metals, the pollution of As, Mn, Cr, and Ni in most samples exceeded National Ambient Air Quality Standards of China. Assessing the risk of inhalation exposure showed that there were potential carcinogenic risks to local residents. [ABSTRACT FROM AUTHOR]

The end-Permian mass extinction (EPME) was the most severe extinction event in the past 540 million years, and the Siberian Traps large igneous province (STLIP) is widely hypothesized to have been the primary trigger for the environmental catastrophe. The killing mechanisms depend critically on the nature of volatiles ejected during STLIP eruptions, initiating about 300 kyr before the extinction event, because the atmosphere is the primary interface between magmatism and extinction. Here we report Ni isotopes for Permian-Triassic sedimentary rocks from Arctic Canada. The δ60Ni data range from −1.09‰ to 0.35‰, and exhibit the lightest δ60Ni compositions ever reported for sedimentary rocks. Our results provide strong evidence for global dispersion and loading of Ni-rich aerosol particles into the Panthalassic Ocean. Our data demonstrate that environmental degradation had begun well before the extinction event and provide a link between global dispersion of Ni-rich aerosols, ocean chemistry changes, and the EPME. The end-Permian mass extinction was the most severe extinction event in the past 540 million years, and the Siberian Traps large igneous province is widely hypothesized to have been the primary trigger for the environmental catastrophe. In this study, Ni isotopes provide the link between Siberian Traps magmatism and early environmental degradation, ultimately leading to the end-Permian extinction. [ABSTRACT FROM AUTHOR]

As two different modalities of medical images, Magnetic Resonance (MR) and Computer Tomography (CT), provide mutually-complementary information to doctors in clinical applications. However, to obtain both images sometimes is cost-consuming and unavailable, particularly for special populations. For example, patients with metal implants are not suitable for MR scanning. Also, it is probably infeasible to acquire multi-contrast MR images during once clinical scanning. In this context, to synthesize needed MR images for patients whose CT images are available becomes valuable. To this end, we present a novel generative network, called CAE-ACGAN, which incorporates the advantages of Variational Auto-Encoder (VAE) and Generative Adversarial Network (GAN) with an auxiliary discriminative classifier network. We apply this network to synthesize multi-contrast MR images from single CT and conduct experiments on brain datasets. Our main contributions can be summarized as follows: 1)We alleviate the problems of images blurriness and mode collapse by integrating the advantages of VAE and GAN; 2) We solve the complicated cross-domain, multi-contrast MR synthesis task using the proposed network; 3) The technique of random-extraction-patches is used to lower the limit of insufficient training data, enabling to obtain promising results even with limited available data; 4) By comparing with other typical networks, we are able to yield nearer-real, higher-quality synthetic MR images, demonstrating the effectiveness and stability of our proposed network. [ABSTRACT FROM AUTHOR]

Compared with common deep learning methods (e.g., convolutional neural networks), transfer learning is characterized by simplicity, efficiency and its low training cost, breaking the curse of small datasets. Medical image analysis plays an indispensable role in both scientific research and clinical diagnosis. Common medical image acquisition methods include Computer Tomography (CT), Magnetic Resonance Imaging (MRI), Ultrasound (US), X-Ray, etc. Although these medical imaging methods can be applied for non-invasive qualitative and quantitative analysis of patients—compared with image datasets in other computer vision fields such like faces—medical images, especially its labeling, is still scarce and insufficient. Therefore, more and more researchers adopted transfer learning for medical image processing. In this study, after reviewing one hundred representative papers from IEEE, Elsevier, Google Scholar, Web of Science and various sources published from 2000 to 2020, a comprehensive review is presented, including (i) structure of CNN, (ii) background knowledge of transfer learning, (iii) different types of strategies performing transfer learning, (iv) application of transfer learning in various sub-fields of medical image analysis, and (v) discussion on the future prospect of transfer learning in the field of medical image analysis. Through this review paper, beginners could receive an overall and systematic knowledge of transfer learning application in medical image analysis. And policymaker of related realm will benefit from the summary of the trend of transfer learning in medical imaging field and may be encouraged to make policy positive to the future development of transfer learning in the field of medical image analysis. [ABSTRACT FROM AUTHOR]

We propose a new method for hearing loss classification from magnetic resonance image (MRI), which can automatically detect tissue-specific features in a given MRI. Sensorineural hearing loss (SHNL) is highly prevalent in our society. Early diagnosis and intervention have a profound impact on patient outcomes. A solution to provide early diagnosis is the use of automated diagnostic systems. In this study, we propose a novel Deep-HLNet framework, based on few-shot learning, for the automated classification of SNHL. This research involves magnetic resonance (MRI) images from 60 participants of three balanced categories: left-sided SNHL, right-sided SNHL, and healthy controls. A convolutional neural network was employed for feature extraction from individual categories, while a neural network and a comparison classifier strategy constituted a tri-classifier for SNHL classification. In terms of experiment results and practicability of the algorithm, the classification performance was significantly better than the standard deep learning methods or other conventional methods, with an overall accuracy of 96.62%. [ABSTRACT FROM AUTHOR]

Earth's habitability is closely tied to its late-stage accretion, during which impactors delivered the majority of life-essential volatiles. However, the nature of these final building blocks remains poorly constrained. Nickel (Ni) can be a useful tracer in characterizing this accretion as most Ni in the bulk silicate Earth (BSE) comes from the late-stage impactors. Here, we apply Ni stable isotope analysis to a large number of meteorites and terrestrial rocks, and find that the BSE has a lighter Ni isotopic composition compared to chondrites. Using first-principles calculations based on density functional theory, we show that core-mantle differentiation cannot produce the observed light Ni isotopic composition of the BSE. Rather, the sub-chondritic Ni isotopic signature was established during Earth's late-stage accretion, probably through the Moon-forming giant impact. We propose that a highly reduced sulfide-rich, Mercury-like body, whose mantle is characterized by light Ni isotopic composition, collided with and merged into the proto-Earth during the Moon-forming giant impact, producing the sub-chondritic Ni isotopic signature of the BSE, while delivering sulfur and probably other volatiles to the Earth. Based on Nickel isotope analysis of meteorites and terrestrial rocks, the authors suggest that the Bulk Silicate Earth has a sub-chondritic Nickel isotope composition. This signature is thought to result from the impact and accretion of a Mercury-like impactor which originated from the innermost Solar System. [ABSTRACT FROM AUTHOR]

Chinese Sign Language (CSL) offers the main means of communication for the hearing impaired in China. Sign Language Recognition (SLR) can shorten the distance between the hearing-impaired and healthy people and help them integrate into the society. Therefore, SLR has become the focus of sign language application research. Over the years, the continuous development of new technologies provides a source and motivation for SLR. This paper aims to cover the most recent approaches in Chinese Sign Language Recognition (CSLR). With a thorough review of superior methods from 2000 to 2019 in CSLR researches, various techniques and algorithms such as scale-invariant feature transform, histogram of oriented gradients, wavelet entropy, Hu moment invariant, Fourier descriptor, gray-level co-occurrence matrix, dynamic time warping, principal component analysis, autoencoder, hidden Markov model (HMM), support vector machine (SVM), random forest, skin color modeling method, k-NN, artificial neural network, convolutional neural network (CNN), and transfer learning are discussed in detail, which are based on several major stages, that is, data acquisition, preprocessing, feature extraction, and classification. CSLR was summarized from some aspect as follows: methods of classification and feature extraction, accuracy/performance evaluation, and sample size/datasets. The advantages and limitations of different CSLR approaches were compared. It was found that data acquisition is mainly through Kinect and camera, and the feature extraction focuses on hand's shape and spatiotemporal factors, but ignoring facial expressions. HMM and SVM are used most in the classification. CNN is becoming more and more popular, and a deep neural network-based recognition approach will be the future trend. However, due to the complexity of the contemporary Chinese language, CSLR generally has a lower accuracy than other SLR. It is necessary to establish an appropriate dataset to conduct comparable experiments. The issue of decreasing accuracy as the dataset increases needs to resolve. Overall, our study is hoped to give a comprehensive presentation for those people who are interested in CSLR and SLR and to further contribute to the future research. [ABSTRACT FROM AUTHOR]

In this paper, whether the coal fines can be induced by shear failure during drainage process has been discussed in detail. By coupling with the percolation theory, the elasticity mechanics were used to construe the extra stresses in the formation surrounding with the hydraulic fracture. The safe window of the bottom hole pressure was also calculated from the failure envelope. The research shows that the formation pressure on the fracture surface of the coal seam is negatively related with the bottom hole pressure, and the induced stress is positively related with the bottom hole pressure during the drainage process of fractured CBM wells. The pore pressure around the fracture changed due to pore-elastic effects, which also caused a significant change of the in situ stresses. In order to avoid the breakout of the coal seam around hydraulic fracture during drainage process, the model of the reasonable bottom hole pressure is also built. [ABSTRACT FROM AUTHOR]

The concept of the Solar Ring mission was gradually formed from L5/L4 mission concept, and the proposal of its pre-phase study was funded by the National Natural Science Foundation of China in November 2018 and then by the Strategic Priority Program of Chinese Academy of Sciences in space sciences in May 2019. Solar Ring mission will be the first attempt to routinely monitor and study the Sun and inner heliosphere from a full 360-degree perspective in the ecliptic plane. The current preliminary design of the Solar Ring mission is to deploy six spacecraft, grouped in three pairs, on a sub-AU orbit around the Sun. The two spacecraft in each group are separated by about 30° and every two groups by about 120°. This configuration with necessary science payloads will allow us to establish three unprecedented capabilities: (1) determine the photospheric vector magnetic field with unambiguity, (2) provide 360-degree maps of the Sun and the inner heliosphere routinely, and (3) resolve the solar wind structures at multiple scales and multiple longitudes. With these capabilities, the Solar Ring mission aims to address the origin of solar cycle, the origin of solar eruptions, the origin of solar wind structures and the origin of severe space weather events. The successful accomplishment of the mission will advance our understanding of the star and the space environment that hold our life and enhance our capability of expanding the next new territory of human. [ABSTRACT FROM AUTHOR]

Magnetic reconnection is a fundamental plasma process, by which magnetic energy is explosively released in the current sheet to energize charged particles and to create bi-directional Alfvénic plasma jets. Numerical simulations predicted that evolution of the reconnecting current sheet is dominated by formation and interaction of magnetic flux ropes, which finally leads to turbulence. Accordingly, most volume of the reconnecting current sheet is occupied by the ropes, and energy dissipation occurs via multiple relevant mechanisms, e.g., the parallel electric field, the rope coalescence and the rope contraction. As an essential element of the reconnecting current sheet, however, how these ropes evolve has been elusive. Here, we present direct evidence of secondary reconnection in the filamentary currents within the ropes. The observations indicate that secondary reconnection can make a significant contribution to energy conversion in the kinetic scale during turbulent reconnection. Magnetic reconnection is a fundamental plasma process of magnetic energy conversion to kinetic energy. Here, the authors show direct evidence of secondary reconnection in the filamentary currents within the flux ropes indicating a significant contribution to energy conversion in the kinetic scale during turbulent reconnection. [ABSTRACT FROM AUTHOR]

Fingerspelling recognition of Chinese sign language rendered an opportunity to smooth the communication barriers of hearing-impaired people and health people, which occupies an important position in sign language recognition. This study proposed an eight-layer convolutional neural network, combined with three advanced techniques: batch normalization, dropout, and stochastic pooling. The output of the stochastic pooling was obtained via sampling from a multinomial distribution formed from the activations of each pooling region. In addition, we used data augmentation method to enhance the training set. In total 10 runs were implemented with the hold-out randomly set for each run. Our method achieved the highest accuracy of 90.91% and overall accuracy of 89.32 ± 1.07%, which was superior to three state-of-the-art approaches compared. [ABSTRACT FROM AUTHOR]

(Aim) For detecting the hearing loss (HL) more accurately and efficiently, the new computer-aid diagnosis (CAD) based on a nature-inspired algorithm (NIAs) is proposed in this study. (Method) First, the discrete wavelet transform (DWT) is used for extracting texture features from the brain images, and then the principle component analysis (PCA) is employed to decrease the dimension of features. Second, the Multi-Layer Perceptron (MLP) is used as a classifier. Traditional gradient-based descent algorithms are vulnerable to get struck at local minima; thus, the NIAs are introduced. The differential evolution algorithm (DE), particle swarm optimization (PSO), artificial bee colony algorithm (ABC), and improved ABC (IABC) are employed to train MLP. Because the ordinary ABC is good at exploration but gives a poor performance at exploitation, therefore a new model of ABC, called IABC is proposed. The K-fold validation method is utilized to measure the performance of the CAD. (Result) To verify the performance of our method, The CAD based on IABC is compared with state-of-the-art-approaches. (Conclusion) The experiment results show that the overall accuracy of our method has the highest overall accuracy among five approaches. Therefore, the proposed CAD is effective method for detecting hearing loss. [ABSTRACT FROM AUTHOR]

Traditional researches assume that brain functional networks are static during the entire scanning process of functional magnetic resonance image (fMRI) in the resting state. However, it is not difficult to ignore the dynamic interaction patterns of brain regions that essentially change across time. In this study, we take the internal weight information of the brain functional network as a calculation condition of module dividing for brain functional networks. The concept of betweenness efficiency is firstly proposed to improve Girvan-Newman (GN) algorithm for a better module dividing result, and the maximum modularity is used as a criterion to classify the brain functional network modules of normal subjects. The effect of the improved method is verified by controlling subjects, parameters, environment and other conditions. Then, the improved method was utilized to separate the modules in brain functional networks of normal subjects, and the template was used to divide the functional network of Alzheimer's disease (AD) patients and mild cognitive impairment (MCI) sufferers. The shortest path length of each module is calculated, and the experimental results are compared with the original GN and weighted GN algorithm improved by Newman. The experimental results demonstrate that, the maximum modularity of the improved method is higher while the dividing effect is better under the same conditions. Meanwhile, the conclusion is consistent with the existing research results when the proposed method is applicable to the analysis of the shortest path length. These results illustrate the viewpoint that the proposed method of module dividing is feasible in the analysis of modular structure of brain functional network. [ABSTRACT FROM AUTHOR]

In this paper, we apply an improved deep convolutional neural network (CNN) in fruit category classification, which is a hotspot in computer vision field. We created an 8-layer deep convolutional neural network, and utilized parametric rectified linear unit to take the place of plain rectified linear unit, and placed dropout layer before each fully-connected layer. Data augmentation was used to help avoid overfitting. Our 8-layer deep convolutional neural network secured an overall accuracy of 95.67%. This proposed 8-layer method performs better than five state-of-the-art methods using traditional machine learning methods and one state-of-the-art CNN method. [ABSTRACT FROM AUTHOR]

Cerebral microbleeds are important biomarkers of many cerebrovascular diseases and cognitive dysfunctions. Their distribution patterns can indicate some underlying aetiologies. Hitherto, few researches tried to detect cerebral microbleeds accurately and automatically. Some improvements have been achieved via traditional machine learning methods. In this paper, we proposed a method based on convolutional neural network (CNN) for further improving the performance. Firstly, sliding neighborhood processing method was applied to generate the input and target datasets based on 10 3D brain images of cerebral autosomal-dominant arteriopathy with subcortical infarcts and Leukoencephalopathy scanned by susceptibility-weighted imaging (SWI). Then, CNN was used to classify the cerebral microbleeds. To exert the full-power of convolutional neural network, almost all hyperparameters of CNN structure that could affect the performance were tested, such as the number of layers, type of activation function, pooling method, and filter size. A fully-optimized convolutional neural network structure for cerebral microbleeds classification was obtained. It performed better than four existed state-of-the-art approaches with a sensitivity of 99.74%, a specificity of 96.89% and an accuracy of 98.32%. [ABSTRACT FROM AUTHOR]

Traditional sensorineural hearing loss identification use the framework of feature extraction and classification. Nevertheless, this framework needs manual feature engineering. In this study, we proposed an improved convolutional neural network model to identify hearing loss. Our nine-layer deep convolutional neural network contains six conv layers and three fully-connected layers. We used batch normalization to reduce the impact caused by Internal Covariate shift and dropout techniques to prevent over-fitting to increase the performance in terms of accuracy. Data augmentation was used to enlarge the size of training set. The average results of 10 runs on test set show our method secured sensitivities of left-sided hearing loss, right-sided hearing loss, and healthy controls are 96.33 ± 2.46%, 96.67 ± 2.22%, and 96.67 ± 2.72%, respectively. The overall accuracy of all three classes was 96.56 ± 0.63%. Deep learning can effectively build the identification model. The performance of our proposed nine-layer convolutional neural network model yields better performance than five state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Alcoholism changes the structure of brain. Several somatic marker hypothesis network-related regions are known to be damaged in chronic alcoholism. Neuroimaging approach can help us better understanding the impairment discovered in alcohol-dependent subjects. In this research, we recruited subjects from participating hospitals. In total, 188 abstinent long-term chronic alcoholic participants (95 men and 93 women) and 191 non-alcoholic control participants (95 men and 96 women) were enrolled in our experiment via computerized diagnostic interview schedule version IV and medical history interview employed to determine whether the applicants can be enrolled or excluded. The Siemens Verio Tim 3.0 T MR scanner (Siemens Medical Solutions, Erlangen, Germany) was employed to scan the subjects. Then, we proposed a 10-layer convolutional neural network for the diagnosis based on imaging, including three advanced techniques: parametric rectified linear unit (PReLU); batch normalization; and dropout. The structure of network is fine-tuned. The results show that our method secured a sensitivity of 97.73 ± 1.04%, a specificity of 97.69 ± 0.87%, and an accuracy of 97.71 ± 0.68%. We observed the PReLU gives better performance than ordinary ReLU, clipped ReLU, and leaky ReLU. The batch normalization and dropout gained enhanced performance as batch normalization overcame the internal covariate shift and dropout got over the overfitting. The results of our proposed 10-layer CNN model show its performance better than seven state-of-the-art approaches. [ABSTRACT FROM AUTHOR]