Search

Liang, Rui-Ru, Zhang, Sheng-Nan, Liu, Jian-Kui (2023): Bambusicola autumnalis sp. nov., a bambusicolous ascomycete from Sichuan province, China. Phytotaxa 601 (3): 203-211, DOI: 10.11646/phytotaxa.601.3.1, URL: http://dx.doi.org/10.11646/phytotaxa.601.3.1

Bambusicola autumnalis R.R. Liang, S.N. Zhang and Jian K. Liu, sp. nov. MycoBank: 847551; FIGURE 2. Etymology: —The epithet “ autumnalis ” refers to the season “autumn” when the fungus was collected. Holotype:—HKAS 126508 Saprobic on dead bamboo branches. Sexual morph: Ascostromata solitary to gregarious, rarely scattered, immersed to erumpent, pseudostromatic, visible as bumped areas with cracks and a central black minute papilla, in vertical section conical to subglobose, ostiolate, periphysate, individual locules 190–325 μm high, 215–295 μm diam (x = 280 × 250 µm, n = 10). Peridium 19–43 μm, composed of several layers of thick-walled, brown cells of textura angularis. Hamathecium 1–1.9 μm wide, trabecular pseudoparaphyses, anastomosing, hyaline, remotely septate. Asci 52–102 × 8–17 µm (x = 71 × 12 µm, n = 20), 8-spored, bitunicate, fissitunicate, long cylindric-clavate, shortly pedicellate, apically rounded with a minute ocular chamber. Ascospores 22–30 × 4.5–7 µm (x = 27 × 5 µm, n = 30), overlapping bi-seriate or multi-seriate, fusiform, 1-septate, constricted at the septum, guttulate, smooth-walled, surrounded by a thin, inconspicuous mucilaginous sheath. Asexual morph: Undetermined. Culture characteristics:—Colonies on PDA reaching 28–32 mm after 4 weeks incubated at 25 °C in dark, circular, dry, mycelium velvety, milky white to pale yellow, reverse yellow to light brown. Material examined:— CHINA, Sichuan province, Chengdu city, Chengdu Botanical Garden, 30°76.48’ N, 104°13.03’ E, 516 m elevation, on dead branches of bamboo in a terrestrial environment, 21 Nov. 2022, R.R. Liang, (HKAS 126508, holotype), ex-holotype living culture CGMCC 3.24280; ibid., HUEST 23.0001, isotype, ex-isotype living culture UESTCC 23.0001. Notes:—Multi-gene phylogenetic analysis showed that our isolates belong to Bambusicola and are closely related to B. guttulata (FIGURE 1). However, it is not able to compare their morphology as the latter species only represented by a coelomycetous asexual morph. Nevertheless, Bambusicola autumnalis differs from B. guttulata in their nucleotide sequences, viz. SSU (6/950), ITS (44/450), LSU (17/800), RPB2 (73/993) and TEF1-α (49/959), respectively. Morphologically, Bambusicola autumnalis resembles B. loculata in having stromatic ascomata, 8-spored, cylindric asci, and narrowly fusiform, 1-septate ascospores surrounded by an inconspicuous sheath (Dai et al. 2015). However, they have different dimensions of asci (52–102 × 8–17 μm vs. 80–105 × 8–13 μm) and ascospores (22–30 × 4.5–7 µm vs. 22–26.5 × 5–6), and the bumped ascostromata with cracks of B. autumnalis also differs from B. loculata. Moreover, the two species are phylogenetically distinct (FIGURE 1)., Published as part of Liang, Rui-Ru, Zhang, Sheng-Nan & Liu, Jian-Kui, 2023, Bambusicola autumnalis sp. nov., a bambusicolous ascomycete from Sichuan province, China, pp. 203-211 in Phytotaxa 601 (3) on page 208, DOI: 10.11646/phytotaxa.601.3.1, http://zenodo.org/record/8141736, {"references":["Dai, D. Q., Bahkali, A. H., Li, W. J., Bhat, D. J., Zhao, R. L. & Hyde, K. D. (2015) Bambusicola loculata sp. nov. (Bambusicolaceae) from bamboo. Phytotaxa 213 (2): 122 - 130. http: // doi. org / 10.11646 / phytotaxa. 213.2.5"]}

We consider the Cauchy problem for the one-dimensional periodic cubic nonlinear fractional Schrödinger equation (FNLS) with initial data distributed via Gibbs measure. We construct global strong solutions with the flow property to FNLS on the support of the Gibbs measure in the full dispersive range, thus resolving a question proposed by Sun-Tzvetkov (2021). As a byproduct, we prove the invariance of the Gibbs measure and almost sure global well-posedness for FNLS.

The multi-source data-driven faulted segment identification methodology has spurred research interest recently. Based on analyzing the post-fault physical features of the three-core cable earthed strap current, this study develops an approach to identify the faulty cable segment(s) of neutral low-resistance grounding distribution grids with distributed generators. To eliminate the potential uncertainty of recorded data, introducing Mixed Integer Linear Programming to construct the fault diagnosis model. The model employs traditional fault indications, including alarms from remote fault indicators and metering signals of zero-sequence current, to promote the validity of identifications. Under the causal logic constraint of events, employing the expected fault scenario with the minimum discrepancy to available data to determine the true faulty cable segment(s). Additionally, the model decision variables can be used as the reference for anomalous data determination. The effectiveness and robustness of the proposed methodology are validated by applying it to a simulation model based on a real-life three-core cable utility feeder.

The optimization of most combinatorial problems is NP-hard, which can be solved by heuristic algorithms to obtain approximately optimal solutions, especially for large-scale problems. Many heuristic algorithms can apply the 2-opt local search to find better solutions. In complete 2-opt local search, every valid neighboring solutions of swapping mechanism will be compared, which is very time consuming especially when a sequential algorithm is adopted. Nowadays, graphic processing units (GPUs) have evolved into powerful and flexible parallel computation platforms that have been widely used to accelerate the solving of NP-hard problems. Therefore, in this study, we focus on how to use GPUs to solve the single-row facility layout problem (SRFLP) with the 2-opt-based simulated annealing (SA) heuristic algorithm. As far as we know, this study is the first to solve SRFLP by using a GPU. After analyzing the fitness function and the move gains calculation between parent and child solutions, we propose a prefix-sum formula table to eliminate a large amount of replicated computation. To calculate move gains for the 2-opt local search during each iteration, many GPU threads are created and they construct and lookup the table in parallel. According to experimental results, if the prefix-sum formula table is not adopted, the GPU version outperforms the sequential CPU counterpart with the best speedup of 123. However, if the table is used in the GPU version, the best speedup can reach up to 3208. Because the proposed parallelization approach with the prefix-sum formula table is based on the features of the 2-opt local search and the SRFLP fitness function, it can be applied to any 2-opt-based heuristic algorithm for solving SRFLP even though the SA algorithm is used in this study.

Extreme learning machine (ELM) has fast learning speed and perfect performance, at the same time, ELM provides a unified learning framework with a widespread type of feature mappings which can be applied in multiclass classification applications directly. These advantages make ELM become one of the best classification algorithms, and ELM has attracted great attention in supervised learning and semi-supervised learning. However, noise and outliers of data are usually existed in the real world, which will affect the performance of ELM. To improve the robustness and classification performance of ELM, we propose the Kernel Risk-Sensitive Mean p-power Loss Based Hyper-graph Regularized Robust Extreme Learning Machine (KRP-HRELM) method. On the one side, as a nonlinear similarity measure defined in the reproducing kernel space, the kernel risk-sensitive mean p-power loss (KRP) can effectively weaken the negative effects caused by noise and outliers. Therefore, the KRP is introduced into ELM to enhance its robustness. Then, the application of hyper-graph can help the ELM to explore higher-order geometric structures among more sampling points, thereby obtaining more comprehensive data information. In addition, to obtain a more sparsity network model, the L2,1-norm is used to constrain the output weight. On the other side, improving the practical application ability of KRP-HRELM is also the focus of our research, so KRP-HRELM is extended to semi-supervised learning, which is called the semi-supervised KRP-HRELM (SS-KRP-HRELM). Notably, the results of the robustness experiment have proved that our method has extraordinary robustness. At the same time, by using four evaluation measures such as accuracy, recall, precision, and F1-measure, to evaluate the classification results, we can find that our method has obtained better classification performance than other advanced methods.

Bioactive ingredients from natural products have always been an important resource for the discovery of drugs for Alzheimer's disease (AD). Senile plaques, which are formed with amyloid-beta (Aβ) peptides and excess metal ions, are found in AD brains and have been suggested to play an important role in AD pathogenesis. Here, we attempted to design an effective and smart screening method based on cheminformatics approaches to find new ingredients against AD from Vaccinium myrtillus (bilberry) and verified the bioactivity of expected ingredients through experiments. This method integrated advanced artificial intelligence models and target prediction methods to realize the stepwise analysis and filtering of all ingredients. Finally, we obtained the expected new compound malvidin-3-O-galactoside (Ma-3-gal-Cl). The in vitro experiments showed that Ma-3-gal-Cl could reduce the OH· generation and intracellular ROS from the Aβ/Cu2+/AA mixture and maintain the mitochondrial membrane potential of SH-SY5Y cells. Molecular docking and Western blot results indicated that Ma-3-gal-Cl could reduce the amount of activated caspase-3 via binding with unactivated caspase-3 and reduce the expression of phosphorylated p38 via binding with mitogen-activated protein kinase kinases-6 (MKK6). Moreover, Ma-3-gal-Cl could inhibit the Aβ aggregation via binding with Aβ monomer and fibers. Thus, Ma-3-gal-Cl showed significant effects on protecting SH-SY5Y cells from Aβ/Cu2+/AA induced damage via antioxidation effect and inhibition effect to the Aβ aggregation.

Frontal impacts are the most frequent crash types and may generate extreme body in white (BIW) deformations. Vehicles are subjected to impact evaluation rating protocols such as 100 % Front Rigid Barrier (FRB) impact, 40 % Offset Deformable Barrier (ODB) impact and 25 % overlap (small overlap) impact. This paper proposes a collaborative optimization process using optimal Latin hypercube design (Opt LHD) and response surface methodology (RSM), to improve the vehicle crashworthiness in the frontal impacts, by considering displacement oriented structure (DOS). Upper and lower engine bay structure were considered herein to increase the lateral vehicle displacement during small overlap impact and decrease the impact force that transferred to the passenger compartment in the frontal impacts. The optimal results indicated that the acceleration of the B-pillar was reduced 0.4g in the FRB impact, the intrusion of the firewall was reduced by 55.13 % in the ODB impact, and the intrusion of passenger compartment during small overlap impact was decreased by an average of 53.88 %, with a maximum percentage of 71.41 % around left toepan, leading to an IIHS rating to acceptable from poor. The proposed crashworthiness design approach is effective in vehicle structure optimization for better frontal impacts performance.

Recently, extreme learning machine (ELM) has attracted a lot of attention due to its high performance and extreme speed. However, how to improve the robustness of ELM has always been a problem. Considering the problems of noise and outliers in the experimental data, in this paper, we introduce the kernel risk-sensitive mean p-power error (KRP) into ELM and propose a robust ELM method named kernel risk-sensitive mean p-power error based robust extreme learning machine (KRPELM), on the basis of the high efficiency and robustness of KRP. In KRPELM, KRP function instead of square loss is integrated into ELM as the loss function, which can improve the robustness of ELM to noise and outliers. We also propose an efficient iterative adjustment strategy to optimize KRPELM. Nine benchmark datasets are utilized to verify the classification performance of the proposed KRPELM. In addition, we apply the proposed method to the classification of cancer samples. The experimental results on five cancer gene expression datasets show that KRPELM can identify different cancer types more accurately.

Objective. The diagnostic value of CT window width technique in primary omentum infarction was evaluated by this study. Methods. The abdominal CT data of 32 patients with clinically diagnosed abdominal omentum infarction were retrospectively selected and analyzed. The fixed window position was 50 HU, and the window width was 135 HU, 250 HU (abdomen), 350 HU (mediastinum), and 500 HU, respectively. The detection rate of lesions was analyzed and compared. Results. Window widths of 135 HU, 250 HU (abdomen), 350 HU (mediastinum), and 500 HU have a detection rate of 12.5% (4 cases), 62.5% (20 cases), 100% (32 cases), 100% (32 cases) for abdominal omental lesions, respectively. However, 500 HU showed worse abdominal bowel and parenchymal organs than 350 HU. Conclusion. According to the comprehensive image quality, the ideal window width for diagnosis of primary omentum infarction is 350HU (mediastinal) window width.

Introduction. Against the background of marketing and globalization of higher education, many organizations and enterprises not only face stiff competition and quality problems in the market. Despite all the current challenges, they need to implement brand management tools. The urgent task for universities was how to apply the brand's marketing strategy to the brand management of universities, implement the optimal distribution of tangible and intangible resources, create a good brand image, expand popularity and reputation and gain competitive advantage.The hypothesis of scientific research is that for the formation of an effective brand it is necessary to use the tools of strategic marketing.The aim is to study the theoretical and methodological foundations of brand formation in educational institutions using modern competitive marketing strategies.The research methodology is a comparative analysis to reflect trends in the development of the brand of education; systematization and classification – for the classification of structural elements of the object of study; system approach – for the formation of theoretical foundations of regulation of brand management processes.Results. Based on the law on higher education and reliance on the mechanism and legislation of the market, this article examines brand management and argues that university brands should be optimized, coordinated and integrated to take full advantage of internal and external resources. It is necessary to strengthen the concept of brand management, explain the position of the university brand, create a strong university brand, strengthen the dominant university brand, create a crisis management system for university brands, increase the basic competitiveness of universities and gain greater benefits.Conclusions. University brand management should be based on the marketing strategy of enterprise brands, make full use of their own resources, strengthen the concept of brand management, determine the position of the university brand and create a strong university brand. universities, the university brand is undoubtedly a dazzling edge of the sword.

Background: In recent years, immunotherapy has changed the therapeutic landscape of hepatocellular carcinoma (HCC). Since the efficacy of immunotherapy is closely related to the tumor microenvironment (TME), in this study, we constructed a prognostic model based on TME to predict the prognosis and immunotherapy effect of HCC patients.Methods: Transcriptome and follow-up data of 374 HCC patients were acquired from the TCGA Cancer Genome Atlas (TCGA) database. The immune/stromal/estimate scores (TME scores) and tumor purity were calculated using the ESTIMATE algorithm and the module most associated with TME scores were screened by the weighted gene co-expression network analysis (WGCNA). A TME score-related prognostic model was constructed and patients were divided into a high-risk group and a low-risk group. Kaplan-Meier survival curves and receiver operator characteristic curve (ROC) were used to evaluate the performance of the TME risk prognostic model and validated with the external database International Cancer Genome Consortium (ICGC) cohort. Combined with clinicopathologic factors, a prognostic nomogram was established. The nomogram’s ability to predict prognosis was assessed by ROC, calibration curve, and the decision curve analysis (DCA). Gene Set Enrichment Analyses (GSEA) were conducted to explore the underlying biological functions and pathways of this risk signature. Moreover, the possible correlation of risk signature with TME immune cell infiltration, immune checkpoint inhibitor (ICI) treatment response, single-nucleotide polymorphisms (SNPs), and drug sensitivity were assessed. Finally, real-time PCR was used to verify the gene expression levels in normal liver cells and cancer cells.Results: KM survival analysis results indicated that high immune/stromal/estimate score groups were closely associated with a better prognosis, while the tumor purity showed a reverse trend (p < 0.01). WGCNA demonstrated that the yellow module was significantly correlated with the TME score. The 5-genes TME risk signature was built to predict the prognosis of patients with HCC including DAB2, IL18RAP, RAMP3, FCER1G, and LHFPL2. Patients with a low-risk score have higher levels of tumor-infiltrating immune cells and higher expression of immune checkpoints, which may be more sensitive to immunotherapy.Conclusion: It provided a theoretical basis for predicting the prognosis and personalized treatment of patients with HCC.

Direct myocardial and vascular injuries due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection-driven inflammation is the leading cause of acute cardiac injury associated with coronavirus disease 2019 (COVID-19). However, in-depth knowledge of the injury characteristics of the heart affected by inflammation is lacking. In this study, using a quantitative spatial proteomics strategy that combines comparative anatomy, laser-capture microdissection, and histological examination, we establish a region-resolved proteome map of the myocardia and microvessels with obvious inflammatory cells from hearts of patients with COVID-19. A series of molecular dysfunctions of myocardia and microvessels is observed in different cardiac regions. The myocardia and microvessels of the left atrial are the most susceptible to virus infection and inflammatory storm, suggesting more attention should be paid to the lesion and treatment of these two parts. These results can guide in improving clinical treatments for cardiovascular diseases associated with COVID-19.

Additional file 6: Figure S3. KOG classification of the Unigenes in O. tormota. Different KOG function classes are shown in different letter and colors, the number of genes are shown along the vertical axis.

Additional file 3: Supplementary Figure 1. (A) The best-fit OS-related lncRNAs were chosen by Lasso regression analysis. (B) The Lasso regression was performed with the optimal value of λ. (C-D) Distribution of risk scores, survival status. Supplementary Figure 2. (A-D) PCA among all genes, immune genes, immune LncRNA, and risk immune LncRNA.

Supplemental Material, sj-pdf-1-ehp-10.1177_0163278720979651 for Validation of the Chinese Version of the Body Image Concern Inventory by Kui Wang, Xin-yang Yu, Chao-ran Yu, Ya-fei Liu, Min-yi Chu, Rui-ting Zhang, Rui Liang, Jue Chen, Heather L. Littleton, David H. K. Shum and Raymond C. K. Chan in Evaluation & the Health Professions

Additional file 7: Figure S4. Heat map and hierarchical clustering of differential expressed genes in four groups. The red represents up-regulated unigenes, and down-regulated unigenes are represented in green.

Additional file 7: Figure S4. Heat map and hierarchical clustering of differential expressed genes in four groups. The red represents up-regulated unigenes, and down-regulated unigenes are represented in green.

Additional file 4: Supplementary Table 3. All the primer sequences.

Additional file 3: Figure S1. Length distribution statistics of the assembly All-unigene. Horizontal axis representsthe All-unigene length interval; Vertical axis indicates the number of All-unigene within the length interval range.

Additional file 6: Figure S3. KOG classification of the Unigenes in O. tormota. Different KOG function classes are shown in different letter and colors, the number of genes are shown along the vertical axis.

Additional file 5: Figure S2. Species distribution based on the best hit of NR blast result. Different species are represented in different colors, and the size of the pie region represents the proportion of unigenes which were annotated to different species.

Additional file 3: Figure S1. Length distribution statistics of the assembly All-unigene. Horizontal axis representsthe All-unigene length interval; Vertical axis indicates the number of All-unigene within the length interval range.

Additional file 5: Figure S2. Species distribution based on the best hit of NR blast result. Different species are represented in different colors, and the size of the pie region represents the proportion of unigenes which were annotated to different species.

Supplemental material, sj-docx-1-tag-10.1177_17562848221102307 for The hospitalization burden of inflammatory bowel disease in China: a nationwide study from 2013 to 2018 by Yi-Ming He, Ren Mao, Gang Yuan, Rui-Ming Liang, Jian-Yan Long, Xiao-Qi Ye, Marietta Iacucci, Subrata Ghosh, Shomron Ben-Horin, Gilaad G. Kaplan, Yao He, Joseph J.Y. Sung, Sui Peng, Hai-Bo Wang and Min-Hu Chen in Therapeutic Advances in Gastroenterology

Supplemental material, sj-docx-1-tag-10.1177_17562848221102307 for The hospitalization burden of inflammatory bowel disease in China: a nationwide study from 2013 to 2018 by Yi-Ming He, Ren Mao, Gang Yuan, Rui-Ming Liang, Jian-Yan Long, Xiao-Qi Ye, Marietta Iacucci, Subrata Ghosh, Shomron Ben-Horin, Gilaad G. Kaplan, Yao He, Joseph J.Y. Sung, Sui Peng, Hai-Bo Wang and Min-Hu Chen in Therapeutic Advances in Gastroenterology

Additional file 2: Supplementary Table 2. 76 prognostic associated lncRNAs.

Additional file 1: Table 1. Clinical characteristics of hepatocellular carcinoma in train and validation cohort.

Additional file 4: Supplementary Table 3. All the primer sequences.

As a fast-growing commercial wood species, poplar often produces tension wood because of the leaning stems of logs, which has a great impact on the production and utilization. The gelatinous layer (G-layer) was the unique component of tension wood, almost entirely composed of cellulose, which was the main reason for the special performance of the poplar tension wood. This study used the transmission electron microscopy, Raman spectroscopy and nanoindentation to examine the ultrastructure and mechanical properties of every wall layer of the poplar cell wall in order to figure out the ultrastructure and properties of G-layer. In particular, the nanoindentation mapping was combined with manual screening to obtain the mechanical performance data of each layer. The fibers in the tension wood were gelatinous wood fibers with obvious G-layers, and the thickness was 2.53 μm, which was much larger than that of other wall layers. Compared with the wood fiber in the opposite wood, the S1 layer was thicker and the S2 layer was thinner, and the entire cell wall thickness was increased in tension wood. G-layer in tension wood had a very high cellulose content, and the Raman strength of lignin was weak. In general, the micromechanical properties of poplar tension wood were better, which was consistent with its macromechanical properties. The characteristics of G-layer, including high cellulose content, small microfibril angle and high crystallinity, made G-layer own better elastic modulus (14.08 GPa) and hardness (0.45 GPa) than other wall layers in the tension wood. Moreover, G-layer had a high proportion in the tension wood cell wall, so the average longitudinal elastic modulus and hardness of wood fiber cells in the tension wood were also higher than those in the opposite wood.

In this paper, the quantitative evaluation of the explosion effect based on the fuzzy comprehensive evaluation method is proposed to describe the qualitative evaluation results. The selected state characteristic parameters are expressed by two kinds of membership functions, fuzzy normal and triangular distribution membership functions, and preliminary evaluation results are obtained. The validity index of the maximum membership principle is used to assess the accuracy of the evaluation results of two algorithms, and a relevant approaching degree is chosen to optimize the results. The entire evaluation process selects eleven indicators to form an evaluation set, including the boulder yield, root rate, flying distance of flyrock, explosive consumption, postcracking distance, detonator unit consumption, vibration velocity, loose coefficient, cast distance, throw rate, and blasted volume per meter of hole. Part of the indicator parameters are derived from field test monitoring, and another part of the indicator parameters are derived from numerical simulation. The simulation process uses the user-defined material interface function provided by LS-DYNA. And the numerical model of slope blasting is established by embedding the evolution relationship of tensile and compressive damage into the elastoplastic constitutive material. The evaluation method proposed in this paper is used to evaluate the postexplosion effect of Zijin Mountain gold-copper mine slope cast blasting. The results demonstrate that the fuzzy normal distribution membership function can correlate the state characteristic information and evaluation index effectively, and the working condition after explosion can be reflected accurately. Additionally, the influencing factors can be ranked by the importance degrees according to the calculated value of the evaluation index.

As an important commodity wood material, poplar has porous structure, which is worthy of being investigated. Especially, the lignin concentration is closely related to the formation and change of pores. Although a few studies reported the relation between the pore volume and the lignin removal of the ground wood (Stone and Scallan in J Polym Sci C Polym Symp 11(1):13–25, 1965), no information of the distribution of pore sizes versus the whole delignification process steps was reported. This study initially explored the effect of delignification level on the pore size distribution and pore structure in poplar cell walls using nitrogen absorption measurement. It was found that delignification increased the N2 adsorption amount and specific surface area of poplar. It also caused a large number of mesopores in the cell wall, mainly in the 2–10 nm pore size range, and decreased the average pore size. The whole delignification process can be roughly divided into three major phases, namely, the initial phase, the transitional phase and the stable phase, in which, the change process was not completely uniform. During the delignification, the size of poplar cell wall pores continued reducing, and finally a large number of uniform pores of about 2.1 nm was formed.

Small overlap accidents of sedans often result in severe injuries. In this article, a crashworthiness design method, including engine-room energy management, suspension safety design and passenger ...

In order to study the influence of the heat of hydration of concrete on the thermal disturbance of the frozen soil around the pile during the construction of cast-in-place piles in permafrost regions, a two-dimensional temperature field numerical model was established with non-steady state based on the heat conduction control equation and the freezing-thawing phase transition process. This model has been used to analyze and discuss the refreezing process of the concrete pile foundation, the heat flow transfer process between the pile foundation and the frozen soil, and the positive temperature curing process of the pile body. The research results have shown that the refreezing process of the pile foundation is a slow process. It took 28 days after the completion of the pile foundation construction that the negative temperature arose at the bottom of the pile. And 196 days after construction the temperature at the center of the pile dropped to negative. The heat brought by the cast-in-place pile is mainly released to the surrounding soil through the horizontal heat flow, and the thermal conductivity of the soil has a greater influence on this process.

ObjectivesTo investigate the value of contrast-enhanced computer tomography (CT)-based on radiomics in discriminating high-grade and low-grade hepatocellular carcinoma (HCC) before surgery.MethodsThe retrospective study including 161 consecutive subjects with HCC which was approved by the institutional review board, and the patients were divided into a training group (n = 112) and test group (n = 49) from January 2013 to January 2018. The least absolute shrinkage and selection operator (LASSO) was used to select the most valuable features to build a support vector machine (SVM) model. The performance of the predictive model was evaluated using the area under the curve (AUC), accuracy, sensitivity, and specificity.ResultsThe SVM model showed an acceptable ability to differentiate high-grade from low-grade HCC, with an AUC of 0.904 in the training dataset and 0.937 in the test dataset, accuracy (92.2% versus 95.7%), sensitivity(82.5% versus 88.0%), and specificity (92.7% versus 95.8%), respectively.ConclusionThe machine learning-based radiomics reflects a better evaluating performance in differentiating HCC between low-grade and high-grade, which may contribute to personalized treatment.

Due to prominent attenuation of travelling waves, double-end fault location method produces considerable errors when applied to ultra-high voltage AC half-wavelength transmission (UAHT) lines. In order to address this problem, this study presents a novel fault location scheme based on the attenuation, refraction and reflection principles of aerial-mode voltage travelling waves (AVTWs). In the scheme, the faulted section of an UAHT line is first determined according to the accumulated amplitude ratio of the pre-determined frequency components of AVTWs asynchronously sampled at both terminals of the line. Then, the Teager energy of the first two wavefronts of the AVTW - measured at the terminal in the faulted section - is used to define two factors for selecting the appropriate accurate fault location method. Finally, the single-end travelling wave method is adopted for locating the faults near the line terminal. The presented double-end method utilises the amplitude ratio of AVTWs to locate the faults far from the line terminal. A 1000 kV UAHT line model is considered with various fault conditions in PSCAD/EMTDC simulations. The simulation results and analysis demonstrate that the proposed fault location scheme is accurate and immune to fault distances, fault types, fault impedances and fault inception angles.

Based on distribution characteristics of the zero-sequence voltages and currents of the distributed lines, this paper proposes a novel single-phase-to-earth fault (SPTEF) location scheme for distribution network. Firstly, the distributed parameter model of distribution feeder lines is appropriately simplified according to Taylor series expansion of the hyperbolic functions in the model. The fault factor is constructed to select the faulted feeder. Then, the distribution features of the zero-sequence voltage (ZV) along the feeder line are analyzed. In each section of the faulted feeder, golden section search algorithm is adopted to distinguish the faulted section. Finally, based on the concept of maximum fault current, accurate fault location is realized by genetic algorithm (GA). A radial distribution network model containing different types of distribution lines is established in PSCAD/EMTDC. The calculation results of MATLAB prove the validity, reliability and accuracy of the proposed fault location scheme.

This article describes how as one of the hot parallel processors, the general-purpose graphics processing unit (GPU) has been widely adopted to accelerate various time-consuming algorithms. Dynamic programming (DP) optimization is a popular method to solve a particular class of complex problems. This article focuses on serial-monadic DP problems onto NVIDIA GPUs. As 0/1 knapsack is one of the most representational problems in this category and it often arises in many other fields of applications. The previous work proposed the compression method to reduce the amount of data transferred, but data in shared memory cannot be reused. This article demonstrates how to apply a more condensed data structure and the inter-block synchronization to efficiently map the serial-monadic DP onto GPUs. Computational experiments reveal that the best performance improvement of the approach is about 100% comparing with the previous work.

Additional file 1: Supplementary Fig. 1. Correlations between NKX6.1 inactivation level in β cells and Age in non-diabetic subjects. a Correlation between NKX6.1 Nuc-Ins+ cells count and percentage with Aging in Non-diabetic subjects (n = 40). b Correlation between NKX6.1cytIns− cell count and percentage with Aging in non-diabetic subjects (n = 40). Supplementary Fig. 2. Correlations between NKX6.1 inactivation level in β cells and BMI in non-diabetic subjects. a Correlation between NKX6.1 Nuc-Ins+ cells count and percentage with BMI in Non-diabetic subjects (n = 40). b Correlation between NKX6.1cytIns− cell count and percentage with BMI in non-diabetic subjects (n = 40). Supplementary Fig. 3. Correlations between NKX6.1 inactivation level in β cells and HbA1c in non-diabetic subjects. a Correlation between NKX6.1 Nuc-Ins+ cells count and percentage with HbA1c in non-diabetic subjects (n = 40). b Correlation between NKX6.1cytIns− cell count and percentage with HbA1c in non-diabetic subjects (n = 40)

In recent years, Extreme Learning Machine (ELM) has attracted extensive attention in various research fields. To improve the performance of ELM, we propose an Extreme Learning Machine Based on Double Kernel Risk-Sensitive Loss (DKRSLELM) method in this paper. The Kernel Risk-Sensitive Loss (KRSL) is integrated into the objective function of ELM. This is because KRSL not only can effectively reduce the influence of noise and outliers, but also can eliminate the redundant neurons of ELM. In the experiment, we conduct a classification experiment on cancer integration data-sets. The experimental results indicate that our method can effectively improve the classification performance of ELM.

Background As a machine learning method with high performance and excellent generalization ability, extreme learning machine (ELM) is gaining popularity in various studies. Various ELM-based methods for different fields have been proposed. However, the robustness to noise and outliers is always the main problem affecting the performance of ELM. Results In this paper, an integrated method named correntropy induced loss based sparse robust graph regularized extreme learning machine (CSRGELM) is proposed. The introduction of correntropy induced loss improves the robustness of ELM and weakens the negative effects of noise and outliers. By using the L2,1-norm to constrain the output weight matrix, we tend to obtain a sparse output weight matrix to construct a simpler single hidden layer feedforward neural network model. By introducing the graph regularization to preserve the local structural information of the data, the classification performance of the new method is further improved. Besides, we design an iterative optimization method based on the idea of half quadratic optimization to solve the non-convex problem of CSRGELM. Conclusions The classification results on the benchmark dataset show that CSRGELM can obtain better classification results compared with other methods. More importantly, we also apply the new method to the classification problems of cancer samples and get a good classification effect.