Search

In this paper, we improve the broad learning system (BLS) by speeding up the incremental learning for added inputs. We propose an efficient implementation for a step that is in the pseudoinverse computation of a partitioned matrix, to reduce the computational complexity. The proposed efficient implementation has two different forms for the cases of q > k and q ≤ k , respectively, where q and k denote the number of additional training samples and the total number of nodes, respectively. The proposed implementation for q > k utilizes the inverse of a sum of matrices to compute only a k × k matrix inverse, instead of a q × q matrix inverse in the original implementation, and the corresponding speedup for the matrix inversion operation in the number of floating-point operations is 1 2 (q / k) 3 . Moreover, it also speeds up two relevant matrix multiplication operations in the original implementation. On the other hand, the proposed implementation for q ≤ k speeds up one matrix multiplication operation in the original implementation. The numerical simulations show that both the proposed and original implementations always achieve the same testing accuracy. On the Modified National Institute of Standards and Technology dataset, the speedups of the proposed efficient BLS implementation over the original BLS implementation in total training time are 1.35–1.57 when q > k and 1.09–1.13 when q < k , while on the NYU object recognition benchmark dataset, the speedups are 1.10–1.35 when q > k and 1.07–1.09 when q < k. [ABSTRACT FROM AUTHOR]

3-phosphoinositide-dependent protein kinase-1 (PDK-1) is a key kinase regulating the activity of the PI3K/AKT pathway and a major regulator of the AGC protein kinase family. It is essential in the physiological activities of cells, embryonic development, individual development and immune response. In this study, we have identified for the first time an analogue of PDK-1 in the most primitive vertebrate, lamprey, and named it PDK-1-like. The protein sequence similarity of lamprey PDK-1-like to human, mouse, chicken, African xenopus and zebrafish PDK-1 were 64.4 %, 64.5 %, 65.0 %, 61.3 % and 63.2 %, respectively. The phylogenetic tree showed that PDK-1-like of lamprey were located at the base of the vertebrate branch, in line with the trend of biological evolution. Meanwhile, homology analysis showed that PDK-1 proteins across species shared a conserved kinase structural domain and a Pleckstrin Homology (PH) domain. Genomic synteny analysis revealed that the large-scale duplication blocks were not found in lamprey genome and neighbor genes of lamprey PDK-1-like presented dramatic differences compared with jawed vertebrates. More importantly, qPCR analysis showed that PDK-1-like was widely expressed in lamprey. Its mRNA expression levels varied in response to different pathogenic stimuli, and its expression was generally up-regulated under Polyinosinic-Polycytidylic acid (Poly(I:C)) stimulation. Pearson's correlation analysis showed that PDK-1-like was involved in co-expressed with MyD88-independent TLR-3 pathway during the immune response of lamprey, instead of MyD88-dependent TLR-3 pathway. In summary, our composite results offer valuable clues to the origin and evolution of PDK-1, and imply that PDK-1 s are among the most ancestral immune regulators in vertebrates. [ABSTRACT FROM AUTHOR]

In 2017, China began to comprehensively control rural water pollution in two steps, and aims to complete the purification of rural domestic sewage by the end of 2030. The short-term goal was to implement rural drinking water safety as soon as possible, and the medium-term and long-term goals were to completely block all kinds of pollution sources. Based on the field survey data, a variety of mathematical models were established to evaluate the progress of rural sewage treatment, the operation efficiency of sewage treatment facilities and the current situation of water pollution in China. The results show that the short-term goal has been basically achieved. The rural tap water penetration rate has reached 98%, and about 95% of the water has reached the drinking standard. About 70% of the groundwater is safe for bathing. However, the rural economy is weak, the residence is scattered and the unit sewage treatment cost is high, so it is very difficult to implement rural sewage treatment. The implementation of medium-term and long-term goals is slow, especially the treatment rate of agricultural non-point source sewage and domestic sewage is generally low, and surface water such as rural ponds and reservoirs basically belongs to Class IV or Class V. The study found that the key to the poor effect and slow progress of sewage treatment is the lack of economic support. It is suggested that the government should further increase investment in rural sewage treatment facilities. [ABSTRACT FROM AUTHOR]

Background: Although the clinical application of osimertinib, a third-generation epidermal growth factor receptor-tyrosine kinase inhibitor (EGFR-TKI), has been a new step forward in the first-line treatment of non-small cell lung cancer (NSCLC), an increasing number of patients with progression on osimertinib represents a great challenge clinically. The patterns of resistance mechanisms and subsequent treatment strategies after first-line osimertinib resistance are not well established. Methods: Between January 1, 2016 and October 31, 2020, a consecutive of 56 EGFR-mutant lung cancer patients treated with osimertinib as first-line therapy at Daping Hospital (Chongqing, China) were retrospective screened. The samples of pre-osimertinib and osimertinib-resistance were all detected by next-generation sequencing (NGS) panels. Statistical analyses were carried out using SPSS 23.0 software. Survival analyses were performed using the Kaplan–Meier method and compared using a log-rank test between groups. Results: Among 47 patients with osimertinib effectiveness analysis, the median progression free survival (mPFS) was 15.4 months (95% confidence interval [CI]: 12.2-24.9 months), and median overall survival (mOS) was 35.5 months (95% CI: 23.9 months -NA). A total of 21 patients underwent repeated NGS tests upon osimertinib resistance. MET amplification was the most common resistance mechanism (6/21, 28.6%), followed by C797S mutation (5/21, 23.8%). A total of 15 patients received subsequent treatments, with mPFS of 7.3 months (95% CI 5.0 months -NA). Among them, 7 patients with EGFR C797 S or/and MET amplification received subsequent second-line targeted therapy, achieving mPFS of 7.3 months (95% CI 4.5 months -NA). Of note, 3 patients received immunotherapy as second- or third-line treatment after osimertinib resistance, achieving median clinical benefit of 37.3 months. Conclusions: MET amplification and C797S mutation are main resistance mechanisms, which could be targeted by crizotinib and gefitinib, respectively. More than 50% patients could receive subsequent anticancer targetable therapies after first-line osimertinib resistance. Immunotherapy may also be an acceptable choice after osimertinib resistance. [ABSTRACT FROM AUTHOR]

Background: The prognosis of non–small‐cell lung cancer (NSCLC) with leptomeningeal metastasis (LM) is poor. Detection of cell‐free DNA (cfDNA) by next generation sequencing (NGS) in cerebrospinal fluid (CSF) may facilitate diagnosis of LM and identification of drug resistance mechanisms, yet its clinical use needs to be further verified. Methods: We performed a retrospective cohort study to assess the genetic profiles of paired CSF and plasma samples in lung cancer patients with LM. Of 17 patients screened, a total of 14 patients with LM and paired NGS tests were enrolled. Results: All patients harbor driver gene mutations, including 12 epidermal growth factor receptor (EGFR) activating mutations, 1 anaplastic lymphoma kinase (ALK) rearrangement, and 1 ROS‐1 fusion. Genetic mutations were detected in CSF cfDNA from 92.9% patients (13/14), which was significantly higher than that from the plasma (9/14, 64.2%). The mutations were highly divergent between CSF and plasma cfDNA, with a concordance rate of 24.38% and 10 mutations shared by the two media. CSF cfDNA could also benefit the analysis of resistance mechanisms to targeted therapies. In five patients who experienced progression on 1st or 2nd generation EGFR‐tyrosine kinase inhibitors (TKIs), RB1 mutation, and amplification of MET and EGFR were detected in CSF cfDNA only. In eight patients with LM progression on osimertinib resistance, EGFR amplification was detected in CSF cfDNA from four patients, whereas no CNVs were detected in the matched plasma samples. Conclusions: In conclusion, CSF could be superior to plasma in providing a more comprehensive genetic landscape of LM to find out drug resistance mechanisms and guide subsequent treatments. [ABSTRACT FROM AUTHOR]

The broad learning system (BLS) is an algorithm that facilitates feature representation learning and data classification. Although weights of BLS are obtained by analytical computation, which brings better generalization and higher efficiency, BLS suffers from two drawbacks: 1) the performance depends on the number of hidden nodes, which requires manual tuning, and 2) double random mappings bring about the uncertainty, which leads to poor resistance to noise data, as well as unpredictable effects on performance. To address these issues, a kernel-based BLS (KBLS) method is proposed by projecting feature nodes obtained from the first random mapping into kernel space. This manipulation reduces the uncertainty, which contributes to performance improvements with the fixed number of hidden nodes, and indicates that manually tuning is no longer needed. Moreover, to further improve the stability and noise resistance of KBLS, a progressive ensemble framework is proposed, in which the residual of the previous base classifiers is used to train the following base classifier. We conduct comparative experiments against the existing state-of-the-art hierarchical learning methods on multiple noisy real-world datasets. The experimental results indicate our approaches achieve the best or at least comparable performance in terms of accuracy. [ABSTRACT FROM AUTHOR]

In recent years, the appearance of the broad learning system (BLS) is poised to revolutionize conventional artificial intelligence methods. It represents a step toward building more efficient and effective machine-learning methods that can be extended to a broader range of necessary research fields. In this survey, we provide a comprehensive overview of the BLS in data mining and neural networks for the first time, focusing on summarizing various BLS methods from the aspects of its algorithms, theories, applications, and future open research questions. First, we introduce the basic pattern of BLS manifestation, the universal approximation capability, and essence from the theoretical perspective. Furthermore, we focus on BLS’s various improvements based on the current state of the theoretical research, which further improves its flexibility, stability, and accuracy under general or specific conditions, including classification, regression, semisupervised, and unsupervised tasks. Due to its remarkable efficiency, impressive generalization performance, and easy extendibility, BLS has been applied in different domains. Next, we illustrate BLS’s practical advances, such as computer vision, biomedical engineering, control, and natural language processing. Finally, the future open research problems and promising directions for BLSs are pointed out. [ABSTRACT FROM AUTHOR]

Long-term survival benefit has been noticed in non-small-cell lung cancer (NSCLC) patients treated with immune checkpoint inhibitors (ICIs), such as PD-1 inhibitors. However, it is still controversial whether patients with EGFR-activating mutations may benefit from ICIs. Recently, in stage IIIA NSCLC, chemo-immunotherapy has led to significant pathological response, yet patients with the presence of known EGFR mutations were excluded from some randomized trials of neoadjuvant therapy. Herein, we report a case of a 50-year-old female patient, who was initially diagnosed as stage IIIA lung squamous cell carcinoma. Immunohistochemistry analysis showed that the patient presented with high PD-L1 expression. Then, chemo-immunotherapy was given to the patient but the disease progressed quickly with distant metastasis. A re-biopsy revealed a poorly differentiated lung adenocarcinoma together with EGFR p.L858R mutation. Then the patient received gefitinib, which resulted in significant regression of primary lung lesion. A detailed examination of pre-treatment tumor sections demonstrated rare infiltration of CD8+ T cells, indicating that the current patient presented with an "immune-cold" microenvironment, which might explain the primary resistance to chemo-immunotherapy. Taken together, our case indicated that comprehensive detection of PD-L1 expression, driver gene status, together with tumor immune microenvironment, may offer a better prediction of treatment efficacy. [ABSTRACT FROM AUTHOR]

The fuzzy broad learning system (FBLS) is a recently proposed neuro-fuzzy model that shares the similar structure of a broad learning system (BLS). It shows high accuracy in both classification and regression tasks and inherits the fast computational nature of a BLS. However, the ensemble of several fuzzy subsystems in an FBLS decreases the possibility of understanding the fuzzy model since the fuzzy rules from different fuzzy systems are difficult to combine together while keeping the consistence. To balance the model accuracy and complexity, a synthetically simplified FBLS with better interpretability, named compact FBLS (CFBLS), is developed in this article, which can generate much fewer and more explainable fuzzy rules for understanding. In such a way, only one traditional Takagi–Sugeno–Kang fuzzy system is employed in the feature layer of a CFBLS, and the input universe of discourse is equally partitioned to obtain the fuzzy sets with proper linguistic labels accordingly. The random feature selection matrix and rule combination matrix are employed to reduce the total number of fuzzy rules and to avoid the “curse of dimensionality.” The enhancement layer is kept in the CFBLS that helps to add a nonlinear transformation of input features to the traditional first-order polynomial used in the consequent part of a fuzzy rule. The pseudoinverse is also used to determine the parameters of CFBLS guaranteeing its fast computational nature. The experiments on the popular UCI and KEEL datasets indicate that the CFBLS can generate a smaller set of comprehensible fuzzy rules and achieve much higher accuracy than some state-of-the-art neuro-fuzzy models. Moreover, the advantage of CFBLS is also verified in a real-world application. [ABSTRACT FROM AUTHOR]

The broad learning system (BLS) has been proved to be effective and efficient lately. In this article, several deep variants of BLS are reviewed, and a new adaptive incremental structure, Stacked BLS, is proposed. The proposed model is a novel incremental stacking of BLS. This invariant inherits the efficiency and effectiveness of BLS that the structure and weights of lower layers of BLS are fixed when the new blocks are added. The incremental stacking algorithm computes not only the connection weights between the newly stacking blocks but also the connection weights of the enhancement nodes within the BLS block. The Stacked BLS is considered as the increment of “layers” and “neurons” dynamically during the training for multilayer neural networks. The proposed architecture along with the training algorithms that utilizes the residual characteristic is very versatile in comparison with traditional fixed architecture. Finally, experimental results on UCI datasets, MNIST dataset, NORB dataset, CIFAR-10 dataset, SVHN dataset, and CIFAR-100 dataset indicate that the proposed method outperforms the selected state-of-the-art methods on both accuracy and training speed, such as deep residual networks. The results also imply that the proposed structure could highly reduce the number of nodes and the training time of the original BLS in the classification task of some datasets. [ABSTRACT FROM AUTHOR]

Purpose: About one-third of nonsmall cell lung cancer (NSCLC) patients develop brain metastases (BM). However, there is an unmet need for early diagnosis and treatment of BM. The precise mechanism for BM is still unknown. However, the genetic heterogeneity between primary tumor and paired BM indicates that sampling from the primary tumor may not be able to fully represent the mutational status in metastases. In this study, the genetic heterogeneity of primary lung adenocarcinoma and paired BM was analyzed. Patients and methods: A total of 11 paired samples of primary tumors and BM from lung cancer patients were included, in which 7 paired samples of patients were finally analyzed. Samples were sequenced by whole-exome sequencing (WES) to investigate the common and unique mutations in the primary tumors and BM, and the similarities and differences in copy number variation (CNV). Results: The consistency of gene mutation between primary lung adenocarcinoma and paired BM was 33% to 86%. FAM129C and ADAMTSs specifically mutated in BM, along with NKX2-1 high amplification and SAMD2/4 copy number deletion. Conclusion: The consistency of gene mutation between primary lung adenocarcinoma and corresponding BM is relatively high, while the individual differences were significant. FAM129C and ADAMTSs mutations and high amplification of NKX2-1 may be related to BM of lung cancer. The loss of copy number of SAMD2/4 may be a potential therapeutic target for BM from lung adenocarcinoma. [ABSTRACT FROM AUTHOR]

Background: Immune‐therapy with anti‐PD1 inhibitors, such as pembrolizumab, is revolutionizing the treatment of non‐small cell lung cancers (NSCLC). However, identifying patients for the potential therapeutic response and predicting therapy resistance and early relapse remains a challenge. Methods: Between 2016 and 2018, 60 patients were treated with pembrolizumab, among who 12 NSCLC patients had both baseline (before treatment) and serial (on treatment) periodical circulating tumor DNA (ctDNA) samples. Those samples were sequenced on a 329 pan cancer‐related gene panel. Analyses of tumor burden, blood tumor mutational burden (bTMB), maximum somatic allele frequency (MSAF), and tumor clonal structure were performed in association with clinical response. Candidate resistance mutations involved in relapse and metastases were further investigated. Results: ctDNA was detected and mutational profiling was performed for each patient. Those with a high baseline bTMB level showed significantly improved progression‐free survival (PFS) after pembrolizumab treatment. Tumor burden and therapeutic response significantly correlated with the MSAF instead of the bTMB. Clone analysis detected tumor progression about 2‐4 months ahead of computed tomography (CT) scan. One mutation in gene PTCH1 (Protein patched homolog 1) and two acquired anti‐PD1 candidate resistance mutations of gene B2M (β2 microglobulin) were identified in association with distant metastasis. The evolutionary tree of a representative patient was also described. Conclusion: This pilot study showed that MSAF could be another good indicator of therapeutic response, and clonal analysis could be clinically useful in monitoring clonal dynamics and detecting remote metastasis and early relapse. [ABSTRACT FROM AUTHOR]

Carbon-containing pellets were prepared with the carbonized product of agricultural wastes and iron concentrate, and an experimental study on the direct reduction was carried out. The experimental results demonstrated that carbon-containing pellets could be rapidly reduced at 1200 to 1300°C in 15 minutes, and the proper holding time at high temperature was 15 to 20 min. The degree of reduction gradually increased with temperature rising, and the appropriate temperature of reducing pellets was 1200°C. The weight loss rate and reduction degree of pellets increased with the rise of carbon proportion, and the relatively reasonable mole ratio of carbon to oxygen was 0.9. A higher content of carbon and an appropriate content of volatile matters in biomass char were beneficial to the reduction of pellets. The carbon-containing pellets could be reduced at high speeds in the air, but there was some reoxidization phenomenon. [ABSTRACT FROM AUTHOR]

In this paper, carbon-containing pellets were prepared by using crop-derived charcoal made from agricultural residuals and iron ore concentrates, and their pelletizing performance and properties were studied. Experimental results showed that the strengths of pellets were related to the particle size of concentrates and the content of moisture, bentonite, and crop-derived charcoal fines in the pelletizing mixture and the temperature of roasting and reduction. That the granularity of raw materials was fine and the bentonite content increased was beneficial to the improvement of pellet strengths. The suitable molar ratio of carbon to oxygen was 1.0 and the proper proportioning ratios of moisture and binder were 8.0% and 6.5%, respectively. The pellet strengths increased accordingly with increasing the reduction temperature, and when the temperature reached 1200°C, accompanied by the fast reduction of iron and the formation of crystal stock, the dropping strength of product pellets was 15 times and the compressive strength was 1650 N; this may be improved by grinding of the concentrate, leading to acceptable strength for the blast furnace. [ABSTRACT FROM AUTHOR]

Advanced Etch‐Stop s tructure In‐Ga‐Zn‐Oxide thin film transistor (A‐ES TFT) using SD and IGZO layer self‐aligned process is lower production cost, less parasitic capacitance than etch‐stopped layer (ESL) Oxide TFTs. We also found excellent electrical properties and bias stability and fabricated a 12.5‐inch liquid crystal panel using A‐ES TFTs by only two photomask. [ABSTRACT FROM AUTHOR]

Two hydrophilic conjugated polymers, PmP-NOH and PmP36F-NOH, with polar diethanol­amine on the side chains and main chain structures of poly( meta-phenylene) and poly( meta-phenylene- alt-3,6-fluorene), respectively, are successfully synthesized. The films of PmP-NOH and PmP36F-NOH show absorption edges at 340 and 343 nm, respectively. The calculated optical bandgaps of the two polymers are 3.65 and 3.62 eV, respectively, the largest ones so far reported for hydrophilic conjugated polymers. PmP-NOH and PmP36F-NOH also possess deep-lying highest occupied molecular orbital levels of −6.19 and −6.15 eV, respectively. Inserting PmP-NOH and PmP36F-NOH as a cathode interlayer in inverted polymer solar cells with a PTB7/PC71BM blend as the active layer, high power conversion efficiencies of 8.58% and 8.33%, respectively, are achieved, demonstrating that the two hydrophilic polymers are excellent interlayers for efficient inverted polymer solar cells. [ABSTRACT FROM AUTHOR]