Propagation Strategy Research Articles

RECIST-Induced Reliable Learning: Geometry-Driven Label Propagation for Universal Lesion Segmentation.

Automatic universal lesion segmentation (ULS) from Computed Tomography (CT) images can ease the burden of radiologists and provide a more accurate assessment than the current Response Evaluation Criteria In Solid Tumors (RECIST) guideline measurement. However, this task is underdeveloped due to the absence of large-scale pixel-wise labeled data. This paper presents a weakly-supervised learning framework to utilize the large-scale existing lesion databases in hospital Picture Archiving and Communication Systems (PACS) for ULS. Unlike previous methods to construct pseudo surrogate masks for fully supervised training through shallow interactive segmentation techniques, we propose to unearth the implicit information from RECIST annotations and thus design a unified RECIST-induced reliable learning (RiRL) framework. Particularly, we introduce a novel label generation procedure and an on-the-fly soft label propagation strategy to avoid noisy training and poor generalization problems. The former, named RECIST-induced geometric labeling, uses clinical characteristics of RECIST to preliminarily and reliably propagate the label. With the labeling process, a trimap divides the lesion slices into three regions, including certain foreground, background, and unclear regions, which consequently enables a strong and reliable supervision signal on a wide region. A topological knowledge-driven graph is built to conduct the on-the-fly label propagation for the optimal segmentation boundary to further optimize the segmentation boundary. Experimental results on a public benchmark dataset demonstrate that the proposed method surpasses the SOTA RECIST-based ULS methods by a large margin. Our approach surpasses SOTA approaches over 2.0%, 1.5%, 1.4%, and 1.6% Dice with ResNet101, ResNet50, HRNet, and ResNest50 backbones.

Common aquarium antiseptics do not cause long-term shifts in coral microbiota but may impact coral growth rates

IntroductionThough bacterial pathogens and parasites can compromise coral health, coral microbiome research increasingly suggests a beneficial role for bacterial species living in coral tissue and mucus. Recent studies suggest the application of targeted antibiotic treatments, while inhibiting the growth of harmful bacteria, may have unintended and persistent impacts on coral health. Land-based coral nurseries use antiseptic treatments such as Lugol’s solution and KoralMD™ dip to reduce infectious agents as part of restoration best practices. These antiseptic treatments often halt tissue loss, but the short- or long-term effects of these treatments on the coral microbiome is unknown.MethodsWe conducted a controlled tank experiment to assess the effects of these broad-spectrum treatments on coral growth rates and microbial communities when used as a prophylactic measure on healthy corals. Sixty individuals from each of two genotypes of the coral species Acropora palmata and Orbicella faveolata were treated with either Lugol’s solution or KoralMD™. Coral tissue, mucus, and skeleton were sampled pre-treatment, during treatment, and 1 and 2 months after treatment to assess microbiome shifts and recovery. The impact of the two treatments on coral growth was assessed using surface area measurements from 3D imagery.ResultsAlthough we found that A. palmata treated with Lugol’s solution had significantly reduced growth rates compared with untreated controls, impacts of antiseptic treatment were otherwise limited and microbiomes were not significantly different by treatment either immediately after application or 2 months thereafter. DiscussionStudy of the effects of these widely-used interventions may have significant repercussions on management and propagation strategies for corals reared in land-based nurseries. Furthermore, our findings indicate that antiseptic treatments can be applied to mitigate coral health issues without long-term harmful effects or significant microbiome shifts.

An analysis of expert-opined strategies in multi-attribute decision-making using effort propagation

Mining the interrelationships and possible hierarchical structures among the factors in real-world multi-attribute decision-making (MADM) problems is considered one of the primary tasks. One major task is determining an optimal strategy to work on the factors to enhance the effect on the goal attribute. This paper proposes two such strategies: parallel and hierarchical effort assignment and propagation strategies. We formally define and describe the concept of effort propagation through a strategy. Both the parallel and hierarchical strategies are divided into sub-strategies based on whether the assignment of efforts to the factors is uniform or depends upon some appropriate heuristics related to the factors in the system. The adapted and discussed heuristics are the factors’ relative significance and effort propagability. The strategies are analyzed for a real-life case study regarding high school administrative factors that enhance students’ performance. Total effort propagation of around 7%–15% to the goal is seen across the proposed strategies, given a total of 1 unit of effort to the directly accessible factors of the system. A comparative analysis is adapted to determine the optimal strategy among the proposed ones to enhance student performance effectively. The highest effort propagation achieved in the work is approximately 14.4348%. The analysis establishes the necessity of research towards the direction of effort propagation analysis in case of decision-making problems.

A Multilayer Framework for Online Metric Learning.

Online metric learning (OML) has been widely applied in classification and retrieval. It can automatically learn a suitable metric from data by restricting similar instances to be separated from dissimilar instances with a given margin. However, the existing OML algorithms have limited performance in real-world classifications, especially, when data distributions are complex. To this end, this article proposes a multilayer framework for OML to capture the nonlinear similarities among instances. Different from the traditional OML, which can only learn one metric space, the proposed multilayer OML (MLOML) takes an OML algorithm as a metric layer and learns multiple hierarchical metric spaces, where each metric layer follows a nonlinear layer for the complicated data distribution. Moreover, the forward propagation (FP) strategy and backward propagation (BP) strategy are employed to train the hierarchical metric layers. To build a metric layer of the proposed MLOML, a new Mahalanobis-based OML (MOML) algorithm is presented based on the passive-aggressive strategy and one-pass triplet construction strategy. Furthermore, in a progressively and nonlinearly learning way, MLOML has a stronger learning ability than traditional OML in the case of limited available training data. To make the learning process more explainable and theoretically guaranteed, theoretical analysis is provided. The proposed MLOML enjoys several nice properties, indeed learns a metric progressively, and performs better on the benchmark datasets. Extensive experiments with different settings have been conducted to verify these properties of the proposed MLOML.

Interpret what a Convolutional Neural Network learns for fault detection and diagnosis in process systems

The focus of this work is on an interpretation strategy on what a Convolutional Neural Network (CNN) has learned for fault detection and diagnosis (FDD) in process systems. Frequency spectra of process variables obtained by Continuous Wavelet Transform (CWT) are adopted as input features. Then, a CNN structure is designed to represent the mappings from input frequency features to different operation conditions. The Layer-wise Relevance Propagation (LRP) strategy is utilized to gain the relevance of each frequency feature to the classification performance. The formulations of relevance propagation for 4 types of CNN layers are presented in detail. The relevance scores are then depicted in heatmaps, where the pixels’ colors denote the contribution degrees and the most significant frequency features are considered as the major bases that the CNN discriminates different operation situations. The proposed interpretation strategy is experimented on the Tennessee Eastman process benchmark. The testing results demonstrate the efficiency of the strategy in interpreting what the CNN has learned to distinguish normal or faulty conditions in the FDD task.

Research on multi-layer network routing selection strategy based on cooperative evolutionary game in IoT environment

As a new technology and application mode, the Internet of Things has an important impact on social life and economic development. In recent years, low-cost optimization of network transmission to solve the congestion problem of multi-layer communication networks has become one of the research hotspots. In this paper, a multi-dimensional communication data transmission model based on a multi-layer network is proposed. It then uses cooperative evolutionary game theory to calculate revenue, update weights, and adapt neighbors. Finally, the attention mechanism is dynamically introduced to share the weights of the multi-layer network, and the multi-dimensional communication propagation and routing strategies in the Internet of Things are studied and analyzed. The experimental results show that the model proposed in this paper has higher game revenue and application value than traditional single-layer network game theory. In particular, the indicators of cooperation rate, stable state, and maximum cooperation rate are better than the latter. The research results of this paper have important reference value for solving the problems of cooperation dilemma, social stickiness, and synergy in multi-layer networks.

Clonal propagation and hardening of epiphytic orchids (Aerides multiflora and Rhyncostylis retusa) for sustainable conservation and utilization

The study involves devising an efficient method for in vitro protocorm-based plant production, as well as ex vitro establishment of Aerides multiflora and Rhynchostylis retusa. Both the orchids fall prey to the unscrupulous collection from the wild as they are widely used in the traditional medicinal system. Tissue culture technology can be effectively used for mass production of both the orchid species if the planned methodology is used for transplanting the regenerated plantlets into field conditions. MS medium supplemented with natural additives proved to be most optimal for protocorm development from seed explants (giving an average of 94 % to 97 % germination) and further multiplication while Mitra medium was best suited for complete plantlet regeneration from cultured protocorms. In A. multiflora an average of 3.33 shoots and 2.67 roots per protocorm, and in R. retusa mean number of 4.33 shoots and 3.10 roots per protocorm could be obtained on Mitra medium. Tissue culture raised plantlets of both the orchids were hardened in vitro and then shifted under polyhouse conditions. Peat moss, coconut husk, bark shavings, charcoal, and brick pieces, in varying combinations, were tried for optimal growth. Based on the best plantlet survival rate, 90 % in A. multiflora and 100 % in R. retusa, the growth medium combination of peat moss with charcoal, and bark shavings was selected to be the most optimal for orchid growth when kept at a temperature of 25–30 °C, relative humidity 60–70 %, and periodic watering only to avoid drying. The standardized methodology can be utilized for the field plantation of these two important orchids as a means of conservation and propagation strategy.

Estratégias de propagação vegetativa em cultivares comerciais de abacaxizeiro

ABSTRACT The production of healthy pineapple seedlings for commercial plantations has been one of the main limiting factors of the species, and the search for improvements to meet this demand has been constant. Therefore, the present study aimed to compare different propagation strategies (natural tillering, stem sectioning, and micropropagation) in three commercial cultivars 'BRS Imperial', 'Pérola', and Gold 'MD-2' of pineapple. Three methods of vegetative propagation were evaluated: conventional, stem sectioning, and micropropagation. The evaluations consisted of the number of seedlings produced by each method and the period for their development until planting in the field. According to the conventional propagation technique results, the young seedling type in the cultivar 'Pérola' was more abundant. In the propagation by sectioning the stem, the 'BRS Imperial' cultivar presented the highest number of seedlings. Regarding the micropropagation technique, the highest production of shoots was observed in the third subculture for all cultivars, especially for ‘BRS Imperial’, which presented the highest total number of shoots in all subcultures. It was concluded that the micropropagation technique proved efficient in producing uniform seedlings on a large scale, in addition to having an advantage in the greater number of seedlings produced compared to the other methods studied.

Optimizing Tectona grandis L.F. (teak) Tree Clonal Propagation via Regrowth and the Application of Plant Growth Regulators

Tectona grandis L.F., known as teak, is a species of high commercial and ecological value for which the development of efficient clonal propagation strategies is being sought. The availability of clonally propagated materials from plus trees is scarce. The main objective of the research was to evaluate the root induction capacity of Tectona grandis L.F. plus tree sprouts using different hormone concentrations of indole-3-butyric acid (IBA) and 1-naphthaleneacetic acid (NAA) selected at 30 and 60 days. A completely randomized design (CRD) with seven treatments, four replicates, and three experimental units per replicate was used to determine statistical differences between treatments. The best results for rooting of teak shoots were observed in treatment T5 (500 mg kg-1 of IBA) with an average of three roots per plant, T1 (control), and T7 (1500 mg kg-1 of IBA) with an average of four leaves per plant. The highest survival rates were observed in T2 (500 mg kg-1 of NAA) and T7 (1500 mg kg-1 of IBA) with 58.33% and in T1 (control), T2 (500 mg kg-1 of NAA) 50% rooting was obtained after 60 days. Although no differences were observed between treatments, the numerical differences allowed to observe that the use of auxins promotes rooting and survival of the shoots of adult trees of T. grandis and opens the possibility of propagation and cloning strategies for this species. This study provides valuable knowledge on vegetative propagation, breeding and conservation of forest species, which is relevant for academia, public and private sectors.

Host proteins Alpha-2-Macroglobulin and LRP1 associate with Chandipura virus

Chandipura Virus is an emerging tropical pathogen with a high mortality rate among children. No mode of treatment or antivirals exists against CHPV infection, due to little information regarding its host interaction. Studying viral pathogen interaction with its host can not only provide valuable information regarding its propagation strategy, but also on which host proteins interact with the virus. Identifying these proteins and understanding their role in the infection process can provide more stable anti-viral targets. In this study, we focused on identifying host factors that interact with CHPV and may play a critical role in CHPV infection. We are the first to report the successful identification of Alpha-2-Macroglobulin (A2M), a secretory protein of the host that interacts with CHPV. We also established that LRP1 (Low-density lipoprotein receptor-related protein 1) and GRP78 (Glucose regulated protein 78), receptors of A2M, also interact with CHPV. Furthermore, we could also demonstrate that knocking out A2M has a severe effect on viral infection. We conclusively show the interaction of these host proteins with CHPV. Our findings also indicate that these host proteins could play a role in viral entry into the host cell.

Transplanting naturally regenerated tree seedlings for tropical forest restoration: a case study of Calophyllum brasiliense and Vochysia guatemalensis in Mexico

Plant species produce far more seedlings than those surviving to adulthood. It would seem reasonable to take advantage of that excess production, by relocating seedlings to desired restoration sites. There is, however, little information available on this issue. In the present study, we collected naturally regenerated seedlings of two native species, Calophyllum brasiliense (“barí”) and Vochysia guatemalensis (“corpo”), in old‐growth forest patches, and transplanted them for the enrichment of a secondary tropical forest in Veracruz, Mexico. For transplanted C. brasiliense seedlings, overall survival was 32% after 20 months, and for V. guatemalensis seedlings it was 66% after 26 months. The mean height (±standard error) of all C. brasiliense and V. guatemalensis seedlings that survived until the end of the study, was 33 ± 2 cm (from initially 12.9 ± 0.2 cm) and 52 ± 5 cm (from 4.6 ± 0.1 cm), respectively. We applied a cost–benefit model to assess the trade‐off between initial‐investment costs and the resulting survival of transplanting naturally regenerated seedlings in the current study, in comparison with direct seeding (with and without the protection of seeds and emerged seedlings) reported in a previous study for the same area and species. For C. brasiliense, the success of transplanting naturally regenerated seedlings was always inferior to direct seeding without protection, whereas for V. guatemalensis it was always superior. Comparing the transplantation of naturally regenerated seedlings with direct seeding with protection, the best propagation strategy depended on the (monetary) value that a treelet is expected to be worth after 2 years.

CGRS: Collaborative Knowledge Propagation Graph Attention Network for Recipes Recommendation

In the age of big data, recipe recommendation is of great significance. It can recommend recipes in line with the user's eating habits in massive data. Compared with other recommendation tasks, recipe recommendation is influenced by multiple aspects and requires fine-grained learning to obtain entity representations. Therefore, the traditional recommendation method cannot meet people's requirements. In this paper, we propose the Collaborative Knowledge Propagation Graph Attention Network for Recipes Recommendation (CGRS). This method designs collaborative information propagation to make full use of user interaction and recipe attribute information to meet the needs of multiple influencing factors. Use the graph attention feature learning network to obtain the high-order feature information of the entity to meet the demand for fine-grained representation. Specifically, the method first obtains the multi-hop triplet sets of users and recipes through a collaborative message propagation strategy. Then utilises a graph attention feature learning layer to learn the topological proximity structure features of the triplet sets. Obtain high-level semantic information of entities by superimposing network layers. Design an attention aggregator at the prediction layer to refine the embedding representation of entities. Finally predict the user-recipe interaction probability. Experimental results prove the advancement and effectiveness of CGRS.

A CRISPR-Cas9-Derived Male Sterility System for Tomato Breeding

Male sterility can reduce cost and enable high seed purity during hybrid seed production. However, the commercial application of male sterility in hybrid seed production has not been widely used in tomatoes. CRISPR/Cas9-mediated gene editing can facilitate acceleration for the practical application of male sterility in hybrid seed production. Here, by using the CRISPR-Cas9 system, two genes DYSFUNCTIONAL TAPETUM1 (SlDYT1) and Glutathione S-transferase (SlGSTAA), which underly the two closely linked loci Male sterile 10 (Ms10) and Anthocyanin absent (AA), were knocked out simultaneously in two tomato parental lines. The generated dyt1gstaa double mutants developed green hypocotyl owing to anthocyanin deficiency and exhibited stable male sterility. Up to 92% effectiveness in selecting male sterility was achieved using green hypocotyl as a morphological marker, and thereafter an efficient and stable propagation strategy of male sterility with the aid of the morphological marker selection was developed. Furthermore, dyt1gstaa-derived hybrid seeds were produced and found to have comparable yield, weight, and germination rate with the corresponding WT-derived F1 seeds. The dyt1gstaa system not only increased hybrid seed purity to 100% but also facilitated its rapid and cost-effective determination. Moreover, this system was discovered to have no evident side effects on important agronomic traits. This study suggested that our CRISPR/Cas9-created dyt1gstaa system can be deployed in tomato hybrid seed production.

A novel image recommendation model based on user preferences and social relationships

Recent work on personalized recommendation advocates that exploring user visual preferences from their interactions with images has become a future endeavor for image recommendation. However, there are inevitable discrepancies between the images favored and the visual preference representations reflected by users, therefore we strive to reveal user visual preferences from their behaviors and social relationships. In this work, we propose a collaborative graph model of user preferences for image recommendation, which establishes a higher-order collaborative relationship between user visual preferences and image semantic information. In user visual preferences, the underlying visual preferences hidden in social relationship is explored by combining the idea of contrast learning. In the higher-order collaboration module, a recursive interest propagation strategy is used to aggregate user preferences and image visual information. The experimental results clearly show that the model has an improvement of about 8.82% and 7.25% on Recall@20 and NDCG@20 compared to the latest methods.

Impact of uncertainties in biometric parameters on intraocular lens power formula predicted refraction using a Monte-Carlo simulation.

The purpose of this study was to investigate the uncertainty in the formula predicted refractive outcome REFU after cataract surgery resulting from measurement uncertainties in modern optical biometers using literature data for within-subject standard deviation Sw. This Monte-Carlo simulation study used a large dataset containing 16 667 preoperative IOLMaster 700 biometric measurements. Based on literature Sw values, REFU was derived for both the Haigis and Castrop formulae using error propagation strategies. Using the Hoya Vivinex lens (IOL) as an example, REFU was calculated both with (WLT) and without (WoLT) consideration of IOL power labelling tolerances. WoLT the median REFU was 0.10/0.12 dpt for the Haigis/Castrop formula, and WLT it was 0.13/0.15 dpt. WoLT REFU increased systematically for short eyes (or high power IOLs), and WLT this effect was even more pronounced because of increased labelling tolerances. WoLT the uncertainty in the measurement of the corneal front surface radius showed the largest contribution to REFU, especially in long eyes (and low power IOLs). WLT the IOL power uncertainty dominated in short eyes (or high power IOLs) and the uncertainty of the corneal front surface in long eyes (or low power IOLs). Compared with published data on the formula prediction error of refractive outcome after cataract surgery, the uncertainty of biometric measures seems to contribute with ⅓ to ½ to the entire standard deviation. REFU systematically increases with IOL power and decreases with axial length.

Cascading Failure Propagation and Mitigation Strategies in Power Systems

In this article, we investigate the local and nonlocal failure propagation patterns in power systems and propose effective mitigation strategies. It is widely acknowledged that the sequence of failure events of component overloading is determined by power flow redistribution. To understand the power flow redistribution process, we analyze the prefault power flow condition using network flow theory and by clustering the system into several regions with balanced local power flow. We observe that faults triggering local power imbalance would lead to remote power transfer and faults causing no local power imbalance would lead to local power redistribution. The improved understanding of the failure process permits us to develop effective mitigating strategies consisting of adaptive power balance restoration and selective edge protection for cascading failure propagation. In addition, we propose a set of indexes to measure the local and nonlocal propagation risks of cascading failures. Numerical simulations on the IEEE 118- and 300-bus systems demonstrate that the proposed mitigation strategies can significantly reduce the blackout sizes and are tolerant to incorrect transmission line tripping.

Assessment of the TRX2p-yEGFP Biosensor to Monitor the Redox Response of an Industrial Xylose-Fermenting Saccharomyces cerevisiae Strain during Propagation and Fermentation.

The commercial production of bioethanol from lignocellulosic biomass such as wheat straw requires utilizing a microorganism that can withstand all the stressors encountered in the process while fermenting all the sugars in the biomass. Therefore, it is essential to develop tools for monitoring and controlling the cellular fitness during both cell propagation and sugar fermentation to ethanol. In the present study, on-line flow cytometry was adopted to assess the response of the biosensor TRX2p-yEGFP for redox imbalance in an industrial xylose-fermenting strain of Saccharomyces cerevisiae during cell propagation and the following fermentation of wheat-straw hydrolysate. Rapid and transient induction of the sensor was recorded upon exposure to furfural and wheat straw hydrolysate containing up to 3.8 g/L furfural. During the fermentation step, the induction rate of the sensor was also found to correlate to the initial ethanol production rate, highlighting the relevance of redox monitoring and the potential of the presented tool to assess the ethanol production rate in hydrolysates. Three different propagation strategies were also compared, and it was confirmed that pre-exposure to hydrolysate during propagation remains the most efficient method for high ethanol productivity in the following wheat-straw hydrolysate fermentations.

A PPP-RTK APPROACH TO MASS-MARKET APPLICATIONS

Abstract. PPP-RTK has been widely investigated to take the advantages of both real-time kinematic (RTK) and precise point positioning (PPP) techniques. Prior to PPP-RTK, the conventional RTK based on the use of a single base station, on the one hand, has been extended to work within a regional network of multiple base stations, known as network RTK (NRTK). The RTK and NRTK enable fast ambiguity resolution over a short baseline or a local region. PPP, on the other hand, eliminates the need to establish any local network like NRTK, which is able to work in a single receiver mode but it suffers long convergence time in ambiguity resolution. PPP-RTK therefore can provide fast ambiguity resolution capability like RTK and NRTK. Current PPP-RTK techniques however still face challenges in supporting mass-market applications such as mobile devices and autonomous vehicles. Although PPP-RTK system (a combination of RTK and PPP technologies) can help expand the coverage of RTK and speed up the ambiguity resolution in PPP, the deployment and maintenance of a dense network of permanent base stations and a central data processing infrastructure for generation of SSR corrections increases not only the system cost but also the system complexity. This is particularly an obstacle for mass-market applications. In this paper, a new RTK approach is described. First it is based on a single base station state-space-representation (SSR) correction generation strategy to support fast ambiguity resolved PPP. Further it presents a new peer-to-peer propagation strategy to form a real-time dynamically generated network of base stations to support mass-market application users with unbounded coverage. As a result, the new approach eliminates the need to deploy and maintain a dense network of permanent base stations and central data processing infrastructure as required in a conventional PP-RTK system.

Recurrent Interaction Network for Stereoscopic Image Super-Resolution

Recently, deep learning-based stereoscopic image super-resolution has attracted extensive attention and made great progress. However, existing methods have not adequately explored the inter-view dependency among two-view multi-level features. In this paper, a recurrent interaction network for stereoscopic image super-resolution (RISSRnet) is proposed to learn the inter-view dependency. To efficiently utilize the relationship between the two views, a recurrent interaction module is designed to achieve recurrent interaction among two-view multi-level features from the regrouped sequences, which are generated by a coupled queue-regroup mechanism. In addition, to recursively enhance features in the recurrent interaction module, an iterative propagation strategy is developed for sufficient interaction. Extensive experimental results demonstrate the effectiveness and superiority of the proposed RISSRnet.

Community Detection Based on Multiobjective Particle Swarm Optimization and Graph Attention Variational Autoencoder

Community detection is an important research direction in complex network analysis that can help us discover valuable network structures. The community detection algorithms based on multiobjective particle swarm optimization encode community membership of nodes in particles and employ evolutionary strategies to search for the optimal community division. Existing algorithms face two challenges: (1) they are inapplicable to large networks because the evolution process is time-consuming; (2) they are easy to fall into local optima. In this paper, we propose a novel algorithm that combines a multiobjective particle swarm optimization algorithm based on label propagation with a graph attention variational autoencoder to realize community detection. On the one hand, the label propagation strategy is involved in the update of a swarm's particles to speed up its evolution. The optimal solutions found by the particle swarm optimization algorithm are embedded into the objective of the autoencoder to improve the embedding vectors quality. On the other hand, the embedding vectors are used to improve the solutions of the particle swarm optimization algorithm to avoid its early convergence. The experiments on artificial and real-world networks demonstrate the feasibility and effectiveness of our algorithm compared with the state-of-the-art algorithms.