Masked hypergraph learning for weakly supervised histopathology whole slide image classification.

BACKGROUND AND OBJECTIVES: Graph neural network (GNN) has been extensively used in histopathology whole slide image (WSI) analysis due to the efficiency and flexibility in modelling relationships among entities. However, most existing GNN-based WSI analysis methods only consider the pairwise correlation of patches from one single perspective (e.g. spatial affinity or embedding similarity) yet ignore the intrinsic non-pairwise relationships present in gigapixel WSI, which are likely to contribute to feature learning and downstream tasks. The objective of this study is therefore to explore the non-pairwise relationships in histopathology WSI and exploit them to guide the learning of slide-level representations for better classification performance. METHODS: In this paper, we propose a novel Masked HyperGraph Learning (MaskHGL) framework for weakly supervised histopathology WSI classification. Compared with most GNN-based WSI classification methods, MaskHGL exploits the non-pairwise correlations between patches with hypergraph and global message passing conducted by hypergraph convolution. Concretely, multi-perspective hypergraphs are first built for each WSI, then hypergraph attention is introduced into the jointed hypergraph to propagate the non-pairwise relationships and thus yield more discriminative node representation. More importantly, a masked hypergraph reconstruction module is devised to guide the hypergraph learning which can generate more powerful robustness and generalization than the method only using hypergraph modelling. Additionally, a self-attention-based node aggregator is also applied to explore the global correlation of patches in WSI and produce the slide-level representation for classification. RESULTS: The proposed method is evaluated on two public TCGA benchmark datasets and one in-house dataset. On the public TCGA-LUNG (1494 WSIs) and TCGA-EGFR (696 WSIs) test set, the area under receiver operating characteristic (ROC) curve (AUC) were 0.9752±0.0024 and 0.7421±0.0380, respectively. On the USTC-EGFR (754 WSIs) dataset, MaskHGL achieved significantly better performance with an AUC of 0.8745±0.0100, which surpassed the second-best state-of-the-art method SlideGraph+ 2.64%. CONCLUSIONS: MaskHGL shows a great improvement, brought by considering the intrinsic non-pairwise relationships within WSI, in multiple downstream WSI classification tasks. In particular, the designed masked hypergraph reconstruction module promisingly alleviates the data scarcity and greatly enhances the robustness and classification ability of our MaskHGL. Notably, it has shown great potential in cancer subtyping and fine-grained lung cancer gene mutation prediction from hematoxylin and eosin (H&E) stained WSIs.

Histopathology language-image representation learning for fine-grained digital pathology cross-modal retrieval.

Large-scale digital whole slide image (WSI) datasets analysis have gained significant attention in computer-aided cancer diagnosis. Content-based histopathological image retrieval (CBHIR) is a technique that searches a large database for data samples matching input objects in both details and semantics, offering relevant diagnostic information to pathologists. However, the current methods are limited by the difficulty of gigapixels, the variable size of WSIs, and the dependence on manual annotations. In this work, we propose a novel histopathology language-image representation learning framework for fine-grained digital pathology cross-modal retrieval, which utilizes paired diagnosis reports to learn fine-grained semantics from the WSI. An anchor-based WSI encoder is built to extract hierarchical region features and a prompt-based text encoder is introduced to learn fine-grained semantics from the diagnosis reports. The proposed framework is trained with a multivariate cross-modal loss function to learn semantic information from the diagnosis report at both the instance level and region level. After training, it can perform four types of retrieval tasks based on the multi-modal database to support diagnostic requirements. We conducted experiments on an in-house dataset and a public dataset to evaluate the proposed method. Extensive experiments have demonstrated the effectiveness of the proposed method and its advantages to the present histopathology retrieval methods. The code is available at https://github.com/hudingyi/FGCR.

Variability in Leaf Color Induced by Chlorophyll Deficiency: Transcriptional Changes in Bamboo Leaves.

The diversity of leaf characteristics, particularly leaf color, underscores a pivotal area of inquiry within plant science. The synthesis and functionality of chlorophyll, crucial for photosynthesis, largely dictate leaf coloration, with varying concentrations imparting different shades of green. Complex gene interactions regulate the synthesis and degradation of chlorophyll, and disruptions in these pathways can result in abnormal chlorophyll production, thereby affecting leaf pigmentation. This study focuses on Bambusa multiplex f. silverstripe, a natural variant distinguished by a spectrum of leaf colors, such as green, white, and green-white, attributed to genetic variations influencing gene expression. By examining the physiological and molecular mechanisms underlying chlorophyll anomalies and genetic factors in Silverstripe, this research sheds light on the intricate gene interactions and regulatory networks that contribute to leaf color diversity. The investigation includes the measurement of photosynthetic pigments and nutrient concentrations across different leaf color types, alongside transcriptomic analyses for identifying differentially expressed genes. The role of key genes in pathways such as ALA biosynthesis, chlorophyll synthesis, photosynthesis, and sugar metabolism is explored, offering critical insights for advancing research and plant breeding practices.

A Key-Points Based Anchor-Free Cervical Cell Detector.

Cervical cell detection is crucial to cervical cytology screening at early stage. Currently most cervical cell detection methods use anchor-based pipeline to achieve the localization and classification of cells, e.g. faster R-CNN and YOLOv3. However, the anchors generally need to be pre-defined before training and the detection performance is inevitably sensitive to these pre-defined hyperparameters (e.g. number of anchors, anchor size and aspect ratios). More importantly, these preset anchors fail to conform to the cells with different morphology at inference phase. In this paper, we present a key-points based anchor-free cervical cell detector based on YOLOv3. Compared with the conventional YOLOv3, the proposed method applies a key-points based anchor-free strategy to represent the cells in the initial prediction phase instead of the preset anchors. Therefore, it can generate more desirable cell localization effect through refinement. Furthermore, PAFPN is applied to enhance the feature hierarchy. GIoU loss is also introduced to optimize the small cell localization in addition to focal loss and smooth L1 loss. Experimental results on cervical cytology ROI datasets demonstrate the effectiveness of our method for cervical cell detection and the robustness to different liquid-based preparation styles (i.e. drop-slide, membrane-based and sedimentation).

Effects of Different Heterogeneous Nutrient Environments on the Growth and Activities of Enzymes in the Roots of Fokienia hodginsii Families.

Currently, research on the F. hodginsii asexual lineage primarily focuses on the screening of growth traits and the control of single fertilizer applications. The effects of the heterogeneity of soil nutrients on root growth and activity have not been studied in detail. Therefore, we propose forest management measures to improve the foraging ability of forest trees in conjunction with stand productivity. In this experiment, annual containerized seedlings of 10 free-pollinated F. hodginsii lines from a primary asexual seed orchard were used as test subjects, and three heterogeneous nutrient environments of nitrogen (N), phosphorus (P), and potassium (K) were constructed. In contrast, homogeneous nutrient environments were used as the control to carry out potting experiments, to study the growth of F. hodginsii lines and the differences in the activities of root enzymes under the three heterogeneous nutrient environments, and to carry out the comprehensive evaluation using the principal component and cluster analysis method. The results were as follows: (1) The seedling height of F. hodginsii family lines under a homogeneous nutrient environment was significantly higher than that of all heterogeneous nutrient environments; the diameter of the ground was the highest under N heterogeneous nutrient environment and significantly higher than that of all the other nutrient environments; the biomass of the root system was the highest under P heterogeneous nutrient environment, which was significantly higher than that of homogeneous nutrient environment and K heterogeneous nutrient environment. The catalase (CAT) activity of F. hodginsii roots was higher than that of homogeneous nutrients in all heterogeneous nutrient environments but not significant, and the superoxide dismutase (SOD) activity was slightly higher than that of K heterogeneous and homogeneous nutrient environments in N and P heterogeneous nutrient environments. SOD activity was slightly higher than that of K heterogeneous and homogeneous nutrient environments under N, and P. peroxidase (POD) activity in the F. hodginsii root system was the highest under the P heterogeneous nutrient environment, which was significantly higher than that of the other nutrient environments. Unlike the activities of the enzymes, the content of malondialdehyde (MDA) in the roots of F. hodginsii was higher in the heterogeneous environment than in all the other nutrient environments. (2) Under N and P heterogeneous nutrient environments, lines 552 and 590 had higher seedling height, ground diameter, and root enzyme activity, while root biomass was highest in line 544; and under K heterogeneous nutrient environments, line 591 had higher seedling height, ground diameter, and root enzyme activity while root biomass was highest in line 551. In contrast to the patterns of seedling height, accumulation of root biomass and activities of root enzymes, family No. 590 had the highest ground diameter of all the F. hodginsii families under the heterogeneous nutrient environments. Family No. 547 had the highest MDA content. In conclusion, it can be seen that N heterogeneous and homogeneous nutrient environments can significantly increase the seedling height and diameter of F. hodginsii compared with P and K heterogeneous nutrient environments, and N and P heterogeneous nutrient environments can also increase the root biomass, root enzyme activities and significantly reduce the MDA content of F. hodginsii. According to the principal component analysis and cluster analysis, it can be seen that among the 10 F. hodginsii family lines, family lines 590 and 552 have higher evaluation in growth, root biomass accumulation, and enzyme activity.

Identification and characterization of the cupin_1 domain-containing proteins in ma bamboo (Dendrocalamus latiflorus) and their potential role in rhizome sprouting.

Cupin_1 domain-containing protein (CDP) family, which is a member of the cupin superfamily with the most diverse functions in plants, has been found to be involved in hormone pathways that are closely related to rhizome sprouting (RS), a vital form of asexual reproduction in plants. Ma bamboo is a typical clumping bamboo, which mainly reproduces by RS. In this study, we identified and characterized 53 Dendrocalamus latiflorus CDP genes and divided them into seven subfamilies. Comparing the genetic structures among subfamilies showed a relatively conserved gene structure within each subfamily, and the number of cupin_1 domains affected the conservation among D. latiflorus CDP genes. Gene collinearity results showed that segmental duplication and tandem duplication both contributed to the expansion of D. latiflorus CDP genes, and lineage-specific gene duplication was an important factor influencing the evolution of CDP genes. Expression patterns showed that CDP genes generally had higher expression levels in germinating underground buds, indicating that they might play important roles in promoting shoot sprouting. Transcription factor binding site prediction and co-expression network analysis indicated that D. latiflorus CDPs were regulated by a large number of transcription factors, and collectively participated in rhizome buds and shoot development. This study significantly provided new insights into the evolutionary patterns and molecular functions of CDP genes, and laid a foundation for further studying the regulatory mechanisms of plant rhizome sprouting.

Kernel Attention Transformer for Histopathology Whole Slide Image Analysis and Assistant Cancer Diagnosis.

Transformer has been widely used in histopathology whole slide image analysis. However, the design of token-wise self-attention and positional embedding strategy in the common Transformer limits its effectiveness and efficiency when applied to gigapixel histopathology images. In this paper, we propose a novel kernel attention Transformer (KAT) for histopathology WSI analysis and assistant cancer diagnosis. The information transmission in KAT is achieved by cross-attention between the patch features and a set of kernels related to the spatial relationship of the patches on the whole slide images. Compared to the common Transformer structure, KAT can extract the hierarchical context information of the local regions of the WSI and provide diversified diagnosis information. Meanwhile, the kernel-based cross-attention paradigm significantly reduces the computational amount. The proposed method was evaluated on three large-scale datasets and was compared with 8 state-of-the-art methods. The experimental results have demonstrated the proposed KAT is effective and efficient in the task of histopathology WSI analysis and is superior to the state-of-the-art methods.

Autofluorescence imaging-assisted medical thoracoscopy in the diagnosis of malignant pleural disease.

BACKGROUND: Medical thoracoscopy (MT) does not always provide a conclusive diagnosis of pleural diseases because the endoscopic appearance of pleural diseases can be misleading. Autofluorescence imaging (AFI) is an effective assistive diagnostic tool. However, its clinical application for pleural disease remains controversial. OBJECTIVES: This prospective study evaluated the clinical usefulness of AFI-assisted MT for diagnosis of malignant pleural diseases. METHODS: Patients with unexplained pleural effusion admitted to our clinics between December 2018 and September 2021 were enrolled. We performed white-light thoracoscopy (WLT) first, and then AFI, during MT. Images of endoscopic real-time lesions were recorded under both modes. Pleural biopsy specimens were analyzed pathologically. Between-groups differences in diagnostic sensitivity, specificity, positive-predictive value (PPV), and negative-predictive value (NPV) were assessed using 95% confidence intervals (CI). Receiver operating characteristic curves and decision curve analyses were employed to analyze the diagnostic efficiency of these two modes. RESULTS: Of 126 eligible patients, 73 cases were diagnosed with malignant pleural disease. A total of 1292 biopsy specimens from 492 pleural sites were examined for pathological changes. The diagnostic sensitivity, PPV, and NPV of AFI were 99.7%, 58.2%, and 99.2%, respectively. AFI was significantly superior to WLT, which had a sensitivity of 79.7%, PPV of 50.7%, and NPV of 62.8%. Subgroup analysis showed that the AFI type III pattern was significantly more specific for pleural malignant disease than that of WLT. CONCLUSIONS: AFI could further improve the diagnostic efficacy of MT by providing better visualization, convenience, and safety.

Toward ultra-tough and heat-resistant biodegradable polylactide/core-shell rubber blends by regulating the distribution of rubber particles with stereocomplex crystallites.

Ultra-tough and heat-resistant poly(l-lactide)/core-shell rubber (PLLA/CSR) blends were fabricated by utilizing stereocomplex (SC) crystallites to effectively regulate the CSR distribution in PLLA matrix. Linear and 3-11 armed poly(d-lactide)s (PDLAs) were synthesized and then melt-mixed with PLLA/CSR blend. Interestingly, the incorporated PDLA chains could collaborate with PLLA chains to form dense SC crystallites network in PLLA/PDLA/CSR blends, thus inducing the CSR particles to transform from uniform distribution structure to network-like structure. With increasing the PDLA arm numbers, the size of CSR clusters in the network-like structure first increased and then decreased, and the continuity of the network-like structure first remained at a high level and then decreased obviously. The formation of CSR network-like structure could remarkably improve the impact strength of PLLA/PDLA/CSR blends without deteriorating their strength and modulus (compared with PLLA/CSR blend), and the CSR network-like structure with larger-sized CSR clusters and higher continuity could help obtain higher impact strength (78.3 kJ/m2). Moreover, the heat resistance of PLLA/PDLA/CSR blends could also be significantly improved (the highest Vicat softening temperature was 131 °C) by the SC crystallites network and CSR network-like structure. This work provides an effective strategy for controlling the rubber network-like morphology and thereby preparing high-performance PLLA materials.

Response of the root morphological structure of Fokienia hodginsii seedlings to competition from neighboring plants in a heterogeneous nutrient environment.

Introduction: Critical changes often occur in Fokienia hodginsii seedlings during the process of growth owing to differences in the surrounding environment. The most common differences are heterogeneous nutrient environments and competition from neighboring plants. Methods: In this study, we selected one-year-old, high-quality Fokienia hodginsii seedlings as experimental materials. Three planting patterns were established to simulate different competitive treatments, and seedlings were also exposed to three heterogeneous nutrient environments and a homogeneous nutrient environment (control) to determine their effect on the root morphology and structure of F. hodginsii seedlings. Results: Heterogeneous nutrient environments, compared with a homogeneous environment, significantly increased the dry matter accumulation and root morphology indexes of the root system of F. hodginsii, which proliferated in nutrient-rich patches, and the P heterogeneous environment had the most significant enhancement effect, with dry matter accumulation 70.2%, 7.0%, and 27.0% higher than that in homogeneous and N and K heterogeneous environments, respectively. Homogeneous environments significantly increased the specific root length and root area of the root system; the dry matter mass and morphological structure of the root system of F. hodginsii with a heterospecific neighbor were higher than those under conspecific neighbor and single-plant treatments, and the root area of the root system under the conspecific neighbor treatment was higher than that under the heterospecific neighbor treatment, by 20% and 23%, respectively. Moreover, the root system under heterospecific neighbor treatment had high sensitivity; the heterogeneous nutrient environment increased the mean diameter of the fine roots of the seedlings of F. hodginsii and the diameter of the vascular bundle, and the effect was most significant in the P heterogeneous environment, exceeding that in the N and K heterogeneous environments. The effect was most significant in the P heterogeneous environment, which increased fine root diameter by 20.5% and 10.3%, respectively, compared with the homogeneous environment; in contrast, the fine root vascular ratio was highest in the homogeneous environment, and most of the indicators of the fine root anatomical structure in the nutrient-rich patches were of greater values than those in the nutrient-poor patches in the different heterogeneous environments; competition promoted most of the indicators of the fine root anatomical structure of F. hodginsii seedlings. According a principal component analysis (PCA), the N, Pm and K heterogeneous environments with heterospecific neighbors and the P heterogeneous environment with a conspecific neighbor had higher evaluation in the calculation of eigenvalues of the PCA. Discussion: The root dry matter accumulation, root morphology, and anatomical structure of F. hodginsii seedlings in the heterogeneous nutrient environment were more developed than those in the homogeneous nutrient environment. The effect of the P heterogeneous environment was the most significant. The heterospecific neighbor treatment was more conducive to the expansion and development of root morphology of F. hodginsii seedlings than were the conspecific neighbor and single-plant treatments.

Genome-wide analysis of the DREB family genes and functional identification of the involvement of BrDREB2B in abiotic stress in wucai (Brassica campestris L.).

Dehydration responsive element binding protein (DREB) is a significant transcription factor class known to be implicated in abiotic stresses. In this study, we systematically conducted a genome-wide identification and expression analysis of the DREB gene family, including gene structures, evolutionary relationships, chromosome distribution, conserved domains, and expression patterns. A total of 65 DREB family gene members were identified in Chinese cabbage (Brassica rapa L.) and were classified into five subgroups based on phylogenetic analysis. Through analysis of the conserved domains of BrDREB family genes, only one exon existed in the gene structure. Through the analysis of cis-acting elements, these genes were mainly involved in hormone regulation and adversity stress. In order to identify the function of BrDREB2B, overexpressed transgenic Arabidopsis was constructed. After different stress treatments, the germination rate, root growth, survival rate, and various plant physiological indicators were measured. The results showed that transgenic Arabidopsis thaliana plants overexpressing BrDREB2B exhibited enhanced tolerance to salt, heat and drought stresses. Taken together, our results are the first to report the BrDREB2B gene response to drought and heat stresses in Chinese cabbage and provide a basis for further studies to determine the function of BrDREBs in response to abiotic stresses.

BcWRKY33A Enhances Resistance to Botrytis cinerea via Activating BcMYB51-3 in Non-Heading Chinese Cabbage.

The transcription factor WRKY33 is a vital regulator of the biological process of the necrotrophic fungus Botrytis cinerea (B. cinerea). However, its specific regulatory mechanism remains to be further investigated. In non-heading Chinese cabbage (NHCC, Brassica campestris (syn. Brassica rapa) ssp. Chinensis), our previous study showed that BcWRKY33A is induced not only by salt stress, but also by B. cinerea infection. Here, we noticed that BcWRKY33A is expressed in trichomes and confer plant defense resistance. Disease symptoms and qRT-PCR analyses revealed that BcWRKY33A-overexpressing and -silencing lines were less and more severely impaired, respectively, than wild type upon B. cinerea treatment. Meanwhile, the transcripts' abundance of indolic glucosinolates' (IGSs) biosynthetic genes is consistent with plants' B. cinerea tolerance. Identification and expression pattern analysis of BcMYB51s showed that BcMYB51-3 has a similar trend to BcWRKY33A upon B. cinerea infection. Moreover, BcWRKY33A directly binds to the BcMYB51-3 promoter, which was jointly confirmed by Y1H, dual-LUC, and EMSA assays. The importance of MYB51, the homolog of BcMYB51-3, in the BcWRKY33A-mediated B. cinerea resistance was also verified using the TRV-based VIGS system. Overall, our data concludes that BcWRKY33A directly activates the expression of BcMYB51-3 and downstream IGSs' biosynthetic genes, thereby improving the B. cinerea tolerance of NHCC plants.

Tourists' Safety Perception Clues in the Urban Forest Environment: Visual Quality, Facility Completeness, Accessibility-A Case Study of Urban Forests in Fuzhou, China.

The service quality and safety perception of urban forests are important factors that influence tourists to choose them as recreation destinations. This study aims to propose a theoretical model of multivariate relationships to explore the relationship between service quality (including visual quality, facility completeness, and accessibility) and safety perception to examine whether visual quality, facility completeness, and accessibility on tourists' safety perception in the urban forest and to explain the specific reasons for the impact. We collected sample data from many urban forest green spaces in Fuzhou through a two-stage field survey (N = 891), and controlling for potential confounders, a structural equation model was used to estimate relationships. Safety perception was divided into safety environment perception, control perception, and safety emotion. Visual quality of an urban forest positively affected safety emotion. Traffic accessibility positively affected control perception. Facility completeness had a positive impact on safety emotion and control perception. Both safety emotion and control perception played an important intermediary role in improving the perception of a safe environment in the multivariate model. Visual quality, facility completeness, and accessibility all had a positive impact on tourists' safety perception of urban forests. The findings suggest that improving the service quality of a green space can effectively improve tourists' evaluation of the safety of the urban forest environment. Specifically, tourists' psychological tolerance to threats and their self-confidence in survival can be enhanced by improving the service quality of a green space.

Allele-aware chromosome-scale assembly of the allopolyploid genome of hexaploid Ma bamboo (Dendrocalamus latiflorus Munro).

Dendrocalamus latiflorus Munro is a woody clumping bamboo with rapid shoot growth. Both genetic transformation and clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) gene editing techniques are available for D. latiflorus, enabling reverse genetic approaches. Thus, D. latiflorus has the potential to be a model bamboo species. However, the genome sequence of D. latiflorus has remained unreported due to its polyploidy and large genome size. Here, we sequenced the D. latiflorus genome and assembled it into three allele-aware subgenomes (AABBCC), representing the largest genome of a major bamboo species. We assembled 70 allelic chromosomes (2, 737 Mb) for hexaploid D. latiflorus using both single-molecule sequencing from the Pacific Biosciences (PacBio) Sequel platform and chromosome conformation capture sequencing (Hi-C). Repetitive sequences comprised 52.65% of the D. latiflorus genome. We annotated 135 231 protein-coding genes in the genome based on transcriptomes from eight different tissues. Transcriptome sequencing using RNA-Seq and PacBio single-molecule real-time long-read isoform sequencing revealed highly differential alternative splicing (AS) between non-abortive and abortive shoots, suggesting that AS regulates the abortion rate of bamboo shoots. This high-quality hexaploid genome and comprehensive strand-specific transcriptome datasets for this Poaceae family member will pave the way for bamboo research using D. latiflorus as a model species.

Desymmetrizing Isomerization of Alkene via Thiazolinyl Iminoquinoline Cobalt Catalysis.

We report a cobalt-catalyzed desymmetrizing isomerization of exo-cyclic alkenes to generate chiral 1-methylcyclohexene derivatives with good yields and enantioselectivities. A novel chiral thiazolinyl iminoquinoline ligand and its cobalt complex were designed and synthesized to control the establishment of tertiary or quaternary carbon centers at a remote position. This protocol is operationally simple, and a model for the stereochemical outcome has been proposed.

Encoding histopathology whole slide images with location-aware graphs for diagnostically relevant regions retrieval.

Content-based histopathological image retrieval (CBHIR) has become popular in recent years in histopathological image analysis. CBHIR systems provide auxiliary diagnosis information for pathologists by searching for and returning regions that are contently similar to the region of interest (ROI) from a pre-established database. It is challenging and yet significant in clinical applications to retrieve diagnostically relevant regions from a database consisting of histopathological whole slide images (WSIs). In this paper, we propose a novel framework for regions retrieval from WSI database based on location-aware graphs and deep hash techniques. Compared to the present CBHIR framework, both structural information and global location information of ROIs in the WSI are preserved by graph convolution and self-attention operations, which makes the retrieval framework more sensitive to regions that are similar in tissue distribution. Moreover, benefited from the graph structure, the proposed framework has good scalability for both the size and shape variation of ROIs. It allows the pathologist to define query regions using free curves according to the appearance of tissue. Thirdly, the retrieval is achieved based on the hash technique, which ensures the framework is efficient and adequate for practical large-scale WSI database. The proposed method was evaluated on an in-house endometrium dataset with 2650 WSIs and the public ACDC-LungHP dataset. The experimental results have demonstrated that the proposed method achieved a mean average precision above 0.667 on the endometrium dataset and above 0.869 on the ACDC-LungHP dataset in the task of irregular region retrieval, which are superior to the state-of-the-art methods. The average retrieval time from a database containing 1855 WSIs is 0.752 ms. The source code is available at https://github.com/zhengyushan/lagenet.

Dermoscopic image retrieval based on rotation-invariance deep hashing.

Dermoscopic image retrieval technology can provide dermatologists with valuable information such as similar confirmed skin disease cases and diagnosis reports to assist doctors in their diagnosis. In this study, we design a dermoscopic image retrieval algorithm using convolutional neural networks (CNNs) and hash coding. A hybrid dilated convolution spatial attention module is proposed, which can focus on important information and suppress irrelevant information based on the complex morphological characteristics of dermoscopic images. Furthermore, we also propose a Cauchy rotation invariance loss function in view of the skin lesion target without the main direction. This function constrains CNNs to learn output differences in samples from different angles and to make CNNs obtain a certain rotation invariance. Extensive experiments are conducted on dermoscopic image datasets to verify the effectiveness and versatility of the proposed module, algorithm, and loss function. Experiment results show that the rotation-invariance deep hashing network with the proposed spatial attention module obtains better performance on the task of dermoscopic image retrieval.

Genome-Wide Identification of MIKCc-Type MADS-Box Family Gene and Floral Organ Transcriptome Characterization in Ma Bamboo (Dendrocalamus latiflorus Munro).

Most bamboos die after flowering, and the molecular mechanisms responsible for flowering is poorly understood. The MIKCc-type MADS-box family gene is involved in the flowering process. To explore the mechanism of the MIKCc-type MADS-box gene and phytohormone regulation in the flowering of Dendrocalamus latiflorus Munro (D. latiflorus), characterized by extremely rapid growth and widely cultivated woody bamboo, we initially did a genome-wide analysis of the MIKCc-type MADS-box gene in D. latiflorus. In the meantime, transcriptome analysis was performed using the floral organs. A total of 170 MIKCc-Type MADS-Box genes were identified and divided into 15 categories. The cis-acting element analysis in promoters regions revealed that MIKC-type MADS-box family genes were associated with hormones, including auxin, abscisic acid (ABA), gibberellin (GA) and jasmonic acid (JA), which was found at 79, 476, 96, 486 sites and cover 61, 103, 73, 128 genes. Genome synteny analysis showed subgenome AA and BB were better than CC and obtained 49, 40, 39 synteny genes compared with Oryza sativa (O. sativa). In transcriptome analysis of floral organs, the enriched pathway from DEGs included circadian, vernalization and gibberellin pathways associated with the flowering process. We found that the jasmonic acid synthesis gene is highly expressed in the pistil, which may be the cause of Ma bamboo pollen abortion. The expression profile showed that most MIKC-type MADS-box genes exhibited high expression in flower organs. The consequences of this study will provide insight into the irregular flowering and low pollen counts of Ma bamboo.

Effect of Various Mulch Materials on Chemical Properties of Soil, Leaves and Shoot Characteristics in Dendrocalamus Latiflorus Munro Forests.

The effectiveness of mulch treatments on soil quality as well as on the yield and growth rates of bamboo are major considerations and require further attention. The present work was aimed at assessing the impacts of three different mulch materials on soil available nutrients, biochemical traits, and growth patterns of Dendrocalamus latiflorus Munro. We found that relative to the control (CK), bamboo leaves (MB) and organic fertilizers (MF) treatments significantly (P < 0.05) increased the number of bamboo shoots (47.5 and 22.7%) and yield (21.4 and 9.1%), respectively. We observed that under MB and MF treatments, the concentrations of soil available nutrients (nitrogen, phosphorus, and potassium) increased and played a key role in the differences in chlorophyll, leaf carbohydrate contents (soluble sugar and starch) and were essential to promote bamboo shoot development. Furthermore, we infer from principal component analysis (PCA), that both MB and MF appear to be a better choice than rice husks (MR) to improve nutrient availability, biochemical traits of the leaves, and increased bamboo shoot productivity. Consequently, we suggest using organic fertilizers and bamboo leaves as mulch materials are effective for soil conservation to attain high-quality bamboo production.