Forces driving transposable element load variation during Arabidopsis range expansion.

Genetic load refers to the accumulated and potentially life-threatening deleterious mutations in populations. Understanding the mechanisms underlying genetic load variation of transposable element (TE) insertion, a major large-effect mutation, during range expansion is an intriguing question in biology. Here, we used 1,115 global natural accessions of Arabidopsis (Arabidopsis thaliana) to study the driving forces of TE load variation during its range expansion. TE load increased with range expansion, especially in the recently established Yangtze River basin population. Effective population size, which explains 62.0% of the variance in TE load, high transposition rate, and selective sweeps contributed to TE accumulation in the expanded populations. We genetically mapped and identified multiple candidate causal genes and TEs, and revealed the genetic architecture of TE load variation. Overall, this study reveals the variation in TE genetic load during Arabidopsis expansion and highlights the causes of TE load variation from the perspectives of both population genetics and quantitative genetics.

Arabidopsis EXECUTER1 interacts with WRKY transcription factors to mediate plastid-to-nucleus singlet oxygen signaling.

Chloroplasts produce singlet oxygen (1O2), which causes changes in nuclear gene expression through plastid-to-nucleus retrograde signaling to increase plant fitness. However, the identity of this 1O2-triggered pathway remains unclear. Here, we identify mutations in GENOMES UNCOUPLED4 (GUN4) and GUN5 as suppressors of phytochrome-interacting factor1 (pif1) pif3 in regulating the photo-oxidative response in Arabidopsis thaliana. GUN4 and GUN5 specifically interact with EXECUTER1 (EX1) and EX2 in plastids, and this interaction is alleviated by treatment with Rose Bengal (RB) or white light. Impaired expression of GUN4, GUN5, EX1, or EX2 leads to insensitivity to excess light and overexpression of EX1 triggers photo-oxidative responses. Strikingly, upon light irradiation or RB treatment, EX1 transiently accumulates in the nucleus and the nuclear fraction of EX1 shows a similar molecular weight as the plastid-located protein. Point mutagenesis analysis indicated that nuclear localization of EX1 is required for its function. EX1 acts as a transcriptional co-activator and interacts with the transcription factors WRKY18 and WRKY40 to promote the expression of 1O2-responsive genes. This study suggests that EX1 may act in plastid-to-nucleus signaling and establishes a 1O2-triggered retrograde signaling pathway that allows plants adapt to changing light environments during chloroplast development.

Signatures of Adaptation and Purifying Selection in Highland Populations of Dasiphora fruticosa.

High mountains harbor a considerable proportion of biodiversity, but we know little about how diverse plants adapt to the harsh environment. Here we finished a high-quality genome assembly for Dasiphora fruticosa, an ecologically important plant distributed in the Qinghai-Tibetan Plateau and lowland of the Northern Hemisphere, and resequenced 592 natural individuals to address how this horticulture plant adapts to highland. Demographic analysis revealed D. fruticosa underwent a bottleneck after Naynayxungla Glaciation. Selective sweep analysis of two pairs of lowland and highland populations identified 63 shared genes related to cell wall organization or biogenesis, cellular component organization, and dwarfism, suggesting parallel adaptation to highland habitats. Most importantly, we found that stronger purging of estimated genetic load due to inbreeding in highland populations apparently contributed to their adaptation to the highest mountain. Our results revealed how plants could tolerate the extreme plateau, which could provide potential insights for species conservation and crop breeding.

Deep learning-assisted diagnosis of parotid gland tumors by using contrast-enhanced CT imaging.

OBJECTIVES: Imaging interpretation of the benignancy or malignancy of parotid gland tumors (PGTs) is a critical consideration prior to surgery in view of therapeutic and prognostic values of such discrimination. This study investigates the application of a deep learning-based method for preoperative stratification of PGTs. MATERIALS AND METHODS: Using the 3D DenseNet-121 architecture and a dataset consisting of 117 volumetric arterial-phase contrast-enhanced CT scans, we developed a binary classifier for PGT distinction and tested it. We compared the discriminative performance of the model on the test set to that of 12 junior and 12 senior head and neck clinicians. Besides, potential clinical utility of the model was evaluated by measuring changes in unassisted and model-assisted performance of junior clinicians. RESULTS: The model finally reached the sensitivity, specificity, PPV, NPV, F1-score of 0.955 (95% CI 0.751-0.998), 0.667 (95% CI 0.241-0.940), 0.913 (95% CI 0.705-0.985), 0.800 (95% CI 0.299-0.989) and 0.933, respectively, comparable to that of practicing clinicians. Furthermore, there were statistically significant increases in junior clinicians' specificity, PPV, NPV and F1-score in differentiating benign from malignant PGTs when unassisted and model-assisted performance of junior clinicians were compared. CONCLUSION: Our results provide evidence that deep learning-based method may offer assistance for PGT's binary distinction.

Transcriptional repression specifies the central cell for double fertilization.

Double fertilization is a key innovation for the evolutionary success of angiosperms by which the two fertilized female gametes, the egg cell and central cell, generate the embryo and endosperm, respectively. The female gametophyte (embryo sac) enclosed in the sporophyte is derived from a one-celled haploid cell lineage. It undergoes successive events of mitotic divisions, cellularization, and cell specification to give rise to the mature embryo sac, which contains the two female gametes accompanied by two types of accessory cells, namely synergids and antipodals. How the cell fate of the central cell is specified has long been equivocal and is further complicated by the structural diversity of female gametophyte across plant taxa. Here, MADS-box protein AGL80 was verified as a transcriptional repressor that directly suppresses the expression of accessory cell-specific genes to specify the central cell. Further genetic rescue and phylogenetic assay of the AGL80 orthologs revealed a possible conserved mechanism in the Brassicaceae family. Results from this study provide insight into the molecular determination of the second female gamete cell in Brassicaceae.

Scaling-up and proteomic analysis reveals photosynthetic and metabolic insights toward prolonged H2 photoproduction in Chlamydomonas hpm91 mutant lacking proton gradient regulation 5 (PGR5).

Clean and sustainable H2 production is crucial to a carbon-neutral world. H2 generation by Chlamydomonas reinhardtii is an attractive approach for solar-H2 from H2O. However, it is currently not large-scalable because of lacking desirable strains with both optimal H2 productivity and sufficient knowledge of underlying molecular mechanism. We hereby carried out extensive and in-depth investigations of H2 photoproduction of hpm91 mutant lacking PGR5 (Proton Gradient Regulation 5) toward its up-scaling and fundamental mechanism issues. We show that hpm91 is at least 100-fold scalable (up to 10 L) with continuous H2 collection of 7287 ml H2/10L-HPBR in averagely 26 days under sulfur deprivation. Also, we show that hpm91 is robust and active during sustained H2 photoproduction, most likely due to decreased intracellular ROS relative to wild type. Moreover, we obtained quantitative proteomic profiles of wild type and hpm91 at four representing time points of H2 evolution, leading to 2229 and 1350 differentially expressed proteins, respectively. Compared to wild type, major proteome alterations of hpm91 include not only core subunits of photosystems and those related to anti-oxidative responses but also essential proteins in photosynthetic antenna, C/N metabolic balance, and sulfur assimilation toward both cysteine biosynthesis and sulfation of metabolites during sulfur-deprived H2 production. These results reveal not only new insights of cellular and molecular basis of enhanced H2 production in hpm91 but also provide additional candidate gene targets and modules for further genetic modifications and/or in artificial photosynthesis mimics toward basic and applied research aiming at advancing solar-H2 technology.

Adaptation and Phenotypic Diversification in Arabidopsis through Loss-of-Function Mutations in Protein-Coding Genes.

According to the less-is-more hypothesis, gene loss is an engine for evolutionary change. Loss-of-function (LoF) mutations resulting in the natural knockout of protein-coding genes not only provide information about gene function but also play important roles in adaptation and phenotypic diversification. Although the less-is-more hypothesis was proposed two decades ago, it remains to be explored on a large scale. In this study, we identified 60,819 LoF variants in 1071 Arabidopsis (Arabidopsis thaliana) genomes and found that 34% of Arabidopsis protein-coding genes annotated in the Columbia-0 genome do not have any LoF variants. We found that nucleotide diversity, transposable element density, and gene family size are strongly correlated with the presence of LoF variants. Intriguingly, 0.9% of LoF variants with minor allele frequency larger than 0.5% are associated with climate change. In addition, in the Yangtze River basin population, 1% of genes with LoF mutations were under positive selection, providing important insights into the contribution of LoF mutations to adaptation. In particular, our results demonstrate that LoF mutations shape diverse phenotypic traits. Overall, our results highlight the importance of the LoF variants for the adaptation and phenotypic diversification of plants.

Transposable elements drive rapid phenotypic variation in Capsella rubella.

Rapid phenotypic changes in traits of adaptive significance are crucial for organisms to thrive in changing environments. How such phenotypic variation is achieved rapidly, despite limited genetic variation in species that experience a genetic bottleneck is unknown. Capsella rubella, an annual and inbreeding forb (Brassicaceae), is a great system for studying this basic question. Its distribution is wider than those of its congeneric species, despite an extreme genetic bottleneck event that severely diminished its genetic variation. Here, we demonstrate that transposable elements (TEs) are an important source of genetic variation that could account for its high phenotypic diversity. TEs are (i) highly enriched in C. rubella compared with its outcrossing sister species Capsella grandiflora, and (ii) 4.2% of polymorphic TEs in C. rubella are associated with variation in the expression levels of their adjacent genes. Furthermore, we show that frequent TE insertions at FLOWERING LOCUS C (FLC) in natural populations of C. rubella could explain 12.5% of the natural variation in flowering time, a key life history trait correlated with fitness and adaptation. In particular, we show that a recent TE insertion at the 3' UTR of FLC affects mRNA stability, which results in reducing its steady-state expression levels, to promote the onset of flowering. Our results highlight that TE insertions can drive rapid phenotypic variation, which could potentially help with adaptation to changing environments in a species with limited standing genetic variation.

Identification of long noncoding natural antisense transcripts (lncNATs) correlated with drought stress response in wild rice (Oryza nivara).

BACKGROUND: Wild rice, including Oryza nivara and Oryza rufipogon, which are considered as the ancestors of Asian cultivated rice (Oryza sativa), possess high genetic diversity and serve as a crucial resource for breeding novel cultivars of cultivated rice. Although rice domestication related traits, such as seed shattering and plant architecture, have been intensively studied at the phenotypic and genomic levels, further investigation is needed to understand the molecular basis of phenotypic differences between cultivated and wild rice. Drought stress is one of the most severe abiotic stresses affecting rice growth and production. Adaptation to drought stress involves a cascade of genes and regulatory factors that form complex networks. O. nivara inhabits swampy areas with a seasonally dry climate, which is an ideal material to discover drought tolerance alleles. Long noncoding natural antisense transcripts (lncNATs), a class of long noncoding RNAs (lncRNAs), regulate the corresponding sense transcripts and play an important role in plant growth and development. However, the contribution of lncNATs to drought stress response in wild rice remains largely unknown. RESULTS: Here, we conducted strand-specific RNA sequencing (ssRNA-seq) analysis of Nipponbare (O. sativa) and two O. nivara accessions (BJ89 and BJ278) to determine the role of lncNATs in drought stress response in wild rice. A total of 1246 lncRNAs were identified, including 1091 coding-noncoding NAT pairs, of which 50 were expressed only in Nipponbare, and 77 were expressed only in BJ89 and/or BJ278. Of the 1091 coding-noncoding NAT pairs, 240 were differentially expressed between control and drought stress conditions. Among these 240 NAT pairs, 12 were detected only in Nipponbare, and 187 were detected uniquely in O. nivara. Furthermore, 10 of the 240 coding-noncoding NAT pairs were correlated with genes enriched in stress responsive GO terms; among these, nine pairs were uniquely found in O. nivara, and one pair was shared between O. nivara and Nipponbare. CONCLUSION: We identified lncNATs associated with drought stress response in cultivated rice and O. nivara. These results will improve our understanding of the function of lncNATs in drought tolerance and accelerate rice breeding.

Deep Learning for Detecting Cerebral Aneurysms with CT Angiography.

Background Cerebral aneurysm detection is a challenging task. Deep learning may become a supportive tool for more accurate interpretation. Purpose To develop a highly sensitive deep learning-based algorithm that assists in the detection of cerebral aneurysms on CT angiography images. Materials and Methods Head CT angiography images were retrospectively retrieved from two hospital databases acquired across four different scanners between January 2015 and June 2019. The data were divided into training and validation sets; 400 additional independent CT angiograms acquired between July and December 2019 were used for external validation. A deep learning-based algorithm was constructed and assessed. Both internal and external validation were performed. Jackknife alternative free-response receiver operating characteristic analysis was performed. Results A total of 1068 patients (mean age, 57 years ± 11 [standard deviation]; 660 women) were evaluated for a total of 1068 CT angiograms encompassing 1337 cerebral aneurysms. Of these, 534 CT angiograms (688 aneurysms) were assigned to the training set, and the remaining 534 CT angiograms (649 aneurysms) constituted the validation set. The sensitivity of the proposed algorithm for detecting cerebral aneurysms was 97.5% (633 of 649; 95% CI: 96.0, 98.6). Moreover, eight new aneurysms that had been overlooked in the initial reports were detected (1.2%, eight of 649). With the aid of the algorithm, the overall performance of radiologists in terms of area under the weighted alternative free-response receiver operating characteristic curve was higher by 0.01 (95% CI: 0.00, 0.03). Conclusion The proposed deep learning algorithm assisted radiologists in detecting cerebral aneurysms on CT angiography images, resulting in a higher detection rate. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kallmes and Erickson in this issue.

Low Expression of Long Noncoding RNA SLC26A4 Antisense RNA 1 Is an Independent Prognostic Biomarker and Correlate of Immune Infiltrates in Breast Cancer.

BACKGROUND Aberrant expression of long noncoding RNA (lncRNA) SLC26A4 antisense RNA 1 (SLC26A4-AS1) plays an important role in some cancer types. However, the clinical significance of SLC26A4-AS1 in patients with breast cancer (BC) and the possible regulatory mechanisms of SLC26A4-AS1 are unclear. MATERIAL AND METHODS Statistical analysis was used to assess the correlation between SLC26A4-AS1 expression and patients' clinical characteristics. The Kaplan-Meier method and Cox regression analysis were used to assess the correlation between SLC26A4-AS1 expression and prognosis. Gene set enrichment analysis (GSEA) and immuno-infiltration analysis were used to investigate the possible regulatory mechanisms of SLC26A4-AS1. RESULTS Low SLC26A4-AS1 expression in BC was associated with age (P<0.001), estrogen-receptor status (P<0.001), PAM50 (P<0.001), and menopause status (P<0.001). Low SLC26A4-AS1 expression predicted a poorer overall survival (OS) (hazard ratio [HR]: 0.56; 95% confidence interval [CI]: 0.40-0.78; P=0.001) and disease-specific survival (DSS) (HR: 0.57; 95% CI: 0.37-0.88; P=0.011). Also, SLC26A4-AS1 expression (HR: 0.298; 95% CI: 0.154-0.579; P<0.001) was independently correlated with OS in patients with BC. SLC26A4-AS1 was related to CYP2E1 reactions, protein export, mitochondrial_ciii_assembly, formation of adenosine triphosphate by chemiosmotic coupling, budding and maturation of HIV virion, cristae formation, biocarta proteasome pathway, endosomal sorting complex required for transport, and histone modification. SLC26A4-AS1 expression was associated with some types of immune infiltrating cells. CONCLUSIONS SLC26A4-AS1 expression was significantly associated with poor survival and immune infiltration in patients with BC. It may be a promising prognostic biomarker for BC.

Rice FLUORESCENT1 Is Involved in the Regulation of Chlorophyll.

Chlorophyll biosynthesis plays essential roles in photosynthesis and plant growth in response to environmental conditions. The accumulation of excess chlorophyll biosynthesis intermediates under light results in the production of reactive oxygen species and oxidative stress. In this study, we identified a rice (Oryza sativa) mutant, oxidation under photoperiod (oxp), that displayed photobleached lesions on its leaves, reduced growth and decreased chlorophyll content during light/dark cycles or following a dark-to-light transition. The oxp mutant accumulated more chlorophyll precursors (5-aminolevulinic acid and protochlorophyllide) than the wild type in the dark, and more singlet oxygen following light exposure. Several singlet-oxygen-responsive genes were greatly upregulated in oxp, whereas the expression patterns of OsPORA and OsPORB, two genes encoding the chlorophyll biosynthesis enzyme NADPH:protochlorop hyllide oxidoreductase, were altered in de-etiolated oxp seedlings. Molecular and complementation studies revealed that oxp is a loss-of-function mutant in LOC_Os01g32730, a homolog of FLUORESCENT (FLU) in Arabidopsis thaliana. Rice PHYTOCHROME-INTERACTING FACTOR-LIKE14 (OsPIL14) transcription factor directly bound to the OsFLU1 promoter and activated its expression. Dark-grown transgenic rice seedlings overexpressing OsPIL14 accumulated more chlorophyll and turned green faster than the wild type upon light illumination. Thus, OsFLU1 is an important regulator of chlorophyll biosynthesis in rice.

Comprehensive transcriptome profiling of soybean leaves in response to simulated acid rain.

As a source of edible oil and protein, soybean is a major globally important economic crop; Improving its production has been an important objective of soybean breeding. Acid rain has been shown to influence soybean growth and productivity, with consequent adverse impacts on its production for use by human populations. In this study, RNA sequencing technology was utilized to examine changes in gene expression when soybean was exposed to simulated acid rain (SAR). We sampled soybean leaves at five time intervals (0, 6, 30, 54, 78, and 102 h), and built the cDNA library. In total, 54,175 expression genes were found, including 2016 genes with differential expression. A total of 416 genes were considered, as they were closely related to the response to SAR. Genes related to the regulation of sulfur and nitrogen metabolism, carbohydrate metabolism, photosynthesis, and reactive oxygen species were among those differentially expressed in response to SAR. In this study, we examined the response mechanisms of soybean under SAR exposure. Our findings will improve our understanding of the molecular mechanisms employed by soybean in responding to abiotic stress, and therefore provides important information in developing soybean breeding to improve tolerance to these stresses.

[Clinical effects of pedicled omentum covering and wrapping the ureteral anastomosis to prevent ureteral anastomotic leakage after surgery of abdominal and pelvic tumors].

OBJECTIVE: To explore the clinical effects of pedicled omentum covering and wrapping the ureteral anastomosis to prevent ureteral anastomotic leakage after surgery of abdominal and pelvic tumors. METHODS: Clinical data of 64 patients with ureteral anastomosis after surgery of abdominal and pelvic tumors treated in our department from May 2005 to May 2012 were retrospectively analyzed. They were assigned into 2 groups. There were 23 patients of ureteral anastomosis combined with pedicled omentum surrounding and wrapping the anastomotic site (optimization group), and 41 cases of ureteral anastomosis alone (control group). The clinical data of all the 64 patients were reviewed and the therapeutic effects of the two treatment approaches were compared. RESULTS: At one week after the operation, there were 8 cases (34.8%, 8/23) with ureteral anastomotic fistula in the optimization group and 31 cases (75.6%, 31/41) in the control group (P = 0.010). In the postoperative days 1-3, the average drainage everyday from abdominal tube around the anastomotic site was 260.4 ml and 320.8 ml, respectively (P = 0.446). The average drainage volume everyday was 80.5 ml and 160.5 ml from the postoperative day 4 to day 7 (P = 0.015). The average time of removal of the peritoneal cavity drainage tube was 18.5 d in the optimization group and 32.6 d postoperatively in the control group (P = 0.015). CONCLUSIONS: Covering and wrapping the ureteral anastomosis with pedicled omentum can promote the rapid adhesion of surrounding tissues to reduce urine leakage and postoperative complications, and shorten the surgical treatment cycle.

WASPSYN: A Challenge for Domain Adaptive Synapse Detection in Microwasp Brain Connectomes.

The size of image volumes in connectomics studies now reaches terabyte and often petabyte scales with a great diversity of appearance due to different sample preparation procedures. However, manual annotation of neuronal structures (e.g., synapses) in these huge image volumes is time-consuming, leading to limited labeled training data often smaller than 0.001% of the large-scale image volumes in application. Methods that can utilize in-domain labeled data and generalize to out-of-domain unlabeled data are in urgent need. Although many domain adaptation approaches are proposed to address such issues in the natural image domain, few of them have been evaluated on connectomics data due to a lack of domain adaptation benchmarks. Therefore, to enable developments of domain adaptive synapse detection methods for large-scale connectomics applications, we annotated 14 image volumes from a biologically diverse set of Megaphragma viggianii brain regions originating from three different whole-brain datasets and organized the WASPSYN challenge at ISBI 2023. The annotations include coordinates of pre-synapses and post-synapses in the 3D space, together with their one-to-many connectivity information. This paper describes the dataset, the tasks, the proposed baseline, the evaluation method, and the results of the challenge. Limitations of the challenge and the impact on neuroscience research are also discussed. The challenge is and will continue to be available at https://codalab.lisn.upsaclay.fr/competitions/9169. Successful algorithms that emerge from our challenge may potentially revolutionize real-world connectomics research and further the cause that aims to unravel the complexity of brain structure and function.

[Common types of massive intraoperative haemorrhage, treatment philosophy and operating skills in pelvic cancer surgery].

OBJECTIVE: To explore the common types of massive intraoperative bleeding, clinical characteristics, treatment philosophy and operating skills in pelvic cancer surgery. METHODS: We treated massive intraoperative bleeding in 19 patients with pelvic cancer in our department from January 2003 to March 2012. Their clinical data were retrospectively analyzed. The clinical features of massive intraoperative bleeding were analyzed, the treatment experience and lessons were summed up, and the operating skills to manage this serious issue were analyzed. RESULTS: In this group of 19 patients, 7 cases were of presacral venous plexus bleeding, 5 cases of internal iliac vein bleeding, 6 cases of anterior sacral venous plexus and internal iliac vein bleeding, and one cases of internal and external iliac vein bleeding. Six cases of anterior sacral plexus bleeding and 4 cases of internal iliac vein bleeding were treated with suture ligation to stop the bleeding. Six cases of anterior sacral and internal iliac vein bleeding, one cases of anterior sacral vein bleeding, and one case of internal iliac vein bleeding were managed with transabdominal perineal incision or transabdominal cotton pad compression hemostasis. One case of internal and external iliac vein bleeding was treated with direct ligation of the external iliac vein and compression hemostasis of the internal iliac vein. Among the 19 patients, 18 cases had effective hemostasis. Their blood loss was 400-1500 ml, and they had a fair postoperative recovery. One patient died due to massive intraoperative bleeding of ca. 4500 ml. CONCLUSIONS: Most of the massive intraoperative bleeding during pelvic cancer surgery is from the presacral venous plexus and internal iliac vein. The operator should go along with the treatment philosophy to save the life of the patient above all, and to properly perform suture ligation or compression hemostasis according to the actual situation, and with mastered crucial operating hemostatic skills.

Molecular mechanisms of adaptive evolution in wild animals and plants.

Wild animals and plants have developed a variety of adaptive traits driven by adaptive evolution, an important strategy for species survival and persistence. Uncovering the molecular mechanisms of adaptive evolution is the key to understanding species diversification, phenotypic convergence, and inter-species interaction. As the genome sequences of more and more non-model organisms are becoming available, the focus of studies on molecular mechanisms of adaptive evolution has shifted from the candidate gene method to genetic mapping based on genome-wide scanning. In this study, we reviewed the latest research advances in wild animals and plants, focusing on adaptive traits, convergent evolution, and coevolution. Firstly, we focused on the adaptive evolution of morphological, behavioral, and physiological traits. Secondly, we reviewed the phenotypic convergences of life history traits and responding to environmental pressures, and the underlying molecular convergence mechanisms. Thirdly, we summarized the advances of coevolution, including the four main types: mutualism, parasitism, predation and competition. Overall, these latest advances greatly increase our understanding of the underlying molecular mechanisms for diverse adaptive traits and species interaction, demonstrating that the development of evolutionary biology has been greatly accelerated by multi-omics technologies. Finally, we highlighted the emerging trends and future prospects around the above three aspects of adaptive evolution.

Local Intensity Order Transformation for Robust Curvilinear Object Segmentation.

Segmentation of curvilinear structures is important in many applications, such as retinal blood vessel segmentation for early detection of vessel diseases and pavement crack segmentation for road condition evaluation and maintenance. Currently, deep learning-based methods have achieved impressive performance on these tasks. Yet, most of them mainly focus on finding powerful deep architectures but ignore capturing the inherent curvilinear structure feature (e.g., the curvilinear structure is darker than the context) for a more robust representation. In consequence, the performance usually drops a lot on cross-datasets, which poses great challenges in practice. In this paper, we aim to improve the generalizability by introducing a novel local intensity order transformation (LIOT). Specifically, we transfer a gray-scale image into a contrast-invariant four-channel image based on the intensity order between each pixel and its nearby pixels along with the four (horizontal and vertical) directions. This results in a representation that preserves the inherent characteristic of the curvilinear structure while being robust to contrast changes. Cross-dataset evaluation on three retinal blood vessel segmentation datasets demonstrates that LIOT improves the generalizability of some state-of-the-art methods. Additionally, the cross-dataset evaluation between retinal blood vessel segmentation and pavement crack segmentation shows that LIOT is able to preserve the inherent characteristic of curvilinear structure with large appearance gaps. An implementation of the proposed method is available at https://github.com/TY-Shi/LIOT.

Cell Localization and Counting Using Direction Field Map.

Automatic cell counting in pathology images is challenging due to blurred boundaries, low-contrast, and overlapping between cells. In this paper, we train a convolutional neural network (CNN) to predict a two-dimensional direction field map and then use it to localize cell individuals for counting. Specifically, we define a direction field on each pixel in the cell regions (obtained by dilating the original annotation in terms of cell centers) as a two-dimensional unit vector pointing from the pixel to its corresponding cell center. Direction field for adjacent pixels in different cells have opposite directions departing from each other, while those in the same cell region have directions pointing to the same center. Such unique property is used to partition overlapped cells for localization and counting. To deal with those blurred boundaries or low contrast cells, we set the direction field of the background pixels to be zeros in the ground-truth generation. Thus, adjacent pixels belonging to cells and background will have an obvious difference in the predicted direction field. To further deal with cells of varying density and overlapping issues, we adopt geometry adaptive (varying) radius for cells of different densities in the generation of ground-truth direction field map, which guides the CNN model to separate cells of different densities and overlapping cells. Extensive experimental results on three widely used datasets (i.e., VGG Cell, CRCHistoPhenotype2016, and MBM datasets) demonstrate the effectiveness of the proposed approach.

Polishing Performance and Removal Mechanism of Core-Shell Structured Diamond/SiO2 Abrasives on Sapphire Wafer.

Corrosive and toxic solutions are normally employed to polish sapphire wafers, which easily cause environmental pollution. Applying green polishing techniques to obtain an ultrasmooth sapphire surface that is scratch-free and has low damage at high polishing efficiency is a great challenge. In this paper, novel diamond/SiO2 composite abrasives were successfully synthesized by a simplified sol-gel strategy. The prepared composite abrasives were used in the semi-fixed polishing technology of sapphire wafers, where the polishing slurry contains only deionized water and no other chemicals during the whole polishing process, effectively avoiding environmental pollution. The experimental results showed that diamond/SiO2 composite abrasives exhibited excellent polishing performance, along with a 27.2% decrease in surface roughness, and the material removal rate was increased by more than 8.8% compared with pure diamond. Furthermore, through characterizations of polished sapphire surfaces and wear debris, the chemical action mechanism of composite abrasives was investigated, which confirmed the solid-state reaction between the SiO2 shell and the sapphire surface. Finally, applying the elastic-plastic contact model revealed that the reduction of indentation depth and the synergistic effect of chemical corrosion and mechanical removal are the keys to improving polishing performance.