Unclear Boundaries Research Articles

Disease Segmentation Using a Robust Method for Unclear Boundaries in Training with Annotations Including Boundary Errors

Disease Segmentation Using a Robust Method for Unclear Boundaries in Training with Annotations Including Boundary Errors

Molecular systematic analysis of the genus Notelaea (Oleaceae): SNPs from high‐density DArT‐sequencing unravel the mystery of the species limits of threatened species of Notelaea

AbstractConservation decision‐making and prioritization of management actions for taxa at high risk of extinction require a clear understanding of systematic relationships and species boundaries. Taxonomic uncertainty surrounds two threatened species of native olive (genus Notelaea) endemic to Australia. Notelaea ipsviciensis is known from only one small population and is listed as critically endangered under Australian and Queensland legislation, while Notelaea lloydii is listed as vulnerable due to its restricted distribution in south‐east Queensland, Australia. Notelaea ipsviciensis occurs in sympatry with N. lloydii, at the only site where it is found, and exhibits intermediate morphology between N. lloydii and Notelaea ovata raising speculation that it is a natural hybrid of the two. To address this uncertainty, we have reconstructed the molecular phylogeny of the genus Notelaea using both single‐gene sequencing and single nucleotide polymorphism (SNP) data. While the chloroplast DNA markers were not found to be informative, the genome‐wide SNPs used in this study have unequivocally resolved the long‐standing question of the hybrid origin of N. ipsviciensis. Results of the phylogenetics and hybrid analyses of SNP data show that N. ipsviciensis is a natural hybrid of N. lloydii and N. ovata, and the taxon's circumscription needs to be re‐evaluated. Our results also revealed unclear species boundaries among numerous other Notelaea species, including the threatened N. lloydii, indicating that further investigation and taxonomic revision may be necessary.

Local Staging Characteristics (T2, T3) of Colorectal Cancer In Multislice Computerized Tomography (MSCT) Imaging

Background: Early determination of the local staging of colorectal cancer is very important to determine the prognostic and optimal management of therapy. The purpose of this study was to determine the local staging characteristics (T2, T3) of colorectal cancer on MSCT imaging. Method: a descriptive design with a retrospective analytical observation approach, namely by evaluating the results of a local staging CT scan (T2, T3) of colorectal cancer. The sample size was 34 cases of colorectal cancer that met the inclusion criteria. Results: 8 cases (23.5%) of colorectal cancer were found with local T2 staging and 26 cases (76.5%) with local T3 staging. The characteristics of local staging T2 are the majority of 100% solid components, 100% clear boundaries, 100% homogeneous enhancement patterns, and 100% not invading surrounding organs. While the characteristics of the local staging of T3 are mostly solid components 61.53%, unclear boundaries 73.07%, irregular margins 96.15%, mass morphology 88.46%, homogeneous enhancement patterns 61.53% and heterogeneous 38.46%, does not invade the surrounding organs 100%. Conclusion: The differences in the characteristics of local T2 and T3 staging lie in the tumor boundaries, tumor shape, and patterns of enhancement.

Co-Saliency Detection of RGBD Image Based on Superpixel and Hypergraph

For the co-saliency detection algorithm of an RGBD image that may have incomplete detection of common salient regions and unclear boundaries, we proposed an improved co-saliency detection method of RGBD images based on superpixels and hypergraphs. First, we optimized the depth map based on edge consistency, and introduced the optimized depth map into the SLIC algorithm to obtain the better superpixel segmentation results of RGBD images. Second, the color features, optimized depth features and global spatial features of superpixels were extracted to construct a weighted hypergraph model to generate saliency maps. Finally, we constructed a weighted hypergraph model for co-saliency detection based on the relationship of color features, global spatial features, optimized depth features and saliency features among images. In addition, in order to verify the impact of the symmetry of the optimized depth information on the co-saliency detection results, we compared the proposed method with two types of models, which included considering depth information and not considering depth information. The experimental results on Cosal150 and Coseg183 datasets showed that our improved algorithm had the advantages of suppressing the background and detecting the integrity of the common salient region, and outperformed other algorithms on the metrics of P-R curve, F-measure and MAE.

Multiresolution Mutual Assistance Network for Cardiac Magnetic Resonance Images Segmentation.

The automatic segmentation of cardiac magnetic resonance (MR) images is the basis for the diagnosis of cardiac-related diseases. However, the segmentation of cardiac MR images is a challenging task due to the inhomogeneity of MR images intensity distribution and the unclear boundaries between adjacent tissues. In this paper, we propose a novel multiresolution mutual assistance network (MMA-Net) for cardiac MR images segmentation. It is mainly composed of multibranch input module, multiresolution mutual assistance module, and multilabel deep supervision. First, the multibranch input module helps the network to extract local and global features more pertinently. Then, the multiresolution mutual assistance module implements multiresolution feature interaction and progressively improves semantic features to more completely express the information of the tissue. Finally, the multilabel deep supervision is proposed to generate the final segmentation map. We compare with state-of-the-art medical image segmentation methods on the medical image computing and computer-assisted intervention (MICCAI) automated cardiac diagnosis challenge datasets and the MICCAI atrial segmentation challenge datasets. The mean dice scores of our method in the left atrium, right ventricle, myocardium, and left ventricle are 0.919, 0.920, 0.881, and 0.960, respectively. The analysis of evaluation indicators and segmentation results shows that our method achieves the best performance in cardiac magnetic resonance images segmentation.

Mapping cashew monocultures in the Western Ghats using optical and radar imagery in Google Earth Engine

Tropical deforestation is increasingly driven by the expansion of agricultural commodity production. Mapping commodity crops is an important step towards monitoring commodity-driven deforestation. Advances in remote sensing technology, such as the availability of high-resolution imagery and the combination of optical and radar imagery have enabled the detection of the tree-like crops which are difficult to distinguish from forest cover. Cashew is an example of a tree-like crop that grows in areas with high forest cover and biodiversity. Cashew is reported to occupy ∼7.1 million ha globally yet mapping it has been constrained by unclear boundaries due to spatial mixing with forests, an indistinct spectral signature, and structural composition that resembles forests. We employed optical, radar, and a combination of the two imagery types to detect and map cashew monocultures in south Maharashtra, India for 2020. We performed a land cover classification on Google Earth Engine using Random Forest, Classification And Regression Trees and Support Vector Machine algorithms. The combination of Sentinel-2 and Sentinel-1 SAR imagery using Random Forest algorithm yielded the highest unbiased overall accuracy (83%) and unbiased producer's and user's accuracies of 71% and 86% respectively for cashew land cover and was considered the best approach. According to our best approach, monoculture cashew plantations occupy 53,350.37 ha of total land area in the Sawantwadi- Dodamarg landscape in India. This study shows that a combination of optical and radar imagery can be used for cashew land cover classification in the Western Ghats, and future studies could modify these methods for cashew mapping in other landscapes. This study contributes to a growing body of literature supporting the use of both optical and radar imagery for detecting tree-like crop cover.

A Land Cover Classification Method for High-Resolution Remote Sensing Images Based on NDVI Deep Learning Fusion Network

High-resolution remote sensing (HRRS) images have few spectra, low interclass separability and large intraclass differences, and there are some problems in land cover classification (LCC) of HRRS images that only rely on spectral information, such as misclassification of small objects and unclear boundaries. Here, we propose a deep learning fusion network that effectively utilizes NDVI, called the Dense-Spectral-Location-NDVI network (DSLN). In DSLN, we first extract spatial location information from NDVI data at the same time as remote sensing image data to enhance the boundary information. Then, the spectral features are put into the encoding-decoding structure to abstract the depth features and restore the spatial information. The NDVI fusion module is used to fuse the NDVI information and depth features to improve the separability of land cover information. Experiments on the GF-1 dataset show that the mean OA (mOA) and the mean value of the Kappa coefficient (mKappa) of the DSLN network model reach 0.8069 and 0.7161, respectively, which have good applicability to temporal and spatial distribution. The comparison of the forest area released by Xuancheng Forestry Bureau and the forest area in Xuancheng produced by the DSLN model shows that the former is consistent with the latter. In conclusion, the DSLN network model is effectively applied in practice and can provide more accurate land cover data for regional ESV analysis.

The observation of anterior segment in children with an R124L mutation corneal dystrophy by anterior segment optical coherence tomography and in vivo confocal microscopy.

To evaluate the anterior segment in children with an R124L mutation corneal dystrophy (CD) using anterior segment optical coherence tomography (AS-OCT) and in vivo confocal microscopy (IVCM). We investigated a family with prevalent CD and an R124L mutation; 59 individuals (14 patients; 6 male and 8 female, aged 2-69 years, 6 children, 2:4 male: female ratio) from four generations were included. We observed corneal lesions through ophthalmologic examinations, AS-OCT, and IVCM. The mean follow-up was 4.60 ± 3.91 years. The mean age for childhood CD onset was 0.90 ± 0.61 years. An Avelino DNA test revealed a heterozygous R124L mutation. Clinical manifestations included recurrent photophobia, tearing, and a foreign body sensation. Recurrence frequency decreased with age. Slit lamp microscopy revealed a rough corneal epithelium. The anterior matrix under the corneal epithelium and the anterior elastic layer were scattered with gray and white opacity. From onset to follow-up, the children's visual acuity decreased from 0.34 ± 0.12 to 0.55 ± 0.17 LogMAR units. AS-OCT showed uneven corneal epithelial thickness. The Bowman's layer was replaced by abnormal substances in the anterior segment. Corneal deposits became increasingly thicker; the average thickness at the last follow-up was 102.78 ± 10.13 μm. IVCM revealed uneven and reflective signals in the corneal upper cortex and subepithelium, with unclear boundaries and a loss of normal cell morphology. We report an early age of onset in a family with prevalent CD due to R124L mutations. AS-OCT is a convenient, quick, and non-contact tool for screening and monitoring the pathological process of CD.

Residual UNet with Dual Attention—An ensemble residual UNet with dual attention for multi‐modal and multi‐class brain MRI segmentation

AbstractIn this paper, we have proposed a variant of UNet for brain magnetic resonance imaging (MRI) segmentation. The proposed model, termed as Residual UNet with Dual Attention (RUDA), addresses the two significant challenges of UNet: extraction of the complex features with unclear boundaries and the problem of over‐segmentation due to the redundancy caused by the skip connection usage. RUDA is constituted upon the residual blocks for extracting the complex structures. It Introduces attention into the skip connections to avoid redundancy and thereby the chance of over‐segmentation. Our model segments brain MRI into white matter (WM), gray matter (GM), and cerebrospinal fluid (CSF) regions, which are considered crucial informative substructures for diagnosing neurological disorders such as Alzheimer's. It has been implemented in an ensemble manner to accommodate the multi‐sequence (T1‐weighted, IR, and T2‐FLAIR) scans. The empirical analysis shows that with an accuracy of 93.80%, RUDA outperforms the two baseline models: UNet (91.37%), ResUNet (91.44%).

A Scoring System That Predicts Difficult Lipoma Resection: Logistic Regression and Tenfold Cross-Validation Analysis.

IntroductionMost lipomas are readily dissected and removed. However, some cases can pose surgical difficulties. This retrospective study sought to identify clinical and radiological risk factors that predict difficult lipoma resection and can be used in a clinically useful scoring system that predicts difficulty preoperatively.MethodsThe study cohort consisted of all consecutive patients who underwent resection of pathology-confirmed lipoma during 2016–2018 at a tertiary care referral center in Tokyo, Japan. Surgical difficulty was defined as difficulty separating some/all of the tumor from the surrounding tissue by hand and inability to extract the tumor in one piece. Descriptive, univariate, and multivariate logistic regression analyses were conducted to identify predictive factors. The predictive accuracy of the scoring system that included these factors was assessed by tenfold cross-validation analysis. Receiver-operating curve (ROC) analysis was conducted to identify the optimal cutoff score for predicting surgical difficulty.ResultsOf the 86 cases, 36% involved surgical difficulty. Multivariate analysis showed that subfascial intramuscular location (odds ratio 42.7, 95% confidence interval 3.0–608.0), broad touching of underlying structures (46.5, 3.7–586.0), in-flowing blood vessels (9.3, 1.1–78.5), and unclear boundaries (109.0, 1.1–1110.0) significantly predicted surgical difficulty. These factors were used to construct a 0–4 point scoring system (with one point per variable). On cross-validation, the accuracy of the scoring system was 82.4% (Cohen’s kappa of 0.57). ROC analysis showed that scores ≥ 2 predicted surgical difficulty with sensitivity and specificity of 55% and 98%, respectively.ConclusionsOur scoring system accurately predicted lipoma resection difficulty and may help operators prepare, thereby facilitating surgery.

THE TEXT AS A REWRITE OF CITATIONS AND TEXT WITHIN TEXT IN THE NOVEL "WHITE PALACES" WRITTEN BY VEHBI KIKAJ

The novel "White Palaces" (“Sarajet e bardha”) by Vehbi Kikaj is one of the well-known works ofcontemporary Albanian children's literature. The fact that the work is dedicated to the child as a reader, underlinesthe importance of perception. The subject of the novel is a family drama which is experienced by the child character.The novel itself as a narrative stands between the author's word that stands out as a separated text. Withoutneglecting the substantive side, this paper aims to present how the phenomenon of putting the text in the text lookspractically. The starting point of the study is that there is no completely pure text since the existence of literatureitself. As the theory of intertextuality states, we consider that each text absorbs other works, is written on them, or isat the crossroads between textual surfaces.The primary, internal texts are in some cases separated from the new(secondary) text, but in other cases have unclear boundaries. In the analysis of this work, we rely on the positions ofwell-known theorists, including Renate Lachmann and Gearad Genette. By revealing the textual interferences, thepaper will reveal the directions of communication that the work pinpoints. These lines have different directionssuch as: folklore, history, even real or vacant texts that the characters talk about. The text that plays the role of theinternal text will be analyzed according to the way/technique of incorporation in the second textual material.Examples will illustrate cases of false citations and those with authorship, some of which are clearly marked. Wefind concrete examples of this in this work in the form of the story within the novel and the dream, which also hasits own internal time and is thus separated from the other text (the novel). Clarifying these distinctions, Barthe'sattitudes also are mentioned. According to him every text is created from preliminary quotations. By summarizingthese features, we aim to underline the importance of this work within contemporary Albanian children's literature.Also, this approach tries to fill the gaps of studies of this nature which the tradition of studying Albanian literaturefor children lacks.

Digital transformation: A review and research agenda

The ongoing and ubiquitous digital transformation challenges the raison d'être of firms and forces managers to rethink business strategies and operations and academics to reconsider related theories. To aid these efforts, we conduct a systematic review of research on firms' digital transformation, generating a database of 537 peer-reviewed academic articles and analyzing it using a novel multi-layered framework. The framework separates three layers: an organization's core activities, its peripheral activities, and its external environment. We find that firms that have come far in their transformations are more embedded in platform ecosystems with unclear business boundaries. Relatedly, we identify a tension between decentralizing versus centralizing power across organizational layers during a firm's digital transformation and how this dynamic affects corporate strategies and firms' internal and external boundaries.

SERNet: Squeeze and Excitation Residual Network for Semantic Segmentation of High-Resolution Remote Sensing Images

The semantic segmentation of high-resolution remote sensing images (HRRSIs) is a basic task for remote sensing image processing and has a wide range of applications. However, the abundant texture information and wide imaging range of HRRSIs lead to the complex distribution of ground objects and unclear boundaries, which bring huge challenges to the segmentation of HRRSIs. To solve this problem, in this paper we propose an improved squeeze and excitation residual network (SERNet), which integrates several squeeze and excitation residual modules (SERMs) and a refine attention module (RAM). The SERM can recalibrate feature responses adaptively by modeling the long-range dependencies in the channel and spatial dimensions, which enables effective information to be transmitted between the shallow and deep layers. The RAM pays attention to global features that are beneficial to segmentation results. Furthermore, the ISPRS datasets were processed to focus on the segmentation of vegetation categories and introduce Digital Surface Model (DSM) images to learn and integrate features to improve the segmentation accuracy of surface vegetation, which has certain prospects in the field of forestry applications. We conduct a set of comparative experiments on ISPRS Vaihingen and Potsdam datasets. The results verify the superior performance of the proposed SERNet.

Diagnostic Value of Color Doppler Ultrasonography in Subacute Thyroiditis.

In order to explore the clinical effect of color Doppler ultrasonography in the diagnosis of subacute thyroiditis, a method for the diagnosis of subacute thyroiditis by color Doppler ultrasonography was proposed. From November 2019 to November 2020, 90 patients with subacute thyroiditis in our hospital were selected as the experimental group; 90 healthy people were selected as the control group during the same period. Both groups were diagnosed by color Doppler ultrasonography and compared. The experimental results showed that patients with subacute thyroiditis showed mild to moderate enlargement of the involved thyroid gland, and local or diffuse inhomogeneous hypoechoic areas may appear in bilateral or unilateral thyroid glands: irregular edges, unclear boundaries, no “ball feel,” mottled changes, and accompanied by tenderness. The blood flow signal around the hypoechoic area is rich, and the internal blood flow signal is less. There was no significant increase in the blood flow velocity of the superior thyroid artery on the affected side. Color Doppler ultrasound not only is simple, economical, and non-invasive but also has a good diagnostic accuracy for subacute thyroiditis, which can provide an important basis for clinical diagnosis and treatment and is worthy of popularization and application.

PENGARUH VERMIKOMPOS DAN KOMPOS KULIT BUAH KAKAO TERHADAP KUALITAS TANAH OXISOL DAN PERTUMBUHAN TANAMAN JAGUNG PULUT (Zea mays ceratina L.)

Oxisol are soils with advanced weathering and have oxic horizon boundaries, very high clay fractions, low cation exchange capacity, and have unclear horizon boundaries. The purpose of this study was to determine the effect of vermicompost and cocoa peel compost on several chemical and biological properties as well as the growth of waxy corn on Oxisol. This study used a randomized block design consisting of four treatments, namely P0: no fertilizer (control), P1: 4% of the weight of the growing media (equivalent to 400 g: 100 g vermicompost + cocoa peel compost 300 g), P2: 8% of the weight of the soil (equivalent to 800 g: 200 g vermicompost + cocoa peel compost 600 g) and P3: 12% of the weight of the planting medium (equivalent to 1200 g: 300 g vermicompost + cocoa peel compost 900 g). Each treatment was repeated four times, so there were 16 experimental units (polybags). The results showed that the use of vermicompost and cocoa peel compost increased soil pH from 6.9 to 7.1; organic-C from 2.63% to 3.62% and total-N from 0.079% to 0.098%, the nematode population varied in each treatment between 14-21 individuals. The tallest plants were recorded by P3 treatment with 12% fertilization by weight of 1200 g growing media after 8 weeks of experimentation.

Power Dynamics within Icelandic Nursing: Walking the Fine Line

Like other Nordic nations, Iceland has a reputation of gender-equality, despite 98% of the country’s nursing profession being women. This paper examines power dynamics within the profession. Fifteen semi-structured interviews with nurses were analyzed with a thematic analysis. Our theoretical framework draws on an ecological perspective highlighting nurses’ vulnerability to power dynamics, and Allen’s work on organizational labor and the invisibility of nurses’ ‘glue work’. The findings reveal that the nurses experience power imbalances when their autonomy is restricted in cooperation with other professionals, demanding their time and disrespecting their professional workspace, and they miss support from their supervisors. They feel their professionalism is belittled, and that the gender imbalance hinders equality. For coping and meeting norms and expectations, the nurses use silencing, which with time pressure and unclear boundaries preserve and enhance stereotypical images. Attracting more male nurses could enhance equality, but additional effort at multiple levels is needed.

Research of Vertical Domain Entity Linking Method Fusing Bert-Binary

To solve the problem of unclear entity boundaries and low recognition accuracy in Chinese text, we construct the crop dataset and propose a Bert-binary-based entity link method. Candidate entity sets are generated through entity matching in multiple data sources. The Bert-binary model is called to calculate the correct probability of the candidate entity, and the entity with the highest score is screened for linking. In comparative experiments with three models on the crop dataset, the F 1 value is increased by 2.5% on the best method or by 8.8% on average. The experimental results show the effectiveness of Bert-binary method in this paper.

Differences between Two Groups of Burmese Vipers (Viperidae: Azemiops) in the Proteomic Profiles, Immunoreactivity and Biochemical Functions of Their Venoms.

Two recently revised Azemiops snakes with apparent differences in their external appearances and skeletal morphologies but unclear genetic boundaries have been proposed. Some researchers have refrained from using the newly proposed taxonomy because these two “species” might be two clades corresponding to different geographical populations of Azemiops feae. To improve the understanding of the kinship of these two Burmese viper groups, more of their characteristics should be explored in depth. We performed a comparative analysis of the proteomic profiles and biochemical activities of snake venoms from these two groups (Sichuan A. feae and Zhejiang A. feae) and evaluated the immunorecognition capacity of commercial antivenoms toward them. Eight protein families were identified in venoms from these two groups, while phospholipase B was only detected in venom from Sichuan A. feae. These protein families displayed varying degrees of differences in relative abundance between venoms, and phospholipase A2 (Sichuan A. feae: 57.15%; Zhejiang A. feae: 65.94%) was the predominated component. Gloydius brevicaudus antivenom exhibited the strongest capacity to immunologically recognize these two venoms, but this was mainly limited to components with high molecular masses, some of which differed between venoms. Additionally, Zhejiang A. feae venom was more toxic than Sichuan A. feae venom, and the venoms expressed remarkable differences in enzymatic activities, probably resulting from the variation in the relative abundance of specific protein families. Our findings unveil differences between the two Burmese viper groups in terms of proteomic profiles, immunoreactivity, and the biochemical functions of their venoms. This information will facilitate the management of snakebites caused by these snakes.

Learning to detect boundary information for brain image segmentation

MRI brain images are always of low contrast, which makes it difficult to identify to which area the information at the boundary of brain images belongs. This can make the extraction of features at the boundary more challenging, since those features can be misleading as they might mix properties of different brain regions. Hence, to alleviate such a problem, image boundary detection plays a vital role in medical image segmentation, and brain segmentation in particular, as unclear boundaries can worsen brain segmentation results. Yet, given the low quality of brain images, boundary detection in the context of brain image segmentation remains challenging. Despite the research invested to improve boundary detection and brain segmentation, these two problems were addressed independently, i.e., little attention was paid to applying boundary detection to brain segmentation tasks. Therefore, in this paper, we propose a boundary detection-based model for brain image segmentation. To this end, we first design a boundary segmentation network for detecting and segmenting images brain tissues. Then, we design a boundary information module (BIM) to distinguish boundaries from the three different brain tissues. After that, we add a boundary attention gate (BAG) to the encoder output layers of our transformer to capture more informative local details. We evaluate our proposed model on two datasets of brain tissue images, including infant and adult brains. The extensive evaluation experiments of our model show better performance (a Dice Coefficient (DC) accuracy of up to 5.3% compared to the state-of-the-art models) in detecting and segmenting brain tissue images.

An Encoder-Decoder Network for Automatic Clinical Target Volume Target Segmentation of Cervical Cancer in CT Images

Cervical cancer is a common gynecological cancer, and its common treatment method radiotherapy depends on target area delineation. The manual delineation work takes a long time and has low accuracy, so automating such delineation is important. At present, some traditional image segmentation algorithms for target area delineation have low accuracy rates. Deep learning algorithms also face some difficulties, such as insufficient data and long training time. As the popular network used in medical image segmentation, U-net still has several disadvantages when handling small targets with unclear boundaries. According to the characteristics of the clinical target volume target segmentation task of cervical cancer, this study modified the U-net structure and optimized the training loss to improve the accuracy of small target detection. The modified structure could handle target boundaries well with operations such as bilinear upsampling. Finally, the proposed algorithm was evaluated on the dataset and compared with several deep learning-based algorithms. Results indicate that the proposed approach has certain superiority.