Weak Boundaries Research Articles

Reaching for a hybrid and networked university through lifelong learning initiatives

Working life is transforming, including an emerging digitalisation of its products and processes. Due to such transformation, competencies need to be developed in organisations that suit the emerging conditions for performing work. In the growth of these competencies, higher education is a key player. Its role is twofold, including preparing students for working life and being involved in professional development through continuous education and lifelong learning initiatives. This role as a key player has the potential to assist in the development of competencies at both an individual and organisational level. Another movement in society is the dissolution of boundaries between organisations, leading to the emergence of networked and hybrid organisations. Digitalisation is a driving force in this phenomenon. Even within universities, this trend has grown and impacted the relationship between the university and the surrounding society. In a hybrid and networked university, its operations are closely linked to the surrounding society, including the interplay of the needs of both parties; a balance of power; and weak boundaries between the university and organisations in the surrounding society. Another trend in the higher education sector is the emphasis on lifelong learning policies. Such policies link lifelong learning to global competition and social inclusion for all. The project reported in this paper builds on the ideas of lifelong learning from a university that aims to be hybrid and networked. The reported initiative builds on data from an early phase of a development project that aims at developing new forms of lifelong learning and professional development. Thematically analysed Post-it Notes from workshops supported the answering of the following research question: How do higher education staff perceive development of lifelong learning? The following four main categories were generated: (1) the hybrid and networked university, (2) pedagogical models, (3) internal organisation, and (4) quality. The conclusion includes staffs’ positive attitude towards lifelong learning initiatives. They emphasise the university as an open space and as a partner for agents in the surrounding society. Moreover, they stress lifelong learning initiatives for staff; further development of educational models; scientific legitimacy and high-quality deliverance; flexible internal organisation and structures; and finally, the importance of collaboration and networking. Further data collection is needed to validate the results and the potential for developing theoretical insights that could inform lifelong learning initiatives in emerging digitally infused societies.

Comparative study on mechanism of hydrogen embrittlement of Fe–18Mn-0.6C twinning-induced plasticity (TWIP) steel subjected to friction-stir welding (FSW) and tungsten inert-gas (TIG) welding

In present study, we investigated the effect of welding technique on resistance of hydrogen (H) embrittlement in Fe–18Mn-0.6C (wt.%) twinning-induced plasticity (TWIP) steel by using an electrochemical H-charging, thermal desorption spectroscope, and electron microscope. The friction-stir welding (FSW) specimen was less sensitive to H embrittlement relative to base metal and tungsten inert-gas (TIG) welding specimens by differences in microstructure and deformation mechanism. During H-charging of the FSW specimen, numerous dislocations/Σ3 boundaries reduced H diffusion into specimen interior, causing shallowest depth of brittle fracture. In contrast, the depth of brittle fracture in the H-charged TIG specimen was much larger due to rapid H diffusion by decreased grain boundaries including Σ3 annealing twin boundaries. During tensile deformation, the H-charged FSW specimen underwent the reduction in stress concentration by inactive TWIP as well as strong resistance of boundary decohesion. It was because of alleviation of H-enhanced localized plasticity (HELP) mechanism, leading to suppression of H-induced crack growth. Conversely, the H-charged base metal and TIG specimens revealed large stress concentration by active TWIP and weak boundaries owing to strong effects of HELP + H-enhanced decohesion (HEDE) mechanisms, exhibiting rapid H-induced crack propagation.

Phylogenetic Diversity, Host Specificity, and Distribution of the Wood-Decaying Fungus Phellinotus teixeirae in Western Colombia’s Seasonally Dry Tropical Forest

Phellinotus (Polyporales) is a common genus of wood-decay fungi in tropical and subtropical areas, endemic to the Seasonally Dry Tropical Forest (SDTF) biome. However, Phellinotus diversity remains unexplored, despite being a major threat to living trees. Therefore, this study is aimed at confirming and characterizing through morphological and molecular data the first isolates of Phellinotus teixeirae in Pithecellobium dulce (Fabaceae) trees (locally referred to as ‘Chiminango’) from the endangered Colombian SDTF biome. Fifteen fungal specimens were recovered from living P. dulce trees, in the urban area and at the Universidad del Valle campus, and classified as P. teixeirae based on taxonomical descriptors. Phylogenetic relationships were inferred from a four-loci dataset (ribosomal and gene-coding regions), including 82 taxa covering 3991 nucleotide positions. The analysis recovered seven highly supported (>90% bootstrapping) monophyletic taxa of the ‘Phellinotus Clade’, and confirmed the new distribution range of P. teixeirae (100% bootstrap support), which extends approx. 1000 km north in the Neotropics. Hierarchical stratified Analysis of MOlecular VAriance (AMOVA) provided a clear genetic distinction between species (70% of variation, p-value = 0.001) and low differentiation among country of origin within species (11%, p-value = 0.044). Discriminant Analysis for Principal Components (DAPC) indicated complex clustering including closely related species, probably a signal of recent radiation and weak species boundaries. Median-joining haplotype network analysis identified unique haplotypes, which may correlate with new host colonization and population expansion (Tajima’s D ≤ −0.5). In conclusion, this study provides the first assessment of the genetic diversity of P. teixeirae in a novel geography (SDTP) and host tree (P. dulce). However, increasing the number of isolates remains critical to understand further the genus’ distribution patterns and drivers of genetic diversity.

Fragmentation behavior of diamond particles with different particle size and ratio under high pressure compaction

The fragmentation behavior of mixed diamond particles with different grain sizes under high pressure is important to guide the design of raw material diamond powder ratios and obtain high performance polycrystalline diamond compact (PDC). In this study, five different grain size diamond powders commonly used for the preparation of PDCs were selected, and the single and mixed diamond particles were cold pressed at 7.7 GPa for 60s. The particle size variations of diamond powders with different principal particle sizes and ratios after cold pressing were analyzed by a laser particle size scanner, and the crushing behavior was observed by scanning electron microscopy. The results showed that the larger the initial grain size, the more significant the degree of fragmentation under high pressure. The crushing morphology of fine-grained diamond showed that the majority of grains were locally fractured at the edge, and the peak of the grain size distribution curve was shifted to the left after high pressure. In comparison, the coarse-grained diamond produced a large number of cracks and fractures along the weak grain boundaries under high pressure, which generated a large number of fine grains, and multiple secondary peaks in the particle size distribution curve. The addition of fine-grained diamond as filler particles to coarse-grained diamond could inhibit the crushing phenomenon to some extent. As the proportion of filler particles increased, the crushing of coarse-grained diamond was inhibited more significantly. When the proportion of filler particles was 30 wt%, the crushing under high pressure was greatly limited, especially when the size difference between filler particles and main particles was large. The results of the cold pressing experiments provided preliminary evidence that 30 wt% of filler particles were too high, and that a better ratio should be between 10 wt% and 20 wt%.

Pupils’ right to inclusion in an urban Finnish lower secondary school: a Bernsteinian analysis of pupils’ social relations

Finnish schools are famous for their egalitarian principles, but they face challenges related to pupils’ equal opportunities and, more broadly, democratic schooling. In this article, we examine the lived consequences of a growing challenge, school segregation, using Basil Bernstein’s concepts of inclusion and classification. Our analysis is based on ethnographic data produced in a lower secondary school located in a relatively disadvantaged residential area in the metropolitan area of Helsinki, Finland. First, we explore how the classification of pupils from different social class and ethnic backgrounds is present in the school. Second, we analyse how the strength of this classification affects pupils’ right to inclusion. We conclude that weak boundaries between pupils, and thus weak classification, construct an inclusive school community. However, the different categories—social class, ethnicity or other—intersect, and the strength of their classification varies, which in turn affects pupils’ peer relations and feelings of belonging.

New improved model for joint segmentation and registration of multi-modality images: application to medical images

Joint segmentation and registration of images is a focused area of research nowadays. Jointly segmenting and registering noisy images and images having weak boundaries/intensity inhomogeneity is a challenging task. In medical image processing, joint segmentation and registration are essential methods that aid in distinguishing structures and aligning images for precise diagnosis and therapy. However, these methods encounter challenges, such as computational complexity and sensitivity to variations in image quality, which may reduce their effectiveness in real-world applications. Another major issue is still attaining effective joint segmentation and registration in the presence of artifacts or anatomical deformations. In this paper, a new nonparametric joint model is proposed for the segmentation and registration of multi-modality images having weak boundaries/noise. For segmentation purposes, the model will be utilizing local binary fitting data term and for registration, it is utilizing conditional mutual information. For regularization of the model, we are using linear curvature. The new proposed model is more efficient to segmenting and registering multi-modality images having intensity inhomogeneity, noise and/or weak boundaries. The proposed model is also tested on the images obtained from the freely available CHOAS dataset and compare the results of the proposed model with the other existing models using statistical measures such as the Jaccard similarity index, relative reduction, Dice similarity coefficient and Hausdorff distance. It can be seen that the proposed model outperforms the other existing models in terms of quantitatively and qualitatively.

Cross co-teaching for semi-supervised medical image segmentation

Excellent performance has been achieved on semi-supervised medical image segmentation, but existing algorithms perform relatively poorly for objects with variable topologies and weak boundaries. In this paper, we propose a novel cross co-teaching framework, called Cross-structure-task Collaborative Teaching (CroCT), which not only can effectively handle variable topologies, but also strengthens the learning for weak boundaries of unlabeled data. Specifically, a new cross-structure-task collaborative teaching mechanism is developed based on our designed “E-Net” structure composed of a shared encoder and two decoder branches with distinct learning paradigms, which asks these two branches to regress topology-aware signed distance functions and densely-predicted segmentation masks for each other. Powered by the collaboration across different structural biases and sequence-related tasks, our CroCT can extract more discriminative yet complementary representations from abundant raw medical data to promote the consistency learning generalization, further boosting the performance for tackling highly diverse shapes and topological changes intra-/inter-slices. Besides, it guarantees the diversities from multi-levels, i.e., structure and task perspectives, to preclude prediction uncertainty. In addition, a novel adaptive boundary enhancing (ABE) module is proposed to introduce compact annularly enhanced boundary features into semi-supervised training, which significantly improves weak boundary perception ability for unlabeled data while facilitating collaborative teaching for efficiently propagating complementary knowledge across different branches. The extensive experiments on three challenging medical benchmarks, employing different labeled settings, demonstrate the superiority of our CroCT over recent state-of-the-art competitors.

A generalized framework for agricultural field delineation from high-resolution satellite imageries

ABSTRACT Accurate digital data on agricultural fields are crucial for various agriculture applications. While deep learning methods have shown promise in delineating fields from high-resolution imagery, there is a lack of research evaluating key techniques for field boundary detection. Challenges also exist in converting detection results into high-quality fields. This study addresses these issues and proposes a generalized framework for agricultural field delineation (GF-AFD). First, we identify three key techniques for field boundary detection and apply them to MPSPNet model. Ablation and comparison experiments demonstrate significant performance enhancements due to techniques. They also prove effective for DeeplabV3+ model, which shares a similar architecture. The modified models outperform U-Net-based models and approach the state-of-the-art models. Second, we show that performing region segmentation on boundary results yields improved field shapes. To address weak boundary loss and unstable parameters issues in existing segmentation-based methods, we introduce the OWT method to enhance weak boundaries directionally before segmentation. We also develop a hierarchical merging method, leveraging the observational hierarchy of fields, resulting in stable parameters across regions and models. The proposed GF-AFD framework was validated cross three diverse Chinese counties. The results demonstrate the framework's robust performance, providing a valuable solution for delineating agricultural fields.

Design and Internal Validation of an Educational-Therapeutic Model for Addressing Co-Dependency Based on Perceived Lived Experiences of Adolescents

Purpose: Adolescence is the third developmental stage, beginning after the completion of childhood. This stage, influenced by developmental changes, exposes individuals to a higher risk of developmental dysfunctions and psychosocial harm compared to other life stages. The present study aimed to design and internally validate an educational-therapeutic model for addressing co-dependency based on the perceived lived experiences of co-dependent adolescents in the Khattam region. Methods and Materials: A sequential exploratory mixed-methods approach was employed. The qualitative part utilized inductive content analysis, while the quantitative part gathered opinions from professors and experts through a survey. In the qualitative section, the statistical sample for semi-structured interviews included 15 co-dependent adolescents from the Khattam region. In the quantitative section, 12 professors and specialists were selected as the sample. Based on the content analysis, 5 main themes and 12 sub-themes were extracted. Findings: The qualitative findings indicated that the themes and sub-themes were divided into two dimensions, individual and social. The individual dimension included reactivity with sub-components (denial, obsession, and anger), the need for control with sub-components (suppression of emotions and difficulty in establishing relationships), and alienation. The social dimension included interpersonal dependency with sub-components (controlling, distrust, abnormal relationships, weak boundaries), and external focus with sub-components (caring for others and being influenced by external factors). Conclusion: It can be concluded that educational system stakeholders should consider the proposed model of this research for designing and implementing effective education aimed at enhancing the self-efficacy of adolescents.

Multiscale Sea-Land Segmentation Networks for Weak Boundaries

Multiscale Sea-Land Segmentation Networks for Weak Boundaries

Selective image segmentation driven by region, edge and saliency functions

Present active contour methods often struggle with the segmentation of regions displaying variations in texture, color, or intensity a phenomenon referred to as inhomogeneities. These limitation impairs their ability to precisely distinguish and outline diverse components within an image. Further some of these methods employ intricate mathematical formulations for energy minimization. Such complexity introduces computational sluggishness, making these methods unsuitable for tasks requiring real-time processing or rapid segmentation. Moreover, these methods are susceptible to being trapped in energy configurations corresponding to local minimum points. Consequently, the segmentation process fails to converge to the desired outcome. Additionally, the efficacy of these methods diminishes when confronted with regions exhibiting weak or subtle boundaries. To address these limitations comprehensively, our proposed approach introduces a fresh paradigm for image segmentation through the synchronization of region-based, edge-based, and saliency-based segmentation techniques. Initially, we adapt an intensity edge term based on the zero crossing feature detector (ZCD), which is used to highlight significant edges of an image. Secondly, a saliency function is formulated to detect salient regions from an image. We have also included a globally tuned region based SPF (signed pressure force) term to move contour away and capture homogeneous regions. ZCD, saliency and global SPF are jointly incorporated with some scaled value for the level set evolution to develop an effective image segmentation model. In addition, proposed method is capable to perform selective object segmentation, which enables us to choose any single or multiple objects inside an image. Saliency function and ZCD detector are considered feature enhancement tools, which are used to get important features of an image, so this method has a solid capacity to segment nature images (homogeneous or inhomogeneous) precisely. Finally, the adaption of the Gaussian kernel removes the need of any penalization term for level set reinitialization. Experimental results will exhibit the efficiency of the proposed method.

Weakening patterns of intervocalic voiced plosives in Japanese

Japanese voiced plosives have been observed to exhibit weakened burst intensity, and in some instances, they may show a complete weakening with no observable burst. Maekawa (2010) investigated the weakening rate of voiced plosives in Japanese and its relationship with surrounding acoustic environments. His study revealed a notable tendency for consonant weakening, particularly at positions characterized by weak prosodic boundaries. However, despite referring to a general phenomenon of consonant weakening, there exist various pronunciation patterns. The articulatory gestures responsible for generating these diverse weakening patterns have not yet been thoroughly investigated. In this research, we conducted recording experiments involving native Japanese speakers, simultaneously capturing tongue movements using an ultrasound device. Our focus was directed towards the patterns of weakening observed in intervocalic Japanese voiced plosives and how to classify them. We identified three major patterns of plosives consonant weakening: (1) Voiced glide pattern, (2) Voiced frication pattern, and (3) Pattern with halted vocal fold vibration and weak aspiration. Variation was found in the frequency of occurrence of these patterns among alveolar, velar, and bilabial plosives. It is important to note that the rate of specific weakening patterns at different articulatory points varies among individuals. Additionally, for alveolar and velar sounds, tongue movement was analyzed using ultrasound imaging.

High‐Resolution Mantle Flow Models Reveal Importance of Plate Boundary Geometry and Slab Pull Forces on Generating Tectonic Plate Motions

AbstractMantle convection models based on geophysical constraints have provided us with a basic understanding of the forces driving and resisting plate motions on Earth. However, existing studies computing the balance of underlying forces are contradicting, and the impact of plate boundary geometry on surface deformation remains unknown. We address these issues by developing global instantaneous 3‐D mantle convection models with a heterogeneous density and viscosity distribution and weak plate boundaries prescribed using different geometries. We find that the plate boundary geometry of the Global Earthquake Model (GEM, Pagani et al., 2018, https://doi.org/10.1177/8755293020931866), featuring open plate boundaries with discrete lithospheric‐depth weak zones in the oceans and distributed crustal faults within continents, achieves the best fit to the observed GPS data with a directional correlation of 95.1% and a global point‐wise velocity residual of 1.87 cm/year. A good fit also requires plate boundaries being 3 to 4 orders of magnitude weaker than the surrounding lithosphere and low asthenospheric viscosities between 5 × 1017 and 5 × 1018 Pa s. Models without asthenospheric and lower mantle heterogeneities retain on average 30% and 70% of the plate speeds, respectively. Our results show that Earth's plate boundaries are not uniform and better described by more discrete plate boundaries within the oceans and distributed faults within continents. Furthermore, they emphasize the impact of plate boundary geometry on the direction and speed of plate motions and reaffirm the importance of slab pull in the uppermost mantle as a major plate driving force.

Mantle fragmentation and incomplete core merging of colliding planetesimals as evidenced by pallasites

Main group pallasites were likely formed by the collision of their parent body with a smaller impactor that caused a mixing of mantle and core material from the target and impactor, respectively. In order to better constrain the collision, we present particle size distribution (PSD) analyses of olivines in seven main group pallasites and of the Eagle station pallasite. The PSDs of two fragmental pallasites (Admire and Huckitta) contain linear segments in bi-logarithmic particle number versus size diagrams that are similar to terrestrial cataclasites or impact-related rocks. On the other hand, the PSDs of four angular pallasites only display linear segments above their respective average grain sizes. We argue that fragmental pallasites record brittle rock deformation close to the impact site of the collision, while angular pallasites represent deeper-lying mantle rocks of the target body that were disintegrated by the downward percolation of core metal from the impactor. High strain-rate deformation experiments with the system olivine – FeS melt ± Au melt produced microstructures and PSDs that are broadly similar to these two textural groups. The experiments also suggest that de-localized metal melt percolation and concomitant mantle disintegration as evidenced in angular pallasites is facilitated by weak grain boundaries caused by a small fraction of previously present metal melt in the mantle as opposed to localized diking that is dominant in a melt-free mantle. The former mechanism is expected to prevent efficient core-merging and instead causes mantle-impregnation with metal melt, which could be active when planetesimal mantles were still warm due to short-lived radiogenic heating. In addition to the parent body of the PMG, asteroid 16 Psyche may be an example of this inefficient core-merging mechanism.

Deep learning network with differentiable dynamic programming for retina OCT surface segmentation.

Multiple-surface segmentation in optical coherence tomography (OCT) images is a challenging problem, further complicated by the frequent presence of weak image boundaries. Recently, many deep learning-based methods have been developed for this task and yield remarkable performance. Unfortunately, due to the scarcity of training data in medical imaging, it is challenging for deep learning networks to learn the global structure of the target surfaces, including surface smoothness. To bridge this gap, this study proposes to seamlessly unify a U-Net for feature learning with a constrained differentiable dynamic programming module to achieve end-to-end learning for retina OCT surface segmentation to explicitly enforce surface smoothness. It effectively utilizes the feedback from the downstream model optimization module to guide feature learning, yielding better enforcement of global structures of the target surfaces. Experiments on Duke AMD (age-related macular degeneration) and JHU MS (multiple sclerosis) OCT data sets for retinal layer segmentation demonstrated that the proposed method was able to achieve subvoxel accuracy on both datasets, with the mean absolute surface distance (MASD) errors of 1.88 ± 1.96μm and 2.75 ± 0.94μm, respectively, over all the segmented surfaces.

Deformation behavior and mechanical properties of rolled carbon fiber reinforced Al-matrix composites via protection with Sn layer

How to break the restriction of elongation mismatch between the continuous carbon fibers and Al matrix in the rolling process has become a major obstacle for developing the rolled carbon fiber reinforced Al matrix (CF/Al) composites with high-performance. Herein, a strategy for liquid protection was proposed to overcome the damage of continuous CFs in rolling process of Al matrix by constructing the CF/Al composites with coating structure of Sn layer ([[email protected]]/Al composites). The design on the stress distribution and CFs damage in the rolling deformation of [[email protected]]/Al composites was optimized based on the numerical simulation. The micro-computed tomography (micro-CT) reconstruction indicated that the implantation of Sn can effectively release the stress concentration in CFs to prevent the CFs broken. After the heat treatment, the transition layer with eutectic Al-Sn structure (~550 nm) formed between the Al-matrix and CF-reinforcement, which can be used as weak boundaries, benefits to the load transfer to the CF-reinforcement with an with an increase of 23% in Vickers hardness. More importantly, the tensile strength of rolled [[email protected]]/Al composites can reach 256 MPa, which are 42.2% and 68.4% higher than that of rolled Al-matrix and rolled CF/Al composites (180 MPa and 152 MPa), respectively. This study could motivate the ingenious design of CFs damage protection for developing the advanced CF/Al composites in the automotive and aerospace industries.

MAE-BG: dual-stream boundary optimization for remote sensing image semantic segmentation

Deep learning has achieved remarkable performance in semantically segmenting remotely sensed images. However, the high-frequency detail loss caused by continuous convolution and pooling operations and the uncertainty introduced when annotating low-contrast objects with weak boundaries induce blurred object boundaries. Therefore, a dual-stream network MAE-BG, consisting of an edge detection (ED) branch and a smooth branch with boundary guidance (BG), is proposed. The ED branch is designed to enhance the weak edges that need to be preserved, simultaneously suppressing false responses caused by local texture. This mechanism is achieved by introducing improved multiple-attention edge detection blocks (MAE). Furthermore, two specific ED branches with MAE are designed to combine with typical deep convolutional (DC) and Codec infrastructures and result in two configurations of MAE-A and MAE-B. Meanwhile, multiscale edge information extracted by MAE networks is fed into the backbone networks to complement the detail loss caused by convolution and pooling operations. This results in smooth networks with BG. After that, the segmentation results with improved boundaries are obtained by stacking the output of the ED and smooth branches. The proposed algorithms were evaluated on the ISPRS Potsdam and Inria Aerial Image Labelling datasets. Comprehensive experiments show that the proposed method can precisely locate object boundaries and improve segmentation performance. The MAE-A branch leads to an overall accuracy (OA) of 89.16%, a mean intersection over union (MIOU) of 80.25% for Potsdam, and an OA of 96.61% and MIOU of 86.63% for Inria. Compared with the results without the proposed edge optimization blocks, the OAs from the Potsdam and Inria datasets increase by 5.49% and 7.64%, respectively.

Effects of cooling rate and cryogenic temperature on the mechanical properties and deformation characteristics of an Al-Mg-Si-Fe-Cr alloy

The Al-Mg-Si-Fe-Cr alloy was subjected to water quenching (WQ) and air quenching (AQ) after solution treatment for 30 min at 843 K followed by a peak artificial aging treatment at 453 K for 8 h. Precipitates of different sizes and precipitate free zone (PFZ) of different widths were obtained. Uniaxial tensile tests were conducted to investigate the effect of the microstructure and deformation temperature on the mechanical properties and deformation behavior of the WQ and AQ samples at 298 K and 77 K. The results showed that the yield strength and work hardening rate of the AQ samples were lower than those of the WQ sample. In addition, the strength and ductility of the two samples are significantly improved at cryogenic temperatures. Furthermore, quasi-in situ electron back scatter diffraction (EBSD) was conducted to investigate the orientation evolution of grains with strains of 1%, 3%, 7% and 10% for the two samples at 298 K and 77 K. The different orientation evolution of the two samples was caused by the nonuniform deformation in the PFZ and grain interior. The dislocation characteristics and fracture morphology were observed by transmission electron microscopy (TEM) and scanning electron microscopy (SEM), respectively. Because the rod-shaped precipitates hinder the dislocation movement, no obvious slip traces could be observed on the deformed surface of the AQ sample. In addition, two slip systems are activated for most grains of the WQ sample at 77 K, but only one slip system is activated at 298 K. Due to the weak grain boundaries deformed at 298 K, the stress applied to grains is consistent with the tensile direction. However, the grain boundaries are strengthened when tested at 77 K, and the stress field of the grain boundaries is more complex, resulting in the stress not only in the tensile direction. More slip systems are activated to coordinate the deformation, improving the ductility. TEM images reveal that the deformation of the PFZ is more severe for the AQ sample at 298 K than at 77 K. The deformation in the precipitate-free zone and within the grain was more homogeneous in the AQ sample at 77 K, which increased its ductility compared to the AQ sample at 298 K.

Boundary-oriented Network for Automatic Breast Tumor Segmentation in Ultrasound Images.

Breast cancer is considered as the most prevalent cancer. Using ultrasound images is a momentous clinical diagnosis method to locate breast tumors. However, accurate segmentation of breast tumors remains an open problem due to ultrasound artifacts, low contrast, and complicated tumor shapes in ultrasound images. To address this issue, we proposed a boundary-oriented network (BO-Net) for boosting breast tumor segmentation in ultrasound images. The BO-Net boosts tumor segmentation performance from two perspectives. Firstly, a boundary-oriented module (BOM) was designed to capture the weak boundaries of breast tumors by learning additional breast tumor boundary maps. Second, we focus on enhanced feature extraction, which takes advantage of the Atrous Spatial Pyramid Pooling (ASPP) module and Squeeze-and-Excitation (SE) block to obtain multi-scale and efficient feature information. We evaluate our network on two public datasets: Dataset B and BUSI. For the Dataset B, our network achieves 0.8685 in Dice, 0.7846 in Jaccard, 0.8604 in Precision, 0.9078 in Recall, and 0.9928 in Specificity. For the BUSI dataset, our network achieves 0.7954 in Dice, 0.7033 in Jaccard, 0.8275 in Precision, 0.8251 in Recall, and 0.9814 in Specificity. Experimental results show that BO-Net outperforms the state-of-the-art segmentation methods for breast tumor segmentation in ultrasound images. It demonstrates that focusing on boundary and feature enhancement creates more efficient and robust breast tumor segmentation.

Effectual accuracy of OCT image retinal segmentation with the aid of speckle noise reduction and boundary edge detection strategy.

Optical coherence tomography (OCT) has shown to be a valuable imaging tool in the field of ophthalmology, and it is becoming increasingly relevant in the field of neurology. Several OCT image segmentation methods have been developed previously to segment retinal images, however sophisticated speckle noises with low-intensity restrictions, complex retinal tissues, and inaccurate retinal layer structure remain a challenge to perform effective retinal segmentation. Hence, in this research, complicated speckle noises are removed by using a novel far-flung ratio algorithm in which preprocessing has been done to treat the speckle noise thereby highly decreasing the speckle noise through new similarity and statistical measures. Additionally, a novel haphazard walk and inter-frame flattening algorithms have been presented to tackle the weak object boundaries in OCT images. These algorithms are effective at detecting edges and estimating minimal weighted paths to better diverge, which reduces the time complexity. In addition, the segmentation of OCT images is made simpler by using a novel N-ret layer segmentation approach that executes simultaneous segmentation of various surfaces, ensures unambiguous segmentation across neighbouring layers, and improves segmentation accuracy by using two grey scale values to construct data. Consequently, the novel work outperformed the OCT image segmentation with 98.5% of accuracy.