U-shaped Structure Research Articles

GNAS-U2Net: A New Optic Cup and Optic Disc Segmentation Architecture With Genetic Neural Architecture Search

Neural architecture search (NAS) has made incredible progress in medical image segmentation tasks, due to its automatic design of the model. However, the search spaces studied in many existing studies are based on U-Net and its variants, which limits the potential of neural architecture search in modeling better architectures. In this study, we propose a new NAS architecture named GNAS-U²Net for the joint segmentation of optic cup and optic disc. This architecture is the first application of NAS in a two-level nested U-shaped structure. The best performance achieved by the joint segmentation model designed by NAS on the REFUGE dataset has an average DICE of 92.88%. Compared to U²-Net and other related work, the model has better performance and uses only 34.79M parameters. We then verify the generalization of the model on two datasets, namely the Drishti-GS dataset and the GAMMA dataset, for which we obtain an average DICE of 92.32% and 92.11% respectively.

SwinSUNet: Pure Transformer Network for Remote Sensing Image Change Detection

Convolutional neural network (CNN) can extract effective semantic features, so it was widely used for remote sensing image change detection (CD) in the latest years. CNN has acquired great achievements in the field of CD, but due to the intrinsic locality of convolution operation, it could not capture global information in space-time. The transformer was proposed in recent years and it can effectively extract global information, so it was used to solve computer vision (CV) tasks and achieved amazing success. In this article, we design a pure transformer network with Siamese U-shaped structure to solve CD problems and name it SwinSUNet. SwinSUNet contains encoder, fusion, and decoder, and all of them use Swin transformer blocks as basic units. Encoder has a Siamese structure based on hierarchical Swin transformer, so encoder can process bitemporal images in parallel and extract their multiscale features. Fusion is mainly responsible for the merge operation of the bitemporal features generated by the encoder. Like encoder, the decoder is also based on hierarchical Swin transformer. Different from the encoder, the decoder uses upsampling and merging (UM) block and Swin transformer blocks to recover the details of the change information. The encoder uses patch merging and Swin transformer blocks to generate effective semantic features. After the sequential process of these three modules, SwinSUNet will output the change maps. We did expensive experiments on four CD datasets, and in these experiments, SwinSUNet achieved better results than other related methods.

Miniaturized Metal-Mountable U-Shaped Inductive-Coupling-Fed UHF RFID Tag Antenna With Defected Microstrip Surface

This study presents a novel miniature ultra-high frequency (UHF) radio frequency identification (RFID) tag for metallic objects. Its arrangement includes a U-shaped feeder, which is inductively coupled to two E-type connected patches. Size reduction is achieved by means of utilizing the U-shaped feeder, and introducing a defection in the connection between the two E-type patches. The defection in the connection area between the patches modifies the current distribution by increasing the electrical length, and the antenna inductance. Furthermore, increasing the length of the U-shaped structure significantly raises the antenna inductance and, consequently, the flexibility of the structure for the impedance matching purpose. Polytetrafluoroethylene (PTFE) is employed for the fabrication of the designed tag. The tag’s size is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$65\,\,mm\times 20\,\,mm\times 1.5\,\,mm$ </tex-math></inline-formula> , and does not require any complex fabrication process such as utilization of metal vias and/or shorting stubs. It is observed that with 4W EIRP, the measured detection distance is <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$6.9~m$ </tex-math></inline-formula> when it is attached onto metal sheets ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$20\,\,cm\times 20\,\,cm$ </tex-math></inline-formula> ), and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$5.54~m$ </tex-math></inline-formula> when it is in free space. The measurement results also are in good agreement with the simulated results.

First structural observation around the hinge of the Mongolian Orocline (Central Asia): Implications for the geodynamics of oroclinal bending and the evolution of the Mongol-Okhotsk Ocean

ABSTRACT To understand the origin of curved subduction zones has been one of the major challenges in plate tectonics. The Mongol-Okhotsk Orogen in Central Asia is characterized by the development of a U-shaped oroclinal structure that was accompanied by the continuous subduction of the Mongol-Okhotsk oceanic plate. Therefore, it provides a natural laboratory to understand why and how a subduction system became tightly curved. In this study, we provide the first structural observation around the hinge of the Mongolian Orocline (the Zag zone in Central Mongolia), with an aim to constrain the oroclinal geometry and to link hinge zone structures with the origin of the orocline. Our results show that rocks in the Zag zone are characterized by the occurrence of a penetrative foliation that is commonly subparallel to bedding. Both bedding and dominant fabric in the Zag zone are steeply dipping, and their strike orientations in a map view follow a simple curve around the hinge of the Mongolian Orocline, thus providing the first structural constraint for 3D geometry of the orocline. A secondary penetrative fabric parallel to the axial plane of the orocline was not observed, indicating a low degree of orogen-parallel shortening during oroclinal bending. Combining with available geological and geophysical data, we conclude that the Mongolian Orocline was developed in a period of Permian to Jurassic, and its origin was linked to the subduction of the Mongol-Okhotsk oceanic slab. We consider that the low-strain oroclinal bending likely resulted from the along-strike variation in trench retreat, which was either triggered by the negative buoyancy of the Mongol-Okhotsk oceanic slab, or driven by the relative rotation of the Siberian and North China cratons. Our results shed a light on 3D geometry and geodynamic mechanisms of large-scale oroclinal bending in an accretionary orogen.

Characteristics of thickness-vibration-mode PZT transducer for acoustic micropumps

The characteristics of a thickness-vibration-mode PZT transducer for acoustic micropumps are evaluated with a U-shaped tube structure. The phenomenon in which the height of the water level can be increased in a U-shaped tube due to the acoustic driving force was used to develop measurement methods for the output pressure and flow rate of the acoustic micropump driven by a PZT transducer. The measurement results showed that the output pressure increases from 22.6 to 37.6 Pa and the flow rate increases from 3.6 to 6.4 ml/min when input voltages of 68–105 Vpp are applied to the PZT transducer and the height of water level is 30 cm above the PZT transducer. To discuss the effect of backpressure, the output characteristics of pressure and flow at water height levels of 31 and 32 cm were measured for comparison with those at 30 cm.

Multistrengthening Module-Based Salient Object Detection

Object detection is a classical research problem in computer vision, and it is widely used in the automatic monitoring field of various production safety. However, current object detection techniques often suffer low detection accuracy when an image has a complex background. To solve this problem, this paper proposes a double U-shaped multireinforced unit structure network (DUMRN). The proposed network consists of a detection module (DM), a reinforced module (RM), and a salient loss function (SLF). Extensive experiments on five public datasets and a practical application dataset are conducted and compared against nine state-of-the-art methods. The experiment results show the superiority of our method over the state of the art.

SUACDNet: Attentional change detection network based on siamese U-shaped structure

Remote sensing image change detection is an essential aspect of remote sensing technology application. Existing change detection algorithms based on deep learning do not distinguish between changed and unchanged areas explicitly, resulting in serious loss of edge detail information during detection. Therefore, a new attentional change detection network based on Siamese U-shaped structure (SUACDNet) is proposed in this paper. In the feature encoding stage, three branches are introduced between the Siamese structure to focus on the global information, difference information and similarity information of bitemporal images respectively. In the feature decoding stage, an U-shaped structure is constructed for upsampling and recovery layer by layer. Multi-scale Convolution Residual Module (MCRM) is a new convolution structure designed for multi-scale feature extraction in the network. In addition, this work also proposes three auxiliary modules to optimize the network, namely Spatial Attention Module (SAM), Feature Fusion Module (FFM) and Cross-scale Global Context Semantic Information Aggregation Module (CGCAM), making the network more sensitive to the changed area while filtering out the background noise. Comparative experiments on three datasets show that our method is superior to the existing methods.

URNet: A U-Shaped Residual Network for Lightweight Image Super-Resolution

It is extremely important and necessary for low computing power or portable devices to design more lightweight algorithms for image super-resolution (SR). Recently, most SR methods have achieved outstanding performance by sacrificing computational cost and memory storage, or vice versa. To address this problem, we introduce a lightweight U-shaped residual network (URNet) for fast and accurate image SR. Specifically, we propose a more effective feature distillation pyramid residual group (FDPRG) to extract features from low-resolution images. The FDPRG can effectively reuse the learned features with dense shortcuts and capture multi-scale information with a cascaded feature pyramid block. Based on the U-shaped structure, we utilize a step-by-step fusion strategy to improve the performance of feature fusion of different blocks. This strategy is different from the general SR methods which only use a single Concat operation to fuse the features of all basic blocks. Moreover, a lightweight asymmetric residual non-local block is proposed to model the global context information and further improve the performance of SR. Finally, a high-frequency loss function is designed to alleviate smoothing image details caused by pixel-wise loss. Simultaneously, the proposed modules and high-frequency loss function can be easily plugged into multiple mature architectures to improve the performance of SR. Extensive experiments on multiple natural image datasets and remote sensing image datasets show the URNet achieves a better trade-off between image SR performance and model complexity against other state-of-the-art SR methods.

SGUNet: Style-guided UNet for adversely conditioned fundus image super-resolution

Image super-resolution from low-resolution fundus image has valuable applications in clinical practices. The popular methods yield unsatisfactory results when the fundus images are contaminated due to the bleeding or plaques caused by eye diseases. To this end, we propose a style-guided UNet (SGUNet) which incorporates a series of style-guided U-shape block (SUB) for fundus image super-resolution. Each SUB consists of trunk and mask branches. The proposed trunk branch is a U-shape structure that intends to enlarge the receptive field by down-sampling via large-stride convolution, and fuses the complementary information under the different receptive fields. The mask branch then dynamically estimates the relative importance of individual potential styles to reweigh the feature maps according to the significance of the potential styles. To fully leverage the hierarchical features, a dense feature fusion scheme is introduced by concatenating the output of preceding SUBs. We extensively validate the proposed network on low-resolution retina dataset with adversely affected by diseases. The experimental results demonstrate that our SGUNet achieves superior performance with excellent robustness and high accuracy by comparing with six popular methods.

Design of a Low-Ripple Double-Modular-Stator Switched Reluctance Machine for Electric Vehicle Applications

To overcome the disadvantages of traditional switched reluctance machine (SRM) with moderate power/torque density and high torque ripple, this article proposes a novel double-stator SRM (DSSRM) with the inner and outer stators both adopting U-shaped segmental stator structure. By optimizing the polarity distribution of the inner and outer stator magnetic fields, the inner and outer stator magnetic fields are decoupled. Then, the relationships of torque ripple coefficient with the mechanical offset angles of the inner and outer rotor teeth are investigated. To optimize the mechanical offset angle and reduce the torque ripple of the motor, the initial design method is proposed, and five main parameters of the machine are optimized by the finite-element analysis. Then, the static magnetic characteristics and dynamic performance of this novel machine are simulated and analyzed. Finally, a 16/18/18/16 DSSRM is prototyped to validate the above simulation and theoretical analysis.

Characteristic and paleoenvironment significance of microbially induced sedimentary structures (MISS) in terrestrial facies across P-T boundary in Western Henan Province, North China

AbstractWe report varied microbially induced sedimentary structures (MISS), and other sedimentary surface textures, from the Induan (Early Triassic) Sunjiagou Formation and Liujiagou Formation in the Xingyang, Dengfeng, Jiyuan and Yiyang areas, western Henan Province, North China. Microanalysis shows that these MISS are characterized by a U-shaped structure, thin clayey laminae, and discontinuous mica sheet that are arranged parallel to the bedding plane, as well as directionally oriented quartz grains floating in lamina, which are indicative of a biogenic origin. The MISS of the studied area were probably affected by four main factors, including the end-Permian mass extinction, the megamonsoon, the adapted sedimentary environment, and the sediment supply, and they possess significant stratigraphic correlation. Abundant microbial-related sedimentary structures from the study area indicate that continental ecosystems were severely devastated in the aftermath of the Permian-Triassic biocrisis. These sedimentary structure assemblages, including MISS, red beds, conglomerate layers, and calcareous concretions in the western Henan Province, show a specific, post-extinction continental ecosystem that was characterized by microflora dominance, monotonous and rare fossils, extreme hot climate, soil ecosystem devastation, and poor vegetation.

A Broadband 90° Balun With Low-Phase-Imbalance Performance Based on Periodic Slow Wave Structure

We propose a method to construct the broadband 90° balun with low-phase-imbalance performance based on a U-shaped periodic slow wave structure (PSWS) transmission line (TL), whose dispersion curve has a linear region in a wide frequency band. By introducing meander lines and U-shaped TLs, the linearity, cutoff frequencies and slope of dispersion curves of PSWS can be controlled flexibly. Simulated and measured results show good agreements. Both reflection coefficient and isolation of the proposed balun are below -15 dB from 1.5 to 2.0 GHz. The phase deviation is within ±3°, covering a fractional bandwidth of 28.6%. Compared with that based on the traditional corrugated periodic slow wave TL unit, the operating bandwidth of the proposed balun is increased by more than six times. Moreover, the proposed PSWS keeps excellent ability of field confinement and therefore improves the isolation between the output ports. In a word, based on the proposed method, the microwave devices and circuits which realize the function of phase shifting can be designed.

Anisotropic wet etching of a novel micro-texture structure for an Al/n-Si/Al metal–semiconductor–metal photodetector fabrication

Micro-electro-mechanical system fabrication involves molding on a single substrate. Chemical wet etching is common in these systems because it can provide a very high etch rate and selectivity. This optical device has been successfully completed fabrication by using the single-step lithography process. The surface was coated with PR and patterned using a single mask to create U-shaped structures by a wide electrode gap on a silicon substrate. Then it was etched using anisotropic wet etching process in potassium hydroxide, tetramethylammonium hydroxide (TMAH), silicic acid and dissolved silicon powder. The etchant created a 58% increase in the light sensitive area of the U-shaped trench in the structure higher compared to the planar structure. Random micro-pyramids, formed on the silicon surface in the U-shaped structure by the silicic acid-added TMAH solution, led to a ∼254 times higher ratio between photon generated current and dark current at reverse bias. This processing technique is a promising technology for improving performance for next-generation optical sensors or silicon photodetector.

Improved Uniformity of Photoresist Ashing for a Half-Inch Wafer with Double U-shaped Antenna Structure in a Microwave-Excited Water Vapor Plasma

Improved Uniformity of Photoresist Ashing for a Half-Inch Wafer with Double U-shaped Antenna Structure in a Microwave-Excited Water Vapor Plasma

Craniofacial Reconstruction via Face Elevation Map Estimation Based on the Deep Convolution Neutral Network

In this study, to achieve the possibility of predicting face by skull automatically, we propose a craniofacial reconstruction method based on the end-to-end deep convolutional neural network. Three-dimensional volume data are obtained from 1447 head CT scans of Chinese people of different ages. The facial and skull surface data are projected onto two-dimensional space to generate a two-dimensional elevation map, and then, use the deep convolution neural network to realize the prediction of skull to face shape in two-dimensional space. The encoder and decoder are composed of first feature extraction through the encoder and then as the input of the decoder to generate the craniofacial restoration image. In order to accurately describe the features of different scales, we adopt an U-shaped codec structure with cross-layer connections. Therefore, the output features are decomposed with the features of the corresponding scales in the encoding stage to achieve the integration of different scales while restoring the feature scales in the compression and decoding stage. Meanwhile, the U-net structures help to avoid the problem of loss of detail features in the downsampling process. We use supervised learning to obtain the prediction model from skull to facial elevation map. Back-projection operation is performed afterwards to generate facial surface data in 3D space. Experiments show that the proposed method in this study can effectively achieve craniofacial reconstruction, and for most part of the face, restoration error is controlled within 2 mm.

Enhancing feature fusion with spatial aggregation and channel fusion for semantic segmentation

Semantic segmentation is crucial to the autonomous driving, as an accurate recognition and location of the surrounding scenes can be provided for the street scenes understanding task. Many existing segmentation networks usually fuse high-level and low-level features to boost segmentation performance. However, the simple fusion may impose a limited performance improvement because of the gap between high-level and low-level features. To alleviate this limitation, we respectively propose spatial aggregation and channel fusion to bridge the gap. Our implementation, inspired by the attention mechanism, consists of two steps: (1) Spatial aggregation relies on the proposed pyramid spatial context aggregation module to capture spatial similarities to enhance the spatial representation of high-level features, which is more effective for the latter fusion. (2) Channel fusion relies on the proposed attention-based channel fusion module to weight channel maps on different levels to enhance the fusion. In addition, the complete network with U-shape structure is constructed. A series of ablation experiments are conducted to demonstrate the effectiveness of our designs, and the network achieves mIoU score of 81.4% on Cityscapes test dataset and 84.6% on PASCALVOC 2012 test dataset.

Thermoluminescence Analysis of Bricks from the so-called Arch of Ali Shah

The application of Thermally (TL) and Optically (OSL) Stimulated Luminescence on bricks used as building material can answer questions regarding the chronology of historical buildings. The remarkable historical reports of the “largest ever made brick vault” known as the Arch of Ali Shah (14th century) invoke the image of a gigantic structure adjoining the u-shaped brick monument in Tabriz. However, there is new scientific data that has led us to consider an alternate hypothesis regarding this monument that contradicts the traditional views of scholars. The attribution of this controversial building to Ali Shah, the great vizier of the Ilkhanid court, has long been considered an historical fact by scholars. To better understand the evolution of this unique structure, thermoluminescence (TL) was used to propose a relative dating for its construction. Surprisingly, the results yielded dates of 512±20, 514±27 and 517±21 AD (TL age, equaling 17th century), indicating that the u-shaped structure was built some 200 years after the Ilkhanid era. These dates, supplemented with historical context and architectural evidence, leads to the conclusion that it was used as a separate building before being integrated into the older building, thereby converting the entire complex into a formidable fortification. The remains of the so called Arch of Ali Shah are clearly of a later date, characterizing a completely different architectural style than those of the Ilkhanid period.

Imaging error compensation method for through-focus scanning optical microscopy images based on deep learning.

Through-focus scanning optical microscopy (TSOM) is a model-based nanoscale metrology technique which combines conventional bright-field microscopy and the relevant numerical simulations. A TSOM image is generated after through-focus scanning and data processing. However, the mechanical vibration and optical noise introduced into the TSOM image during image generation can affect the measurement accuracy. To reduce this effect, this paper proposes a imaging error compensation method for the TSOM image based on deep learning with U-Net. Here, the simulated TSOM image is regarded as the ground truth, and the U-Net is trained using the experimental TSOM images by means of a supervised learning strategy. The experimental TSOM image is first encoded and then decoded with the U-shaped structure of the U-Net. The difference between the experimental and simulated TSOM images is minimised by iteratively updating the weights and bias factors of the network, to obtain the compensated TSOM image. The proposed method is applied for optimising the TSOM images for nanoscale linewidth estimation. The results demonstrate that the proposed method performs as expected and provides a significant enhancement in accuracy.

Advances in Molecular Understanding of α-Helical Membrane-Active Peptides.

Biological membranes separate the interior of cells or cellular compartments from their outer environments. This barrier function of membranes can be disrupted by membrane-active peptides, some of which can spontaneously penetrate through the membranes or open leaky transmembrane pores. However, the origin of their activity/toxicity is not sufficiently understood for the development of more potent peptides. To this day, there are no design rules that would be generally valid, and the role of individual amino acids tends to be sequence-specific.In this Account, we describe recent progress in understanding the design principles that govern the activity of membrane-active peptides. We focus on α-helical amphiphilic peptides and their ability to (1) translocate across phospholipid bilayers, (2) form transmembrane pores, or (3) act synergistically, i.e., to produce a significantly more potent effect in a mixture than the individual components.We refined the description of peptide translocation using computer simulations and demonstrated the effect of selected residues. Our simulations showed the necessity to explicitly include charged residues in the translocation description to correctly sample the membrane perturbations they can cause. Using this description, we calculated the translocation of helical peptides with and without the kink induced by the proline/glycine residue. The presence of the kink had no effect on the translocation barrier, but it decreased the peptide affinity to the membrane and reduced the peptide stability inside the membrane. Interestingly, the effects were mainly caused by the peptide's increased polarity, not the higher flexibility of the kink.Flexibility plays a crucial role in pore formation and affects distinct pore structures in different ways. The presence of a kink destabilizes barrel-stave pores, because the kink prevents the tight packing of peptides in the bundle, which is characteristic of the barrel-stave structure. In contrast, the kink facilitates the formation of toroidal pores, where the peptides are only loosely arranged and do not need to closely assemble. The exact position of the kink in the sequence further determines the preferred arrangement of peptides in the pore, i.e., an hourglass or U-shaped structure. In addition, we demonstrated that two self-associated (via termini) helical peptides could mimic the behavior of peptides with a helix-kink-helix motif.Finally, we review the recent findings on the peptide synergism of the archetypal mixture of Magainin 2 and PGLa peptides. We focused on a bacterial plasma membrane mimic that contains negatively charged lipids and lipids with negative intrinsic curvature. We showed that the synergistic action of peptides was highly dependent on the lipid composition. When the lipid composition and peptide/lipid ratios were changed, the systems exhibited more complex behavior than just the previously reported pore formation. We observed membrane adhesion, fusion, and even the formation of the sponge phase in this regime. Furthermore, enhanced adhesion/partitioning to the membrane was reported to be caused by lipid-induced peptide aggregation.In conclusion, the provided molecular insight into the complex behavior of membrane-active peptides provides clues for the design and modification of antimicrobial peptides or toxins.

STUDY ON EFFICACY OF ULTRASOUND AND FOB IN CONFIRMING THE PLACEMENT OF LARYNGEAL MASK AIRWAY

Ultrasound is a safe and non invasive technique for assessing the airway . In this observational study, 60 patients belonging to ASAClasses I and II scheduled for elective general surgery without any airway abnormality were included. Squared wave of capnogramTransverse neck ultrasound was performed using linear probe (8- 13Hz) started from hyoid bone (hyperechoic supercial inverted Ushaped linear structure)probe moved caudally with slight cephalad angle visualization of vocal cord &amp; arytenoids. The Ultrasound arytenoid grade had a signicant correlation with LMA rotation grade.The incidence of LMAMalposition by ultrasound was found to be 80% .With regard to the test performance ,US was found to have a sensitivity of 92%,with a positive predictive value of 94% to detect the LMA rotation of any number of degrees. The specicity was 62% with a negative predictive value of 55%. FOB Examination is the gold standard tool for conrmation of the correct placement of LMA. ,we evaluated USG as a tool to conrm the placement comparing with the FOB examinationThe FOB examination is invasive &amp; interrupts the ventilation during procedure. It cannot be done frequently intra-operatively compromising ventilation. These problems can be overcome by using ultrasound.USG detects the rotation of LMAeven if it is positioned at optimal depth thereby predicting the chance of dislodgement during the course of surgery.