Segmental Mode Research Articles

High-performance deep learning segmentation for non-destructive testing of X-ray tomography

For non-destructive testing (NDT) with X-ray computed tomography (XCT), the complex morphological segmentation of homogeneous defects is impossible for traditional threshold-based segmentation algorithm, due to their same absorbances. A variety of defect segmentation using convolutional neural networks (CNNs) has been suitable for such problems, and its accuracy is determined by image quality and comprehensiveness of training set. To achieve a high-performance measurement for quantitative defect evaluation, a modified U-Net-based segmentation model of XCT was proposed suitable for defects with various image quality resulting from different experimental efficiency. The dataset consisting of 24 subsets was acquired from the condition-varying measurements of aluminum alloy samples produced by laser 3D metal printing additive manufacturing. The structure and distribution of segmented pore and crack defects were accurately visualized by our model. Compared with traditional and main CNNs-based segmentation methods, our model exhibited a better recognition accuracy for pores and cracks, the Mean Intersection over Union and Pixel accuracy metrics reaching to 84.83 % and 98.87 % respectively. Practically, the option of efficiency or accuracy priority was quantitatively determined to carry out such a high-performance defect-related NDT. The proposed XCT-based segmentation mode has a great potential in fields of engineering, material science, biomedicine.

Convolutional neural network for automated tooth segmentation on intraoral scans

BackgroundTooth segmentation on intraoral scanned (IOS) data is a prerequisite for clinical applications in digital workflows. Current state-of-the-art methods lack the robustness to handle variability in dental conditions. This study aims to propose and evaluate the performance of a convolutional neural network (CNN) model for automatic tooth segmentation on IOS images.MethodsA dataset of 761 IOS images (380 upper jaws, 381 lower jaws) was acquired using an intraoral scanner. The inclusion criteria included a full set of permanent teeth, teeth with orthodontic brackets, and partially edentulous dentition. A multi-step 3D U-Net pipeline was designed for automated tooth segmentation on IOS images. The model’s performance was assessed in terms of time and accuracy. Additionally, the model was deployed on an online cloud-based platform, where a separate subsample of 18 IOS images was used to test the clinical applicability of the model by comparing three modes of segmentation: automated artificial intelligence-driven (A-AI), refined (R-AI), and semi-automatic (SA) segmentation.ResultsThe average time for automated segmentation was 31.7 ± 8.1 s per jaw. The CNN model achieved an Intersection over Union (IoU) score of 91%, with the full set of teeth achieving the highest performance and the partially edentulous group scoring the lowest. In terms of clinical applicability, SA took an average of 860.4 s per case, whereas R-AI showed a 2.6-fold decrease in time (328.5 s). Furthermore, R-AI offered higher performance and reliability compared to SA, regardless of the dentition group.ConclusionsThe 3D U-Net pipeline was accurate, efficient, and consistent for automatic tooth segmentation on IOS images. The online cloud-based platform could serve as a viable alternative for IOS segmentation.

Aquaculture Ponds Identification Based on Multi-Feature Combination Strategy and Machine Learning from Landsat-5/8 in a Typical Inland Lake of China

Inland aquaculture ponds, as an important land use type, have brought great economic benefits to local people but at the same time have caused many environmental problems threatening regional ecology security. Therefore, understanding the spatiotemporal pattern of aquaculture ponds and its potential influence on water quality is vital for the sustainable development of inland lakes. In this study, based on Landsat5/8 images, three types of land features, namely spectral features, index features, and texture features, and five machine learning algorithms, namely random forest (RF), extreme gradient boosting (XGBoost), artificial neural network (ANN), k-nearest neighbor (KNN), and Gaussian naive Bayes (GNB), were combined to identify aquaculture ponds and some other primary land use types around a typical inland lake of China. The results demonstrated that the XGBoost algorithm that integrated the three features performed the best among all groups of the five machine learning algorithms and the three features, with an overall accuracy of up to 96.15%. In particular, the texture features provided additional useful information besides the spectral features to allow more accurately separation of aquaculture ponds from other land use types and thus improve the land use mapping ability in complex inland lakes. Next, this study examined the tendency of aquaculture ponds and found a segmented increase mode, namely sharp increase during 1984–2003 and then slow elevation since 2003. Further positive correlation detected between the area of aquaculture ponds and the phytoplankton population dynamics suggest a likely influence of aquaculture activity on the lake water quality. This study provides an important scientific basis for the sustainable management and ecological protection of inland lakes.

Mechanism of chip segmentation transition from shear slip to shear fracture of Zirconium-based bulk metallic glass in mechanical machining

An in-depth understanding of material cutting deformation forms the foundation for manufacturing Zirconium-based bulk metallic glass (Zr-based BMG) parts. This study explores the transition in chip segmentation mode from shear slip to shear fracture in Zr-based BMG during mechanical machining. The investigation covers chip micromorphology, cyclic cutting force, machined surface quality, and the themo-mechanical properties under various chip segmentation modes linked to cutting conditions. A size-dependent transition in chip segmentation has been observed in the cutting of Zr-based BMG. At a specific cutting speed, a smaller uncut chip thickness favors chip formation via shear slip, resulting in a smooth primary shear zone (PSZ). Conversely, a larger uncut chip thickness leads to chip shear fracture, with the PSZ displaying fishbone-like and dimple patterns. The integration of the modified cutting kinematic model with the free volume-dependent shear transformation zone (STZ) model suggests that the transition from shear slip to fracture is due to an increase in uncut chip thickness, which causes a greater STZ volume in the PSZ. When the STZ volume surpasses a threshold, the intrinsic structural recovery of material from shearing cannot fully compensate for the free volume softening, leading to excessive free volume that ultimately triggers shear fracture. During the chip shear fracture process, the evolution from fishbone-like patterns to dimples in the PSZ is attributed to faster crack propagation at greater uncut chip thickness or higher cutting speeds. In practical applications, it is advisable to avoid chip shear fracture during machining to maintain high-quality machined surface quality.

Design and Experiment of Variable-Spray System Based on Deep Learning

In response to issues of the low pesticide-utilization rate caused by the traditional constant spraying method, a variable-spraying system for wheat-field weeds was proposed in this study based on real-time segmentation by deep learning. In this study, the weed density within the operational area was obtained by using the improved DeepLab V3+ semantic segmentation mode, and a variable spray-level model based on real-time weed density and speed was constructed by using PWM variable-spraying technology to adjust the spray volume. The lightweight MobileNet V2 network was selected as its backbone network, and the CA attention mechanism was integrated into the feature extraction module. The mean intersection over the union (MIoU) and mean pixel accuracy (MPA) of the improved DeepLab V3+ were 73.34% and 80.76%, respectively, and the segmentation time for a single image was 0.09 s. The results of field verification tests showed that (1) compared with constant spraying, variable spraying can increase droplet density and save the amount of pesticides, with the droplet density increased by 38.87 droplets/cm2 and the pesticide consumption saved by 46.3%; and (2) at the same speed, the average droplet-coverage rate in the areas with sparse weed density decreased by 13.98% compared with the areas with dense weeds. Under the same plant density, the average coverage rate of 0.5 m/s increased by 2.91% and 6.59% compared with 1 m/s and 1.5 m/s, respectively. These results further demonstrated that the system can automatically adjust the spray volume based on different travel speeds and weed densities. This research can provide theoretical and reference support for the development of new precision-spray plant-protection machinery for wheat fields.

Underwater dynamic modeling and experiments of flexible structure with harmonic actuation of macro fiber composites

Flexible structures driven by smart actuators are promising alternatives for aquatic bionic propulsion vehicles. However, the hydrodynamic effects induced by viscous fluids on the dynamic response of flexible structures remain an ongoing challenge. Thus, this article presents a fluid-structure interaction dynamic model of a flexible underwater structure actuated by macro fiber composite actuators, and the underwater multimodal vibration characteristics are studied. The model is based on the extended Hamilton’s principle, which considers the effects of hydrodynamic forces. The segmented mode shape functions of the flexible structure are derived using the assumed mode method. The study shows that the first two mode shapes of the beam predicted by the model agrees with the experimental results. The underwater dynamic responses of the flexible structure in a board band covering the first two resonance frequencies are also investigated at different actuation levels. The underwater results indicate that the first two resonance frequencies are 1.05 and 6.1 Hz respectively, basically consistent with the simulation ones (1.07 and 6.61 Hz). Correspondingly, the maximum transverse deflections at the end of the beam are 4.81 and 1.31 mm respectively, which are slightly lower than the predicted values of 5.75 and 1.62 mm. Therefore, the validity of the proposed coupled dynamic model is verified, which can be utilized to predict the multimodal dynamic responses of underwater flexible structures actuated by MFC or other smart actuators.

Research on the design method of mode transition control law for Ma6 external parallel TBCC engine

Abstract To achieve a fast and reliable mode transition of the Ma 0–6 external parallel TBCC engine, a design method for the transition state control law is developed. The design parameters of the TBCC engine are obtained based on the mission requirements of hypersonic aircraft. For the key mode transition process of the combination engine, a segmented mode transition strategy is formulated with the objectives of smooth thrust and continuous flow. The mode transition control law of the combination engine is designed and optimized using the Hooke-Jeeve Direct Search Method. The results demonstrate that the proposed design method for the mode transition control law can effectively enable a fast and smooth mode transition of the combination engine, with a total duration of 7 s. During the mode transition process, the total thrust fluctuation during the transition from the turbofan windmill state to the turbofan shutdown state is the largest, and the maximum relative error between the total thrust and the expected value during the mode transition process is 0.4 %.

Hi-SAM: Marrying Segment Anything Model for Hierarchical Text Segmentation.

The Segment Anything Model (SAM), a profound vision foundation model pretrained on a large-scale dataset, breaks the boundaries of general segmentation and sparks various downstream applications. This paper introduces Hi-SAM, a unified model leveraging SAM for hierarchical text segmentation. Hi-SAM excels in segmentation across four hierarchies, including pixel-level text, word, text-line, and paragraph, while realizing layout analysis as well. Specifically, we first turn SAM into a high-quality pixel-level text segmentation (TS) model through a parameter-efficient fine-tuning approach. We use this TS model to iteratively generate the pixel-level text labels in a semi-automatical manner, unifying labels across the four text hierarchies in the HierText dataset. Subsequently, with these complete labels, we launch the end-to-end trainable Hi-SAM based on the TS architecture with a customized hierarchical mask decoder. During inference, Hi-SAM offers both automatic mask generation (AMG) mode and promptable segmentation (PS) mode. In the AMG mode, Hi-SAM segments pixel-level text foreground masks initially, then samples foreground points for hierarchical text mask generation and achieves layout analysis in passing. As for the PS mode, Hi-SAM provides word, text-line, and paragraph masks with a single point click. Experimental results show the state-of-the-art performance of our TS model: 84.86% fgIOU on Total-Text and 88.96% fgIOU on TextSeg for pixel-level text segmentation. Moreover, compared to the previous specialist for joint hierarchical detection and layout analysis on HierText, Hi-SAM achieves significant improvements: 4.73% PQ and 5.39% F1 on the text-line level, 5.49% PQ and 7.39% F1 on the paragraph level layout analysis, requiring 20× fewer training epochs. The code is available at Hi-SAM.

Uncovering gender-based violence and harassment in public transport: Lessons for spatial and transport justice

When suitably organized, public transport plays an important role in social inclusion and equity by providing crucial access to social, political, and economic opportunities. Notwithstanding, a growing body of evidence points to gender-based harassment and violence associated with public transport journeys as significant barriers to women's travel. This raises important issues regarding spatial and transport justice.Using a survey in Santiago, Chile, this study explores gender-based violence and harassment experienced while accessing, egressing, and using public transport. We use zero-inflated and logistic regression models to explore individual, behavioral and spatial factors associated with gender violence, particularly sexual harassment, defined as unwanted touching, filming or photography, stalking or masturbation, and rape. We examine these factors differentiating by trip segment (access to and egress from the system versus in-vehicle experience) and public transport mode (bus and Metro).Our results show that 17.5% experienced four or more harassment situations. Public masturbation or rape primarily occurred on the journey to and from public transport (8.6%) and on board a bus (7.4%). Our models indicate a spatial component of harassment which intersects with behavioral and individual factors as trip frequency, trip purpose, gender, age, and income, reducing public transit's ability to serve women, particularly those in low-income living in more peripheral areas. This article contributes to knowledge regarding how these factors interact in a major city in the Global South, which has been under-explored to date. Our findings join other studies challenging theories of spatial and transport justice, which to date pay little attention to violence in general and gender-based violence in particular. We make suggestions for reinforcing theory in order to develop stronger solutions to ensure transport systems improve equity rather than locking in current discrimination and exclusions.

DECC: Achieving Low Latency in Data Center Networks With Deep Reinforcement Learning

Data Center Networks (DCNs) suffer from synchronized bursts for segmentation and aggregation modes, leading to buffer overflows at switches and increasing network delay. To overcome this problem, some congestion control algorithms like DCTCP use Explicit Congestion Notification (ECN) to notify in-network congestion and reduce switch buffer occupancy. However, the traditional Additive Increase Multiplicative Decrease (AIMD) method causes high fluctuation of round-trip time (RTT) in DCNs. Some intelligent congestion control algorithms designed for Internet can achieve great flexibility, but are not applicable in DCNs for a lack of accurate congestion feedback. In this paper, we analyze the deficiencies of utilizing RTT as congestion signals and the applicability of learning algorithms in DCNs. Then, we propose DECC, a smart TCP congestion control algorithm for DCNs, which combines Deep Reinforcement Learning (DRL) with ECN to achieve high bandwidth utilization as well as low queuing delay. DECC fully utilizes precise in-network feedback and formulates several QoS requirements to a multi-objective function. Meanwhile, it decouples cwnd adjustment with DRL decision making to gradually learn the optimal congestion control policy in real-time. We evaluate the performance of DECC in various scenarios. Simulation results show that DECC can reduce the queue length at bottleneck switches by more than 50% compared to DCTCP, while maintaining high bandwidth utilization and reducing Flow Completion Time (FCTs) under burst traffic.

Simultaneous determination of tipiracil, trifluridine and its metabolite 5-trifluoromethyluracil in human plasma using segmented polarity LC-MS/MS: A fully validated assay with clinical application

Trifluridine (FTD) and tipiracil (TPI) hydrochloride tablets (TAS-102) were used for the treatment of patients with metastatic rectal cancer that was resistant to conventional chemotherapy drugs. In this study, a rapid and sensitive liquid chromatography-tandem mass spectrometry method was developed and fully validated for the simultaneous determination of TPI, FTD, and the metabolite 5-trifluoromethyluracil (FTY) of FTD in human plasma. The plasma samples were prepared by protein precipitation. The chromatography separation was performed using ACE Excel 3 AQ (100 × 2.1 mm i.d., 1.7 µm, ACE, England) column protected by a security guard cartridge (4.0 × 2.0 mm i.d., 5 µm, Phenomenex, USA) with a gradient elution of 0.05% acetic acid in water and methanol at a flow rate of 0.35 mL/min. The MS/MS analysis was performed by using multiple reaction monitoring with the segmented polarity (positive for TPI: m/z 243.1→183.0, and negative for FTD: m/z 295.1→252.0 and FTY: m/z 178.9→158.9) electrospray ionization mode. The segmented polarity mode was designed to achieve two advantages: better sensitivity and simultaneous determination of the analytes with different ion polarities. The calibration ranges were as follows: 1.00–250 ng/ for TPI, 8.00–8000 ng/mL for FTD and 5.00–1250 ng/mL for FTY. The selectivity, accuracy, precision, matrix effect, recovery, carryover, dilution integrity and stability test results meet ICH acceptance criteria. The method was evaluated using the RGB model and successfully applied to a clinical study in patients with solid tumors. For TPI, FTD and FTY, the maximum plasma concentration was 137–147 ng/mL, 6160–6240 ng/mL and 724–725 ng/mL, respectively; the plasma elimination half-life was 1.69–1.78 h, 1.70 h, and 3.09–3.14 h, respectively, after an oral administration of 60 mg TAS-102.

A novel dealloying assisted machining method to improve the machinability of NiTi alloy as a typical high toughness difficult-to-machine material

High quality and high efficiency cutting process is of great significance to the components manufacturing related to machinability of workpiece materials. As a typical difficult-to-machine material induced mainly by high toughness, NiTi alloy faces challenges such as high cutting force and difficult chip breaking during the cutting process. To solve these problems, this paper presents a novel surface dealloying assisted machining method, i.e., the preparation of nanoporous structures on the workpiece surface with dealloying treatment, to improve the machinability of NiTi alloy. Based on high speed in-situ imaging of material removal process, the presence of dealloyed layer leads to a segmented flow mode instead of homogeneous severe plastic deformation without dealloying treatment. The change in material removal mode is beneficial to decrease the cutting force and energy and improve the machined surface quality. The embrittlement effect of dealloyed layer inhibits material plastic flow and promotes periodic crack nucleation and propagation. An analytical model is developed to characterize the brittle crack propagation, which provides theoretical basis for explaining embrittlement effect of dealloyed layer. This work verifies the feasibility and effectiveness of the proposed method to modify the machinability of NiTi alloy and provides new ideas for the development of manufacturing processes.

Pre-earthquake anomaly extraction from borehole strain data based on machine learning

Borehole strain monitoring plays a critical role in earthquake precursor research. With the accumulation of observation data, traditional data processing methods struggle to handle the challenges of big data. This study proposes a segmented variational mode decomposition method and a GRU-LUBE deep learning network based on machine learning theory. The algorithm enhances data correlation during decomposition and effectively predicts borehole strain data changes. We extract pre-earthquake anomalies from four-component borehole strain data of the Guza station for two major earthquakes in Sichuan (Wenchuan and Lushan earthquakes), obtaining more comprehensive anomalies than previous studies. Statistical analysis reveals similar abnormal phenomena in the Guza station’s borehole strain data before both earthquakes, suggesting shared crustal stress accumulation and release patterns. These findings highlight the need for further research to improve earthquake prediction and preparedness through understanding underlying mechanisms.

The Effect of Dose-Fractionation Mode on the Immune Response and Therapeutic Effect of Lewis Lung Cancer Mice Model Based on Tumor Microenvironment

The therapeutic effect of different groups of Lewis lung cancer mice and the expression of immune cells in peripheral blood and tumor tissues of mice were observed after different dose-segmentation modes of treatment, so as to select the most suitable for stimulating immune effect with good prognosis and small side effects. Female homologous inbred C57BL/6 male mice at 6-8 weeks were selected and inoculated subcutaneously on the right thigh according to LLC: L929: lymphocyte = tumor cell suspension with PBMC(inactivated lymphocyte) = 5:1:1 ratio; after tumor growth to the size of 100-200mm3, different dose-segmentation modes were used for irradiation, which was divided into 5 groups, group A: no radiotherapy; Group B: 2Gy/qd, continuous radiotherapy for 30 days; Group C: 2Gy/bid, continuous radiotherapy for 13 days; Group D: 3Gy/bid, continuous radiotherapy for 10 days; Group E: 6Gy/qd, continuous radiotherapy for 8 days. On the 3rd day after radiotherapy, some mice were sacrificed, peripheral blood lymphocytes of mice were isolated with lymphocyte separation solution, and the transplanted tumor of mice was prepared into single-cell suspension. Then, the number of CD3+T cells, CD3+B cells, NK1.1 cells, CD3+CD4+T cells, CD3+CD8+T cells, CD4+Foxp3+Treg cells in peripheral blood and single-cell suspension of mice was detected by flow cytometry. HE staining was used to observe the tumor tissue structure of mice leg transplantation. The expressions of CD4, CD8, Foxp3 and CD31 in tumor tissues of each radiotherapy group were observed by immunohistochemical staining (IHC) in paraffin sections. The survival time of mice in group C (2Gy/ bidx13d) was significantly improved compared with that in group B (2Gy/qdx30d), and the twin-daily segmentation mode could improve the elevation of Treg cells after once-daily segmentation mode radiotherapy in both peripheral blood and tumor tissues. There was no significant difference in the survival time of mice in group E (6Gy/qdx8d) and group D (3Gy/bidx10d), the number of Treg cells in peripheral blood of mice in group E was significantly lower than that in group D, but there were more Treg cells in tumor tissue of mice in group E than group D. Compared with the control mice, the overall immunity of the experimental group was inhibited. The twice-daily segmentation mode in the conventional dose radiotherapy group significantly improved Treg in peripheral blood and tumor immune microenvironment and promoted neovascularization. The twice-daily segmentation mode in the high-dose radiotherapy group improved the immunosuppression of lymphocytes in peripheral blood but increased immunosuppression in tumor microenvironment.

Segment-powered Control System of Linear Motor Based on Long-Mover Segmented Mode and Hybrid Switching

This paper proposes a segment-powered control system of a linear motor. It is based on long-mover segmented mode and hybrid switching for the segment-powered control requirements of a high-thrust permanent magnet linear synchronous motor. Firstly, the segment-powered principle of the long-mover segmented mode is analyzed. Then the design of the hybrid switching control system is completed. Finally, the experimental verification proves that the control system has good control stability and control accuracy, which can meet the segment-powered control requirements of a high-thrust linear motor.

Numerical investigation of multistage cavity shedding around a cavitating hydrofoil based on different turbulence models

Multistage shedding in cavity flow processes is a challenging and crucial topic in cavitating flows. This paper employs three different turbulence models to obtain a more comprehensive understanding of the physical details of multistage shedding in compressible cavitating flows and compare the differences between the turbulence models. The study employs the Schnerr-Sauer cavitation model in combination with the volume of fluid (VOF) method to capture the details of the unsteady cavitating flow around NACA66. The results are compared to experimental data and validate the numerical method. The study reveals that the primary shedding process is dominated by the re-entrant jet mechanism, whereas the pressure wave mechanism dominates the second shedding process. The second shedding process generates multiple pressure waves that disrupt the stable vortex structure of the sheet cavity. Furthermore, wavelet analysis indicates that the higher-order coherent structure mainly originates from the second shedding stage, and the second shedding process is shorter than the primary shedding process. The study proposes the segmented dynamic mode decomposition (SDMD), which obtains the mode characteristics at different time scales within one cycle. Under the RANS model, the small-scale cavity collapse and vortex structure cannot be observed during the second shedding process. However, the LES model can predict more realistic multi-scale flow field information.

Lesion size of early cerebral infarction on a Non-Contrast CT influences detection ability in Cascade Mask Region-Convolutional neural networks

BackgroundCurrently, convolutional neural network (CNN) methods for early non-contrast brain CT (NCCT) infarction detection have limited accuracy. We suspect this may be due to the significant scale variance of lesion size, which would result in small infarctions not being properly detected. Materials and MethodsPatients’ early NCCT infarction signs were labeled and divided into training/validation and test datasets. The 1st Cascade Mask Region-Convolutional Neural Network (Cascade Mask R-CNN) model (1st model) was train/validated. The test dataset, with added normal controls, was used to evaluate the performances. We further divided the lesions into large and small ones by statistics. We filtered the test dataset’s ‘small’ lesions and re-examined the 1st model. Finally, we filtered the ‘small’ lesions of the training/validation datasets to train/validate the 2nd Cascade Mask R-CNN (2nd model), and the performances were examined by the filtered test dataset. ResultsIn total, 266 ischemic stroke patients (Male/Female 163/103, age 69.9 ± 14.1 years) were included, and the duration from last known well to NCCT was 28.5 ± 39.3 h. Two hundred twenty patients were placed in training/validation datasets and 46 were assigned to the test dataset. Twenty normal control cases (Male/Female 14/6, age 68.7 ± 10.8 years) were added into the test dataset. The results were Sensitivity: 74.0%; Specificity: 91.8%; Accuracy: 86.1%; Precision: 80.9%; and F1-score: 77.3%. Based on the histogram, large lesions were > 2,012 pixels (4.8 cm2), and small lesions were ≤ 2,012 pixels. The 1st model’s performances by the filtered test dataset were Sensitivity: 96.9%; Specificity: 92.9%; Accuracy: 93.6%; Precision: 74.1%; and F1-score: 84.0%. The 2nd model’s performances were Sensitivity: 97.1%; Specificity: 96.4%; Accuracy: 96.5%; Precision: 85.7%; and F1-score: 91.0%. ConclusionsThe target objects in most benchmark databases do not have the property of significant scale variance, which means commonly developed AI models focus on the issue of translational variance. The conventional CNN-based AI models do have the property of translation equivariance. The impact of scale variance has rarely been explored in real applications. For patients with ischemic stroke, the lesion size may vary significantly. Experimental results showed that NCCT’s early infarction size can influence commonly used conventional CNN-based object detection and image segmentation modes, such as Cascade Mask R-CNN’s performance.

High resolution single-shot myocardial imaging using bSSFP with wave encoding.

Single-shot balanced steady-state free precession (bSSFP) sequence is widely used in cardiac imaging. However, the limited scan time in one heartbeat greatly hinders its spatial resolution compared to the segmented acquisition mode. Therefore, a highly accelerated single-shot bSSFP imaging technology is needed for clinical use. To develop and evaluate a wave-encoded bSSFP sequence with high acceleration rates for single-shot myocardial imaging. The proposed Wave-bSSFP method is implemented by adding a sinusoidal wave gradient in the phase encoding direction during the readout of bSSFP sequence. Uniform undersampling is used for acceleration. Its performance was first validated via phantom studies by comparison with conventional bSSFP. Then it was evaluated in volunteer studies via anatomical imaging, T2 -prepared bSSFP, and T1 mapping in in-vivo cardiac imaging. All methods were compared with accelerated conventional bSSFP reconstructed using iterative SENSE and compressed sensing (CS), to demonstrate the advantage of wave encoding in suppressing the noise amplification and artifacts induced by acceleration. The proposed Wave-bSSFP method achieved a high acceleration factor of 4 for single-shot acquisitions. The proposed method showed lower average g-factors than bSSFP, and fewer blurring artifacts than CS reconstruction. The Wave-bSSFP with R=4 achieved higher spatial and temporal resolutions compared with the conventional bSSFP with R=2 in several applications such as T2 -prepared bSSFP and T1 mapping, and could be applied in systolic imaging. Wave encoding can be used to highly accelerate 2D bSSFP imaging with single-shot acquisitions. Compared with the conventional bSSFP sequence, the proposed Wave-bSSFP method can effectively reduce the g-factor and aliasing artifacts in cardiac imaging.

Evolution of enhanced magmatism at the ultraslow spreading Southwest Indian Ridge between 46°E and 53.5°E

The Indomed-Gallieni supersegment (46-52°E) of the Southwest Indian Ridge (SWIR) has been of interest as a melt-rich endmember of ultraslow spreading mid-ocean ridges, attributed to enhanced magmatism since 11 Ma. This provides an excellent opportunity to study the evolution and the impact of enhanced magmatism at mid-ocean ridges. Here we combine shipboard bathymetry and gravity data over the SWIR 46-53.5°E. We illustrate that the enhanced magmatism initiated in the western part of the Indomed-Gallieni supersegment with decreasing crustal productions and no propagation. In the eastern part, the enhanced magmatism propagated to the east at decreasing propagating velocities, producing an increasingly thick crust. The present propagation tip is at 51.9°E, ∼30 km to the west of the Gallieni transform fault (TF). The most magmatically-robust area of the propagation has been located near 50.5°E since ∼6 Ma. We hypothesize that a mantle melting anomaly with heterogenous upwelling rates and/or geometry was responsible for the change of the enhanced magmatism. The enhanced magmatism plays significant roles in ridge segmentation and seafloor spreading modes. The degree of along-axis melt focusing was lowered, causing the segmentation to be typical of slow spreading ridges. As magmatism dramatically decreases over 160 km along the ridge axis from the most robust (producing a 9.5-km-thick crust at 50.5°E) to nearly-amagmatic near the Gallieni TF, this ridge section exhibits the full spectrum of ultraslow spreading modes, changing from magmatically-controlled (intense volcanism and almost no faulting) to tectonically-dominated (smooth ridges and extensive peridotites outcrops) modes.

An Adaptive CU Split Method for VVC Intra Encoding

In order to reduce the coding complexity and ensure that versatile video coding (H.266/VVC) effectively maintains the equilibrium between coding efficiency and quality in the coding process, an adaptive coding unit (CU) split method for intra encoding is proposed in this paper. First, the coding speed of CU division is accelerated by an iterative algorithm based on the interval relation of texture complexity energy. Then, according to the relationship between texture features and CU segmentation patterns, the Scharr edge gradient operator algorithm is used to describe the texture information and the threshold. Texture uniformity, directionality, and subblock discrepancy are analyzed and discussed, respectively. Finally, the optimal segmentation pattern is determined based on the texture information to adaptively skip unnecessary segmentation modes. The results show that the algorithm can save 45.2% of the overall coding time on average with an average cost of 0.51% Bjontegaard delta bit rate (BDBR). The algorithm ensures the coding quality while reducing the computational complexity.