Meta-learning based blind image super-resolution approach to different degradations.

Although recent studies on blind single image super-resolution (SISR) have achieved significant success, most of them typically require supervised training on synthetic low resolution (LR)-high resolution (HR) paired images. This leads to re-training necessity for different degradations and restricted applications in real-world scenarios with unfavorable inputs. In this paper, we propose an unsupervised blind SISR method with input underlying different degradations, named different degradations blind super-resolution (DDSR). It formulates a Gaussian modeling on blur degradation and employs a meta-learning framework for solving different image degradations. Specifically, a neural network-based kernel generator is optimized by learning from random kernel samples, referred to as random kernel learning. This operation provides effective initialization for blur degradation optimization. At the same time, a meta-learning framework is proposed to resolve multiple degradation modelings on the basis of alternative optimization between blur degradation and image restoration, respectively. Differing from the pre-trained deep-learning methods, the proposed DDSR is implemented in a plug-and-play manner, and is capable of restoring HR image from unfavorable LR input with degradations such as partial coverage, noise addition, and darkening. Extensive simulations illustrate the superior performance of the proposed DDSR approach compared to the state-of-the-arts on public datasets with comparable memory load and time consumption, yet exhibiting better application flexibility and convenience, and significantly better generalization ability towards multiple degradations. Our code is available at https://github.com/XYLGroup/DDSR.

Endoscopic resection for esophageal gastrointestinal stromal tumors: a multi-center feasibility study.

Background: Esophageal gastrointestinal stromal tumors (E-GISTs) are highly uncommon and have not been thoroughly examined. Objectives: The objective of this multi-center study was to assess the viability of endoscopic resection (ER) in the treatment of E-GISTs and to explore its clinical implications. Design: This was a multi-center retrospective study. Consecutive patients referred to the four participating centers. Methods: E-GISTs among the consecutive subepithelial tumors (SETs) treated by ER methods were enrolled from April 2019 to August 2022. Clinicopathological, endoscopic, and follow-up data were collected and analyzed. Results: A total of 23 patients with E-GISTs were included for analysis, accounting for 1.9% of all the esophageal SETs (1243 patients). The average size of the tumor lesions was 2.3 cm (range 1.0-4.0 cm). We observed that tumors larger than 2.0 cm were more likely to grow deeper, with a statistically significant difference (p < 0.001). End bloc resection was achieved in all 23 patients. The mean operation time was 53.6 min (range 25-111 min). One patient experienced significant intraoperative bleeding, which was promptly managed endoscopically without necessitating surgery. The average hospital stay was 4.5 days (range 3-8 days). The overall median follow-up period was 31 months (range 13-47 months). No tumor recurrence, residual tumor, distal metastasis, or death was observed during the follow-up period. Conclusion: Based on our limited data, our study indicates that ER may be a feasible and effective option for treating esophageal GISTs measuring 4 cm or less. We suggest submucosal tunnel endoscopic resection as the preferred approach, as all E-GISTs in our study were situated in the muscularis propria layer. Additionally, tumors larger than 2 cm were more prone to deeper growth or extraluminal extension.

Blind Super-Resolution Via Meta-Learning and Markov Chain Monte Carlo Simulation.

Learning based approaches have witnessed great successes in blind single image super-resolution (SISR) tasks, however, handcrafted kernel priors and learning based kernel priors are typically required. In this paper, we propose a Meta-learning and Markov Chain Monte Carlo based SISR approach to learn kernel priors from organized randomness. In concrete, a lightweight network is adopted as kernel generator, and is optimized via learning from the MCMC simulation on random Gaussian distributions. This procedure provides an approximation for the rational blur kernel, and introduces a network-level Langevin dynamics into SISR optimization processes, which contributes to preventing bad local optimal solutions for kernel estimation. Meanwhile, a meta-learning based alternating optimization procedure is proposed to optimize the kernel generator and image restorer, respectively. In contrast to the conventional alternating minimization strategy, a meta-learning based framework is applied to learn an adaptive optimization strategy, which is less-greedy and results in better convergence performance. These two procedures are iteratively processed in a plug-and-play fashion, for the first time, realizing a learning-based but plug-and-play blind SISR solution in unsupervised inference. Extensive simulations demonstrate the superior performance and generalization ability of the proposed approach when comparing with state-of-the-arts on synthesis and real-world datasets.

CrossHomo: Cross-Modality and Cross-Resolution Homography Estimation.

Multi-modal homography estimation aims to spatially align the images from different modalities, which is quite challenging since both the image content and resolution are variant across modalities. In this paper, we introduce a novel framework namely CrossHomo to tackle this challenging problem. Our framework is motivated by two interesting findings which demonstrate the mutual benefits between image super-resolution and homography estimation. Based on these findings, we design a flexible multi-level homography estimation network to align the multi-modal images in a coarse-to-fine manner. Each level is composed of a multi-modal image super-resolution (MISR) module to shrink the resolution gap between different modalities, followed by a multi-modal homography estimation (MHE) module to predict the homography matrix. To the best of our knowledge, CrossHomo is the first attempt to address the homography estimation problem with both modality and resolution discrepancy. Extensive experimental results show that our CrossHomo can achieve high registration accuracy on various multi-modal datasets with different resolution gaps. In addition, the network has high efficiency in terms of both model complexity and running speed.

Association of Circulating Carbohydrate Antigen 19-9 Level with Type 2 Diabetic Kidney Disease in Chinese Adults: A Cross-Sectional Study.

Objective: Very few and conflicting data are available regarding the correlation between circulating carbohydrate antigen 19-9 (CA19-9) levels and diabetic kidney disease (DKD) and its components including albuminuria and a low estimated glomerular filtration rate (eGFR). This study aimed to examine the association of circulating CA19-9 and DKD in Chinese patients with type 2 diabetes mellitus (T2DM). Methods: A total of 402 hospitalized T2DM patients between September 2017 and December 2021 were included in this cross-sectional study. There were 224 and 178 subjects in non-DKD and DKD groups, respectively. Serum CA19-9 was measured by chemiluminescence method, and its potential relationship with DKD was evaluated by multivariate logistic regression and correlation analyses, and receiver operating characteristic (ROC) curve analysis. Results: T2DM patients with DKD had significantly higher serum CA19-9 levels than those without, and serum CA19-9 levels were positively related to urinary albumin-to-creatinine ratio and negatively to eGFR (P<0.01). Multivariate regression analysis revealed that serum CA 19-9 was an independent factor of DKD [odds ratio (OR), 1.018; 95% confidence interval (CI), 1.002-1.035; P<0.05]. Moreover, an increased progressively risk of DKD with an increase in serum CA19-9 quartiles was observed (P for trend <0.001), and T2DM patients in the highest serum CA19-9 quartile were associated with an increased likelihood of DKD when compared to those in the lowest quartile (OR: 2.936, 95% CI 1.129-7.633, P<0.05). Last, the analysis of ROC curves suggested that serum CA 19-9 at a cut of 25.09 U/mL resulted in the highest Youden index with sensitivity 43.8% and 75.4% specificity to predict the presence of DKD. Conclusion: These results showed that high circulating CA19-9 was related to DKD and may serve as a useful biomarker of DKD in hospitalized Chinese T2DM patients.

Assessing Face Image Quality: A Large-Scale Database and a Transformer Method.

The amount of face images has been witnessing an explosive increase in the last decade, where various distortions inevitably exist on transmitted or stored face images. The distortions lead to visible and undesirable degradation on face images, affecting their quality of experience (QoE). To address this issue, this paper proposes a novel Transformer-based method for quality assessment on face images (named as TransFQA). Specifically, we first establish a large-scale face image quality assessment (FIQA) database, which includes 42,125 face images with diversifying content at different distortion types. Through an extensive crowdsource study, we obtain 712,808 subjective scores, which to the best of our knowledge contribute to the largest database for assessing the quality of face images. Furthermore, by investigating the established database, we comprehensively analyze the impacts of distortion types and facial components (FCs) on the overall image quality. Accordingly, we propose the TransFQA method, in which the FC-guided Transformer network (FT-Net) is developed to integrate the global context, face region and FC detailed features via a new progressive attention mechanism. Then, a distortion-specific prediction network (DP-Net) is designed to weight different distortions and accurately predict final quality scores. Finally, the experiments comprehensively verify that our TransFQA method significantly outperforms other state-of-the-art methods for quality assessment on face images.

Increased Systemic Immune-Inflammation Index Was Associated with Type 2 Diabetic Peripheral Neuropathy: A Cross-Sectional Study in the Chinese Population.

Background: Systemic immune-inflammation index (SII), a novel inflammatory marker, has been demonstrated to be associated with type 2 diabetes mellitus (T2DM) and its vascular complications, however, the relation between SII and diabetic peripheral neuropathy (DPN) has been never reported. We aimed to explore whether SII is associated with DPN in Chinese population. Methods: A cross-sectional study was conducted among 1460 hospitalized patients with T2DM. SII was calculated as the platelet count × neutrophil count/lymphocyte count, and its possible association with DPN was investigated by correlation and multivariate logistic regression analysis, and subgroup analyses. Results: Patients with higher SII quartiles had higher vibration perception threshold and prevalence of DPN (all P<0.01), and SII was independently positively associated with the prevalence of DPN (P<0.01). Multivariate logistic regression analysis showed that the risk of prevalence of DPN increased progressively across SII quartiles (P for trend <0.01), and participants in the highest quartile of SII was at a significantly increased risk of prevalent DPN compared to those in the lowest quartile after adjustment for potential confounding factors (odds rate: 1.211, 95% confidence intervals 1.045-1.404, P<0.05). Stratified analysis revealed positive associations of SII quartiles with risk of prevalent DPN only in men, people less than 65 years old, with body mass index <24 kg/m2, duration of diabetes >5 years, hypertension, dyslipidaemia, poor glycaemic control, and estimated glomerular filtration rate <90 mL/min/1.73 m2 (P for trend <0.01 or P for trend <0.05). The receiver operating characteristic curve analysis revealed that the optimal cut-off point of SII for predicting DPN was 617.67 in patients with T2DM, with a sensitivity of 45.3% and a specificity of 73%. Conclusion: The present study showed that higher SII is independently associated with increased risk of DPN, and SII might serve as a new risk biomarker for DPN in Chinese population.

Untargeted Metabolomics Reveals Alterations of Rhythmic Pulmonary Metabolism in IPF.

Idiopathic pulmonary fibrosis (IPF) is a chronic and progressive condition characterized by the impairment of alveolar epithelial cells. Despite continued research efforts, the effective therapeutic medication is still absent due to an incomplete understanding of the underlying etiology. It has been shown that rhythmic alterations are of significant importance in the pathophysiology of IPF. However, a comprehensive understanding of how metabolite level changes with circadian rhythms in individuals with IPF is lacking. Here, we constructed an extensive metabolite database by utilizing an unbiased reference system culturing with 13C or 15N labeled nutrients. Using LC-MS analysis via ESI and APCI ion sources, 1300 potential water-soluble metabolites were characterized and applied to evaluate the metabolic changes with rhythm in the lung from both wild-type mice and mice with IPF. The metabolites, such as glycerophospholipids and amino acids, in WT mice exhibited notable rhythmic oscillations. The concentrations of phospholipids reached the highest during the fast state, while those of amino acids reached their peak during fed state. Similar diurnal variations in the metabolite rhythm of amino acids and phospholipids were also observed in IPF mice. Although the rhythmic oscillation of metabolites in the urea cycle remained unchanged, there was a significant up-regulation in their levels in the lungs of IPF mice. 15N-ammonia in vivo isotope tracing further showed an increase in urea cycle activity in the lungs of mice with IPF, which may compensate for the reduced efficiency of the hepatic urea cycle. In sum, our metabolomics database and method provide evidence of the periodic changes in lung metabolites, thereby offering valuable insights to advance our understanding of metabolic reprogramming in the context of IPF.

Interpretable Multi-modal Image Registration Network Based on Disentangled Convolutional Sparse Coding.

Multi-modal image registration aims to spatially align two images from different modalities to make their feature points match with each other. Captured by different sensors, the images from different modalities often contain many distinct features, which makes it challenging to find their accurate correspondences. With the success of deep learning, many deep networks have been proposed to align multi-modal images, however, they are mostly lack of interpretability. In this paper, we first model the multi-modal image registration problem as a disentangled convolutional sparse coding (DCSC) model. In this model, the multi-modal features that are responsible for alignment (RA features) are well separated from the features that are not responsible for alignment (nRA features). By only allowing the RA features to participate in the deformation field prediction, we can eliminate the interference of the nRA features to improve the registration accuracy and efficiency. The optimization process of the DCSC model to separate the RA and nRA features is then turned into a deep network, namely Interpretable Multi-modal Image Registration Network (InMIR-Net). To ensure the accurate separation of RA and nRA features, we further design an accompanying guidance network (AG-Net) to supervise the extraction of RA features in InMIR-Net. The advantage of InMIR-Net is that it provides a universal framework to tackle both rigid and non-rigid multi-modal image registration tasks. Extensive experimental results verify the effectiveness of our method on both rigid and non-rigid registrations on various multi-modal image datasets, including RGB/depth images, RGB/near-infrared (NIR) images, RGB/multi-spectral images, T1/T2 weighted magnetic resonance (MR) images and computed tomography (CT)/MR images. The codes are available at https://github.com/lep990816/Interpretable-Multi-modal-Image-Registration.

Endoscopic Submucosal Dissection for Large Duodenal Laterally Spreading Lesions is Feasible: A Multi-Center Retrospective Study.

BACKGROUND AND AIMS: This retrospective study aimed to evaluate the effect and safety of endoscopic submucosal dissection (ESD) for large laterally spreading lesions located in the descending duodenum based on multi-center experiences. METHODS: This multicentric retrospective study included 3 hospitals in China. Fifty-one patients with laterally spreading lesions of the duodenum who underwent ESD between February 2019 and December 2020 were enrolled. The en bloc resection rates, en bloc R0 resection rates, complication rates, and local recurrence after ESD were evaluated. RESULTS: Of the 51 patients, the median age was 62 years old (ranging from 37 to 76 years old); among them, 29 were male and 22 were female. The average lesion size was 2.3 cm (ranging from 1.5 to 4.0 cm). All 51 lesions achieved en bloc R0 resection successfully, with the procedure time ranging from 20 to 117 min (median: 45.5 min). The hospital length of stay ranged from 4 to 90 days (median: 8.0 d). Two patients experienced delayed bleeding 3 days after ESD and 2 other patients were diagnosed with delayed perforation. Tumor residual and local recurrence did not occur during a short follow-up period. CONCLUSIONS: ESD for laterally spreading lesions of the descending duodenum is feasible.

Reciprocal GAN Through Characteristic Functions (RCF-GAN).

The integral probability metric (IPM) equips generative adversarial nets (GANs) with the necessary theoretical support for comparing statistical moments in an embedded domain of the critic, while stabilising their training and mitigating the mode collapse issues. For enhanced intuition and physical insight, we introduce a generalisation of IPM-GANs which operates by directly comparing probability distributions rather than their moments. This is achieved through characteristic functions (CFs), a powerful tool that uniquely comprises all information about any general distribution. For rigour, we first theoretically prove the ability of the CF loss to compare probability distributions, and proceed to establish the physical meaning of the phase and amplitude of CFs. An optimal sampling strategy is then developed to calculate the CFs, and an equivalence between the embedded and data domains is proved under the reciprocal theory. This makes it possible to seamlessly combine IPM-GAN with an auto-encoder structure by an advanced anchor architecture, which adversarially learns a semantic low-dimensional manifold for both generation and reconstruction. This efficient reciprocal CF GAN (RCF-GAN) structure, uses only two modules and a simple training strategy to achieve the state-of-the-art bi-directional generation. Experiments demonstrate the superior performance of RCF-GAN on both regular (images) and irregular (graph) domains.

Von Mises-Fisher Elliptical Distribution.

Modern probabilistic learning systems mainly assume symmetric distributions, however, real-world data typically obey skewed distributions and are thus not adequately modeled through symmetric distributions. To address this issue, a generalization of symmetric distributions called elliptical distributions are increasingly used, together with further improvements based on skewed elliptical distributions. However, existing approaches are either hard to estimate or have complicated and abstract representations. To this end, we propose a novel approach based on the von-Mises-Fisher (vMF) distribution to obtain an explicit and simple probability representation of skewed elliptical distributions. The analysis shows that this not only allows us to design and implement nonsymmetric learning systems but also provides a physically meaningful and intuitive way of generalizing skewed distributions. For rigor, the proposed framework is proven to share important and desirable properties with its symmetric counterpart. The proposed vMF distribution is demonstrated to be easy to generate and stable to estimate, both theoretically and through examples.

Metalearning-Based Alternating Minimization Algorithm for Nonconvex Optimization.

In this article, we propose a novel solution for nonconvex problems of multiple variables, especially for those typically solved by an alternating minimization (AM) strategy that splits the original optimization problem into a set of subproblems corresponding to each variable and then iteratively optimizes each subproblem using a fixed updating rule. However, due to the intrinsic nonconvexity of the original optimization problem, the optimization can be trapped into a spurious local minimum even when each subproblem can be optimally solved at each iteration. Meanwhile, learning-based approaches, such as deep unfolding algorithms, have gained popularity for nonconvex optimization; however, they are highly limited by the availability of labeled data and insufficient explainability. To tackle these issues, we propose a meta-learning based alternating minimization (MLAM) method that aims to minimize a part of the global losses over iterations instead of carrying minimization on each subproblem, and it tends to learn an adaptive strategy to replace the handcrafted counterpart resulting in advance on superior performance. The proposed MLAM maintains the original algorithmic principle, providing certain interpretability. We evaluate the proposed method on two representative problems, namely, bilinear inverse problem: matrix completion and nonlinear problem: Gaussian mixture models. The experimental results validate the proposed approach outperforms AM-based methods.

Association of systemic immune-inflammation index with diabetic kidney disease in patients with type 2 diabetes: a cross-sectional study in Chinese population.

Objective: Systemic immune-inflammation index (SII), a novel inflammatory marker, has been reported to be associated with diabetic kidney disease (DKD) in the U.S., however, such a close relationship with DKD in other countries, including China, has not been never determined. We aimed to explore the association between SII and DKD in Chinese population. Methods: A total of 1922 hospitalized patients with type 2 diabetes mellitus (T2DM) included in this cross-sectional study were divided into three groups based on estimated glomerular filtration rate (eGFR) and urinary albumin-to-creatinine ratio (ACR): non-DKD group, DKD stages 1-2 Alb group, and DKD-non-Alb+DKD stage 3 Alb group. The possible association of SII with DKD was investigated by correlation and multivariate logistic regression analysis, and receiver-operating characteristic (ROC) curves analysis. Results: Moving from the non-DKD group to the DKD-non-Alb+DKD stage 3 Alb group, SII level was gradually increased (P for trend <0.01). Partial correlation analysis revealed that SII was positively associated with urinary ACR and prevalence of DKD, and negatively with eGFR (all P<0.01). Multivariate logistic regression analysis showed that SII remained independently significantly associated with the presence of DKD after adjustment for all confounding factors [(odds ratio (OR), 2.735; 95% confidence interval (CI), 1.840-4.063; P < 0.01)]. Moreover, compared with subjects in the lowest quartile of SII (Q1), the fully adjusted OR for presence of DKD was 1.060 (95% CI 0.773-1.455) in Q2, 1.167 (95% CI 0.995-1.368) in Q3, 1.266 (95% CI 1.129-1.420) in the highest quartile (Q4) (P for trend <0.01). Similar results were observed in presence of DKD stages 1-2 Alb or presence of DKD-non- Alb+DKD stage 3 Alb among SII quartiles. Last, the analysis of ROC curves revealed that the best cutoff values for SII to predict DKD, Alb DKD stages 1- 2, and DKD-non-Alb+ DKD stage 3 Alb were 609.85 (sensitivity: 48.3%; specificity: 72.8%), 601.71 (sensitivity: 43.9%; specificity: 72.3%), and 589.27 (sensitivity: 61.1%; specificity: 71.1%), respectively. Conclusion: Higher SII is independently associated with an increased risk of the presence and severity of DKD, and SII might be a promising biomarker for DKD and its distinct phenotypes in Chinese population.

Knowledge and awareness of nicotine, nicotine replacement therapy, and electronic cigarettes among general practitioners with a special interest in respiratory medicine in China.

Objectives: This study aimed to assess the knowledge and awareness of nicotine, nicotine replacement therapy (NRT), and electronic cigarettes (e-cigarettes) among general practitioners with a special interest (GPwSIs) in respiratory medicine. Methods: A cross-sectional study was conducted from October 2021 to February 2022. Knowledge and awareness were compared among smokers and non-smokers, as well as different age and gender groups. Results: The study consisted of 102 GPwSIs from 21 cities in Sichuan Province, China. Most respondents would recommend NRT for long-term use. Only a few believed that e-cigarettes are an effective means of smoking cessation and 71.6% would not recommend e-cigarettes as a substitute for cigarettes to their patients. Additionally, the majority did not regularly provide extensive help to assist patients in quitting smoking and needed smoking cessation counseling training. Conclusion: GPwSIs in respiratory medicine in China could have a relatively low level of knowledge and awareness regarding nicotine, NRT, and e-cigarettes. The study highlights the need for smoking cessation training among GPwSIs to improve their knowledge and provide better assistance to patients who want to quit smoking.

Submucosal Tunneling Endoscopic Resection for Submucosal Tumors in the Proximal Esophagus.

BACKGROUND: Submucosal tunneling endoscopic resection (STER) is widely applied for treatment of gastrointestinal submucosal tumors (SMTs) originating from the muscularis propria layer. However, the tumor location within the proximal esophagus makes STER a challenge for the endoscopists. The aim of this study was to summarize the technique skill and evaluate the outcomes of proximal esophageal STER. STUDY DESIGN: A total of 72 patients with SMTs in the proximal esophagus undergoing STER were included from February 2019 to March 2021. Imaging 3-dimensional reconstruction was used for patients with large SMTs. Clinicopathological, endoscopic, and follow-up data were collected and analyzed. RESULTS: In this study, all the tumors were removed completely and no gross disease was remaining. The en bloc resection was achieved in 90.28% of patients, and the complications rate was 6.95%. Three-dimensional reconstruction was used for 30 patients (41.67%) with large SMTs (transverse diameter >2.0 cm). Based on statistical analysis, tumors with irregular shape and larger size were the significant contributors to piecemeal resection. Larger tumors increase the risk of long operation time, and irregular tumor shapes increase the risk of complications. The median hospitalization time was 4 days. All of the complications were cured by conservative treatment. A median follow-up of 12 months was available, and all patients were free from local recurrence or distant metastasis during the study period. CONCLUSIONS: STER is an effective and safe methodology for the resection of proximal esophageal SMTs. Tumor size and shape mainly impact the piecemeal resection rate, STER-related complications, and procedural difficulty.

Blind VQA on 360° Video via Progressively Learning from Pixels, Frames and Video.

Blind visual quality assessment (BVQA) on 360° video plays a key role in optimizing immersive multimedia systems. When assessing the quality of 360° video, human tends to perceive its quality degradation from the viewport-based spatial distortion of each spherical frame to motion artifact across adjacent frames, ending with the video-level quality score, i.e., a progressive quality assessment paradigm. However, the existing BVQA approaches for 360° video neglect this paradigm. In this paper, we take into account the progressive paradigm of human perception towards spherical video quality, and thus propose a novel BVQA approach (namely ProVQA) for 360° video via progressively learning from pixels, frames and video. Corresponding to the progressive learning of pixels, frames and video, three sub-nets are designed in our ProVQA approach, i.e., the spherical perception aware quality prediction (SPAQ), motion perception aware quality prediction (MPAQ) and multi-frame temporal non-local (MFTN) sub-nets. The SPAQ sub-net first models the spatial quality degradation based on spherical perception mechanism of human. Then, by exploiting motion cues across adjacent frames, the MPAQ sub-net properly incorporates motion contextual information for quality assessment on 360° video. Finally, the MFTN sub-net aggregates multi-frame quality degradation to yield the final quality score, via exploring long-term quality correlation from multiple frames. The experiments validate that our approach significantly advances the state-of-the-art BVQA performance on 360° video over two datasets, the code of which has been public in https://github.com/yanglixiaoshen/ProVQA.

Phase Formation, Microstructure, and Magnetic Properties of Nd14.5Fe79.3B6.2 Melt-Spun Ribbons with Different Ce and Y Substitutions.

Phase formation and microstructure of (Nd1-2xCexYx)14.5Fe79.3B6.2 (x = 0.05, 0.10, 0.15, 0.20, 0.25) alloys were studied experimentally. The results reveal that (Nd1-2xCexYx)14.5Fe79.3B6.2 annealed alloys show (NdCeY)2Fe14B phase with the tetragonal Nd2Fe14B-typed structure (space group P42/mnm) and rich-RE (α-Nd) phase, while (Nd1-2xCexYx)14.5Fe79.3B6.2 ribbons prepared by melt-spun technology are composed of (NdCeY)2Fe14B phase, α-Nd phase and α-Fe phase, except for the ribbon with x = 0.25, which consists of additional CeFe2 phase. On the other hand, magnetic properties of (Nd1-2xCexYx)14.5Fe79.3B6.2 melt-spun ribbons were measured by a vibrating sample magnetometer (VSM). The measured results show that the remanence (Br) and the coercivity (Hcj) of the melt-spun ribbons decrease with the increase of Ce and Y substitutions, while the maximum magnetic energy product ((BH)max) of the ribbons decreases and then increases. The tendency of magnetic properties of the ribbons could result from the co-substitution of Ce and Y for Nd in Nd2Fe14B phase and different phase constitutions. It was found that the Hcj of the ribbon with x = 0.20 is relatively high to be 9.01 kOe, while the (BH)max of the ribbon with x = 0.25 still reaches to be 9.06 MGOe. It suggests that magnetic properties of Nd-Fe-B ribbons with Ce and Y co-substitution could be tunable through alloy composition and phase formation to fabricate novel Nd-Fe-B magnets with low costs and high performance.

A Universal Framework for Learning the Elliptical Mixture Model.

Mixture modeling using elliptical distributions promises enhanced robustness, flexibility, and stability over the widely employed Gaussian mixture model (GMM). However, existing studies based on the elliptical mixture model (EMM) are restricted to several specific types of elliptical probability density functions, which are not supported by general solutions or systematic analysis frameworks; this significantly limits the rigor in the design and power of EMMs in applications. To this end, we propose a novel general framework for estimating and analyzing the EMMs, achieved through the Riemannian manifold optimization. First, we investigate the relationships between Riemannian manifolds and elliptical distributions, and the so established connection between the original manifold and a reformulated one indicates a mismatch between these manifolds, a major cause of failure of the existing optimization for solving general EMMs. We next propose a universal solver that is based on the optimization of a redesigned cost and prove the existence of the same optimum as in the original problem; this is achieved in a simple, fast and stable way. We further calculate the influence functions of the EMM as theoretical bounds to quantify robustness to outliers. Comprehensive numerical results demonstrate the ability of the proposed framework to accommodate EMMs with different properties of individual functions in a stable way and with fast convergence speed. Finally, the enhanced robustness and flexibility of the proposed framework over the standard GMM are demonstrated both analytically and through comprehensive simulations.