Search

Male breast cancer (MBC) is a rare disease. Due to limited information, MBC has always been understudied. We conducted a retrospective population-based cohort study by using the 2010–2014 Surveillance, Epidemiology, and End Results (SEER) program data. The clinical and biological features of female breast cancer (FBC) patients were compared with MBC patients. Survival curves were constructed with the Kaplan-Meier method. Multivariate Cox regression models and competitive risk models were separately built to identify factors associated with survival time in the MBC and FBC group. Our retrospective study showed that MBC patients suffered with higher TNM stages, higher grades, and more percent of hormone receptor–positive tumors, compared with FBC patients. In addition, the site of the primary breast cancer varied greatly different between genders. FBC patients demonstrated superior overall survival than MBC patients on Kaplan-Meier analysis. In multivariate COX analysis for FBC patients, older age, black race, higher T, N, M-stages, higher grades, estrogen receptor (ER)/progesterone receptor (PR) negative and human epidermal growth factor receptor 2 (HER2) negative were associated with higher risk of death. MBC patients had similar independent prognostic factors, but PR and HER-2 status did not appear to independently influence survival. Interestingly, primary site location was an independent prognostic factor for both MBC patients and FBC patients, which should be considered by clinicians as a prognostic factor.

Producing the embedding of a sentence in an unsupervised way is valuable to natural language matching and retrieval problems in practice. In this work, we conduct a thorough examination of pretrained model based unsupervised sentence embeddings. We study on four pretrained models and conduct massive experiments on seven datasets regarding sentence semantics. We have there main findings. First, averaging all tokens is better than only using [CLS] vector. Second, combining both top andbottom layers is better than only using top layers. Lastly, an easy whitening-based vector normalization strategy with less than 10 lines of code consistently boosts the performance.

© 1965-2012 IEEE. This paper introduces an elemental operator with biaxial anisotropy to simulate the scalar hysteresis phenomenon of the ferromagnetic materials. The equilibrium position of the magnetization for the elemental operator can be determined by energy minimization. To directly describe the magnetic properties of each operator, an improved analytical expression is deduced by the partial approximate substitutions. Moreover, this approach utilizes the concept of distribution function density in order to consider the interaction field and the coercive force of the elemental operators. To verify the presented model, the magnetic hysteresis of two different magnetic materials under alternating excitations is measured by the magnetic property measurement system and calculated by this elemental operator method, respectively. The comparisons suggest that this elemental operator is effective and can be a useful tool to simulate the scalar magnetic properties of ferromagnetic materials.

© 2018, Electrical Technology Press Co. Ltd. All right reserved. The soft magnetic composite materials have undergone significant development due to their unique advantages such as low eddy current loss, quasi-isotropy of mechanical and magnetic properties, great design flexibility, low cost and low material consumption during the production process. These materials are suitable for some special electromagnetic devices with 3-D magnetic flux paths, such as transverse flux, claw pole, and axial flux permanent magnet motors. Since the soft magnetic composite materials have been widely used with satisfaction, the magnetic properties, i.e., major hysteresis loops, minor hysteresis loops, and reversal curves, need to be fully understood for developing high performance electromagnetic devices. In this paper, the limiting loop method based on the Preisach model is introduced to simulate the magnetic properties of soft magnetic composite material. A new formulation of the normal Preisach model was derived based on a graphical description of the Preisach theory of magnetic hysteresis. With the help of the Preisach diagrams for the limiting hysteresis loop, the difficulty of identifying the elementary dipole distribution function was circumvented. The new parameter identification method only requires the limiting hysteresis loop as the input data. Besides, to simulate the magnetic properties under different stress, the effect of stress on the magnetization process is taken into account in the improved limiting loop method. Finally, to verify the proposed model, the magnetic properties of SOMALOYTM 500, a classical type of SMC material, are simulated and compared with the experimental results. The accuracy and effectiveness of the model are validated by the actual measurement.

© 2018, Electrical Technology Press Co. Ltd. All right reserved The paper introduces a vector elemental operator with biaxial anisotropy based on the underlying magnetization mechanisms of the magnetic material. Detailed analysis of the magnetic properties of one elemental operator under rotational magnetization has been presented thereafter. With the help of the proposed vector elemental operator, the rotational magnetic hysteresis of soft magnetic composite material is simulated. The simulation results are compared with the experimental results obtained by the 3-D magnetic property measurement system. The good agreement shows the validity and practicability of the elemental operator method in the modelling of the rotational magnetic hysteresis.

BACKGROUND Marfan syndrome (MFS) is an autosomal dominant disease caused by mutations in the Fibrillin (FBN)1 gene and characterized by disorders in the cardiovascular, skeletal, and visual systems. The diversity of mutations and phenotypic heterogeneity of MFS make prenatal molecular diagnoses difficult. In this study, we used pre-implantation genetic diagnosis (PGD) to identify the pathogenic mutation in a male patient with MFS and to determine whether his offspring would be free of the disease. MATERIAL AND METHODS The history and pedigree of the proband were analyzed. Mutation analysis was performed on the couple and immediate family members. The couple chose IVF treatment and 4 blastocysts were biopsied. PGD was carried out by targeted high-throughput sequencing of the FBN1 gene in the embryos, along with single-nucleotide polymorphism haplotyping. Sanger sequencing was used to confirm the causative mutation. RESULTS c.2647T>C (p.Trp883Arg) was identified as the de novo likely pathogenic mutation in the proband. Whole-genome amplification and sequencing of the 3 embryos revealed that they did not carry the mutation, and 1 blastocyst was transferred back to the uterus. The amniocentesis test result analyzed by Sanger sequencing confirmed the PGD. A premature but healthy infant free of heart malformations was born. CONCLUSIONS The de novo mutation c.2647T>C (p.Trp883Arg) in FBN1 was identified in a Chinese patient with MFS. Embryos without the mutation were identified by PGD and resulted in a successful pregnancy.

© 2016 IEEE. This paper simulates the melting process and liquid titanium droplet formation in electromagnetic levitation melting (EML) process by using the finite element method. The modelling of the physical EML process involves the coupling of electromagnetic, thermal, and fluid flow fields. The bidirectional interactions of different fields are also consided. The electrical parameters, such as frequency, current, and number of turns of the excitation coils, which might affect the size of droplet, are studied. The arbitrary Lagrangian-Eulerian formulation is employed to model the deformation of metal sample during the melting process. The results show that the method adopted in this paper could dynamically predict the melting process, and size of the droplet could be controlled by modifying the electrical parameters.

© 2015 AIP Publishing LLC. This paper presents a new Preisach type hysteresis model for the high temperature superconductor. This model requires only the limiting hysteresis loop as the input data, and for this model, the limiting hysteresis loop is first separated into two limiting M-H loops based on the mechanisms, which can then be modeled by two separate modified Preisach algorithms. The area integrations of the Preisach distribution functions are determined only based on the limiting M-H loops. The validity and accuracy of this model is confirmed by comparing the simulation and experiment results of Bi-2223 and YBa2Cu3Ox superconducting tapes with external magnetic fields.

BSIM4 model is not suitable for distortion analysis of circuits with bias point VDS = 0, such as passive mixers and RF switches, due to discontinuities in higher order derivatives. Surface potential based PSP model has continuous derivatives to at least third order, and is therefore suitable for intermodulation products simulation. This paper presents the BSIM4 to PSP model conversion flow and comparison of passive mixer simulation results. Simulation results show good agreement of original BSIM4 and converted PSP model CV, IV, compression point, conversion loss and noise figure, which validates the conversion flow. The converted model has a correct IM3 slope of 3, allowing the simulation of intermodulation products, which was not possible with the original BSIM4 model.

This paper proposes an improved extended finite element method (XFEM) for modeling electromagnetic devices with multiple nearby geometrical interfaces. In regions near these interfaces, the magnetic vector potential approximation is enriched by incorporat-ing multiple derivative discontinuous fields based on the partition of unity method such that the interfaces can be represented inde-pendent of the mesh. The support of a node or an element can be cut by several interfaces. This method results in the high accuracy in the approximation field and the derivative field. Numerical examples applied to the iron core in 1D eddy current field involving level set based parts, error analysis and electromagnetic field computations are provided to demonstrate the utility of the proposed approach.

This paper reports the results of thermal diffusivity of kiwifruit measured using the modified Fitch method. The three unique structural components of the fruit, i.e. skin, flesh and core, were used in the measurement. The thermal diffusivities of skin, flesh and core are 0.78±0.08 x 10‐7, 1.14±0.06 x 10‐7 and 0.95±0.02 x 10‐7 (m2s‐1) respectively.

Objectives.: To assess willingness to participate (WTP) in hypothetical Phase III preventive HIV vaccine trials, and the impact of trial attributes on WTP, among low socioeconomic, ethnically diverse adults from communities at elevated risk for HIV infection. Method.: Participants (n = 123; median age = 38; 69% male; 37% Latino; 14% African-American) were recruited in Los Angeles in 2003 using multi-site, venue-based sampling. WTP was assessed for eight hypothetical HIV vaccine trials that varied across seven dichotomous attributes, using a 27-4fractional factorial experimental design. Individual-specific impact of vaccine trial attributes on WTP was estimated using within-individual ANOVA and then meta-analyzed across individuals. Results.: Mean WTP for eight hypothetical vaccine trials ranged from 1.74 to 3.81 (1 = highly unlikely, 5 = highly likely). Lower WTP was associated with vaccine-induced infection risk (impact = 0.88, p < 0.0001), false HIV-positives (0.53, p < 0.0001), no provision of free HIV medications (0.52, p < 0.0001), and longer trial duration (0.27; p = 0.0002). Conclusion.: HIV vaccine trial attributes may strongly influence WTP. Although existing candidate vaccines cannot cause HIV infection, perceptions of risk may impede WTP. Eliciting trial preferences and concerns prior to trial implementation may enable accommodation of participant preferences and support tailored interventions to address concerns and misconceptions to facilitate enrollment in safe and ethical trials among vulnerable communities. © 2007 Elsevier Inc. All rights reserved.

Ferroelectric/metal composites have been characterized for their electric field dependence of the dielectric and ferroelectric properties. The composites show an interesting change of dielectric properties with increasing metal concentration. A relatively high dielectric constant, up to four times as high as that of monolithic lead zirconate titanate (PZT), is observed in the investigated composition range. The dielectric properties are clearly electric field dependent. The tunability of the composite, defined as the difference between the dielectric constant with no applied voltage and the dielectric constant with an applied voltage, divided by the dielectric constant with no applied voltage, was found to be ∼ 17, whilst for the monolithic PZT ceramics it is about 3 only. This offers perspectives for application of these composites in lighting electronics and voltage controllers. Ferroelectric hysteresis measurements on these composites revealed a square-shaped hysteresis loop with remnant polarization comparable with single PZT ceramics, while the coercive field was much lower than that of pure PZT ceramics. This might be interesting for application of these composites as a ferroelectric memory material with decreased switching voltage.

A new experimental set-up controlled by computer has been made to measure the pyroelectric charge of ferroelectric materials with a relatively high conductivity at slow rates of temperature variation. It allowed us to obtain the polarisation vs. temperature behaviour of PZT with various compositions in the whole range of temperatures with a high accuracy. The ferroelectric Zr-rich region of the PZT phase diagram, where the isostructural phase transition FRL-FRH takes place, has been studied., Comment: 3 pages postscript file. Published 1998. Late inclusion in the archives due to journal diffusion limitations

In this paper, we introduce XGLUE, a new benchmark dataset that can be used to train large-scale cross-lingual pre-trained models using multilingual and bilingual corpora and evaluate their performance across a diverse set of cross-lingual tasks. Comparing to GLUE(Wang et al., 2019), which is labeled in English for natural language understanding tasks only, XGLUE has two main advantages: (1) it provides 11 diversified tasks that cover both natural language understanding and generation scenarios; (2) for each task, it provides labeled data in multiple languages. We extend a recent cross-lingual pre-trained model Unicoder(Huang et al., 2019) to cover both understanding and generation tasks, which is evaluated on XGLUE as a strong baseline. We also evaluate the base versions (12-layer) of Multilingual BERT, XLM and XLM-R for comparison.

Recent research demonstrates the effectiveness of using pretrained language models (PLM) to improve dense retrieval and multilingual dense retrieval. In this work, we present a simple but effective monolingual pretraining task called contrastive context prediction (CCP) to learn sentence representation by modeling sentence level contextual relation. By pushing the embedding of sentences in a local context closer and pushing random negative samples away, different languages could form isomorphic structure, then sentence pairs in two different languages will be automatically aligned. Our experiments show that model collapse and information leakage are very easy to happen during contrastive training of language model, but language-specific memory bank and asymmetric batch normalization operation play an essential role in preventing collapsing and information leakage, respectively. Besides, a post-processing for sentence embedding is also very effective to achieve better retrieval performance. On the multilingual sentence retrieval task Tatoeba, our model achieves new SOTA results among methods without using bilingual data. Our model also shows larger gain on Tatoeba when transferring between non-English pairs. On two multi-lingual query-passage retrieval tasks, XOR Retrieve and Mr.TYDI, our model even achieves two SOTA results in both zero-shot and supervised setting among all pretraining models using bilingual data.