CentroidNet：轻量快速的乳腺癌Ki67细胞核中心点检测模型

目的 Ki67分数是乳腺癌预后评估的重要指标,计算该分数的关键步骤是检测阴性与阳性癌细胞核。人工检测面临疲劳与主观差异的问题。卷积神经网络有望实现高质量、自动化的细胞核检测,然而需要病理专家为其标注细胞核。为了减轻标注的工作量,不少研究者提出以中心点标注训练卷积神经网络。然而这些方法采用过于复杂的卷积神经网络和后处理流程,未能充分提高易用性和效率、发挥卷积神经网络的质量。对此,提出CentroidNet模型,旨在提高中心点检测的质量、效率和易用性。方法 CentroidNet模型在图像上放置均匀排布的锚点,为每个锚点预测一个候选点,一部分候选点通过基于阈值的筛选策略成为预测点。本文提出最近锚点匹配策略用于生成训练标签,既保证了端到端推理,又规避了其他一对一标签匹配算法所具有的标签抖动问题。本文建议锚点间距应尽可能接近训练集答案点间最短距离的第一百分位数,并指出这样的锚点间距能够在前景标签数、坐标回归难度与效率之间取得良好的平衡。本文在设计卷积神经网络的结构时,没有采纳广为使用的U-Net或特征金字塔（feature pyramid network, FPN）中的多级上采样与旁路连接,反...     

Study of high-pressure and high-temperature behaviors and α-to-β phase transition of forsterite by first-principles and quasi-harmonic Debye model

The high-pressure and/or high-temperature behavior and α/β phase transition of forsterite (Mg₂SiO₄) were investigated by first principle calculations: local density approximation (LDA), generalized gradient approximation (GGA) and the quasi-harmonic Debye model in consideration of the temperature effect. At zero pressure and zero temperature, the calculated values of lattice parameters and elastic constants are well concordant with experimental data. The calculated values of cell volume at different pressures and temperatures are compatible with experiments. The results calculated by LDA are more concordant with the experimental results than those by GGA for describing Mg₂SiO₄ crystals. The boundaries of the P-T plots for α/β phase transition are calculated to be 0.00779 GPa/K by LDA and 0.00772 GPa/K by GGA, suggesting the upper and lower limits of the phase boundary for α/β forsterite. At temperature of less than 1000 K, the calculated results agree reasonably with experimental ones. Thus, the present results indicate that combination of first-principles and quasi-harmonic Debye model is an efficient approach to simulate the behavior of minerals at high pressure and/or high temperature.     

基于贝叶斯推理与PFNET理论的认知模型

针对从固定认知结构中生成认知模型的局限性,提出在认知元素固定而认知结构不固定的学习中使用贝叶斯推理方法和PFNET理论从以往学习者的样本信息中按学习者的要求生成"最佳Ki结点集合"和"最优Ki认知链",获得B-P认知模型;对B-P认知模型的生成原理进行说明并通过实例验证该模型的有效性和可行性。     

Prognostic significance of interim ~(18)F-FDG PET/CT SUV reduction associated with Ki67 in patients with diffuse large B-cell lymphoma

To detect the prognostic significance of interim F-FDG PET/CT SUV(standard uptake value) reduction(△SUV_(max)) associated with Ki67 in patients with diffuse large B cell lymphoma(DLBCL).47 DLBCL patients underwent PET/CT before initiation and after 2-4 cycles of chemotherapy were included.The SUV_(max)of the dominant lesions were calculated.Ki67 positive indices were provided by enzyme-labeled immunohistochemistry.SPSS17.0 was used for statistical analysis.ASUV_(max) of different groups were compared by t test.Receiver-operator characteristic analysis was performed to determine the optimal cutoff values.KaplanMeier analyses of PFS(Progression-free survival) were compared using log-rank test.The average of △SUV_(max)and △SUV_(max)%were 11.53 and 69.10%,respectively.The optimal cutoff values of △SUV_(max)x and △SUV_(max)%were 11.45 and 82.92%,respectively.Higher △SUV_(max) and △SUV_(max)%indicated longer PFS(p 0.001).The optimal cutoff value of Ki67 was 55%.Ki6755%was revealed to be an indicator of shorter PFS(p = 0.019).Either △SUV_(max)≤11.45 or Ki6755%was defined as an indicator to poor outcome and scored 1 point.The PFS rate was 100%in patients scored 0 point,yet 0%in patients scored 2 points.PFS tended to be shorter along with the score getting higher(p = 0.006).△SUV_(max) and Ki67 positive index were both of significance in DLBCL prognosis.The prognostic value may be confirmed when △SUV_(max) was accordant with Ki67.     

Ki67 and doublecortin positive cells in the human prefrontal cortices of normal aging and vascular dementia

Immunohistochemical localizing of the proliferation of Ki67 nuclei and doublecortin positive cells were performed in the prefrontal cortex of normal aged and vascular dementia (multiple infarct dementia) patients. Positive Ki67 nuclei and doublecortin positive cells were observed in both groups, with slightly higher density in the prefrontal cortex of vascular dementia. When the Ki67 sites were superimposed with the neuronal specific enolase localizations, only about 5% of the cells was doubly labeled, indicating few proliferating cells were neurons. This percentage did not vary between specimens of normal aging and those of vascular dementia.     

Lysophosphatidic Acid Mediates Imiquimod-Induced Psoriasis-like Symptoms by Promoting Keratinocyte Proliferation through LPAR1/ROCK2/PI3K/AKT Signaling Pathway

Psoriasis is a chronic inflammatory skin disease. Recently, lysophosphatidic acid (LPA)/LPAR5 signaling has been reported to be involved in both NLRP3 inflammasome activation in macrophages and keratinocyte activation to produce inflammatory cytokines, contributing to psoriasis pathogenesis. However, the effect and molecular mechanisms of LPA/LPAR signaling in keratinocyte proliferation in psoriasis remain unclear. In this study, we investigated the effects of LPAR1/3 inhibition on imiquimod (IMQ)-induced psoriasis-like mice. Treatment with the LPAR1/3 antagonist, ki16425, alleviated skin symptoms in IMQ-induced psoriasis-like mouse models and decreased keratinocyte proliferation in the lesion. It also decreased LPA-induced cell proliferation and cell cycle progression via increased cyclin A2, cyclin D1, cyclin-dependent kinase (CDK)2, and CDK4 expression and decreased p27^Kip1 expression in HaCaT cells. LPAR1 knockdown in HaCaT cells reduced LPA-induced proliferation, suppressed cyclin A2 and CDK2 expression, and restored p27^Kip1 expression. LPA increased Rho-associated protein kinase 2 (ROCK2) expression and PI3K/AKT activation; moreover, the pharmacological inhibition of ROCK2 and PI3K/AKT signaling suppressed LPA-induced cell cycle progression. In conclusion, we demonstrated that LPAR1/3 antagonist alleviates IMQ-induced psoriasis-like symptoms in mice, and in particular, LPAR1 signaling is involved in cell cycle progression via ROCK2/PI3K/AKT pathways in keratinocytes.     

电信智能卡集中管理支撑方案探讨

结合中国电信运营商智能卡的业务现状,从智能卡制卡方式、支撑架构、支撑流程等方面论述了电信智能卡的集中支撑方案,最后给出了集中式管理带来的其他问题。     

Digital Image Analysis Applied to Tumor Cell Proliferation,Aggressiveness, and Migration-Related Protein Synthesis in Neuroblastoma 3D Models

Patient-derived cancer 3D models are a promising tool that will revolutionize personalized cancer therapy but that require previous knowledge of optimal cell growth conditions and the most advantageous parameters to evaluate biomimetic relevance and monitor therapy efficacy. This study aims to establish general guidelines on 3D model characterization phenomena, focusing on neuroblastoma. We generated gelatin-based scaffolds with different stiffness and performed SK-N-BE(2) and SH-SY5Y aggressive neuroblastoma cell cultures, also performing co-cultures with mouse stromal Schwann cell line (SW10). Model characterization by digital image analysis at different time points revealed that cell proliferation, vitronectin production, and migration-related gene expression depend on growing conditions and are specific to the tumor cell line. Morphometric data show that 3D in vitro models can help generate optimal patient-derived cancer models, by creating, identifying, and choosing patterns of clinically relevant artificial microenvironments to predict patient tumor cell behavior and therapeutic responses.     

Relativistic three-quark bound states in separable two-quark approximation

Baryons as relativistic bound states in 3-quark correlations are described by an effective Bethe-Salpeter equation when irreducible 3-quark interactions are neglected and separable 2-quark correlations are assumed. We present an efficient numerical method to calculate the nucleon mass and its covariant wave function in this quantum field theoretic quark-diquark model with quark-exchange interaction. Expanding the components of the spinorial wave function in terms of Chebyshev polynomials, the four-dimensional integral equations are in a first step reduced to a coupled set of one-dimensional ones. This set of linear and homogeneous equations defines a generalised eigenvalue problem. Representing the eigenvector corresponding to the largest eigenvalue, the Chebyshev moments are then obtained by iteration. The nucleon mass is implicitly determined by the eigenvalue, and its covariant wave function is reconstructed from the moments within the Chebyshev approximation.     

The 3D Amlouk-Boubaker expansivity-energy gap-Vickers hardness abacus: A new tool for optimizing semiconductor thin film materials

In this study, we propose a new abacus based on a synthetic parameter: the Amlouk-Boubaker expansivity ψ_AB. This abacus gathers the band gap energy E_g., Vickers microhardness and ψ_AB. This abacus is presented as a guide to evaluate and optimize photovoltaic-thermal semiconductor thin films thermal and optical performance. The definition of this parameter ψ_AB takes into account the thermal diffusivity and the optical effective absorptivity of the material. As a first approximation, this parameter, could be considered as a 3D velocity of the transmitted heat inside the material.New functional semiconductor materials have been significantly classified using this abacus.