面部表情的深度卷积级联森林识别

面部表情识别广泛应用于各种研究领域,针对面部表情识别使用深度神经网络方法结构复杂、可解释性差和传统机器学习方法特征提取缺乏多样性、识别率低的问题.提出了一种新的深度卷积级联森林(Deep Convolution Cascade Forest,DCCF)方法用于人脸面部表情识别,该方法通过卷积神经网络深度学习人脸面部显著特征,并采用基于随机森林的级联结构森林逐层学习识别出不同的面部表情特征,提高了人脸表情的识别准确率.DCCF在JAFFE、CK+和Fer2013 3个公开面部表情数据集进行了实验,并对面部表情提取的5种特征和7种分类方法进行了比较分析,结果显示DCCF在对比的算法中人脸表情识别性能最好,3个数据集的准确率分别达到91.4%,98.7%,71.6%.     

Soft computing approach to fault diagnosis of centrifugal pump

Fault detection and isolation in rotating machinery is very important from an industrial viewpoint as it can help in maintenance activities and significantly reduce the down-time of the machine, resulting in major cost savings. Traditional methods have been found to be not very accurate. Soft computing based methods are now being increasingly employed for the purpose. The proposed method is based on a genetic programming technique which is known as gene expression programming (GEP). GEP is somewhat a new member of the genetic programming family. The main objective of this paper is to compare the classification accuracy of the proposed evolutionary computing based method with other pattern classification approaches such as support vector machine (SVM), Wavelet-GEP, and proximal support vector machine (PSVM). For this purpose, six states viz., normal, bearing fault, impeller fault, seal fault, impeller and bearing fault together, cavitation are simulated on centrifugal pump. Decision tree algorithm is used to select the features. The results obtained using GEP is compared with the performance of Wavelet-GEP, support vector machine (SVM) and proximal support vector machine (PSVM) based classifiers. It is observed that both GEP and SVM equally outperform the other two classifiers (PSVM and Wavelet-GEP) considered in the present study.     

一种高效挖掘冠心病医案关联规则的仿真研究

研究中医冠心病医案,高效挖掘有益信息规则问题,由于中医医案数据量大、关联性强,针对传统的关联规则挖掘算法处理中医医案数据时存在效率低、收敛速度慢及漏报规则等问题,提出一种小生境技术和基因表达式编程相结合的挖掘关联规则的方法。通过惩罚函数设置支持度阈值,利用小生境技术执行小生境演化、融合算法,结合基因表达式编程算法操作简单、鲁棒性强的优势搜索强关联规则,有效避免了算法早熟,解决了规则冗余。针对治疗冠心病的中医医案进行了验证性实验,实验结果表明,改进算法在提取有效信息的效率上有较大的提高,挖掘结果对冠心病中医临床诊治具有一定的参考价值。     

原发性高血压病靶器官损害的预测仿真研究

研究原发性高血压病靶器官损害准确预测问题,由于原发性高血压病靶器官损害的影响因子之间存在着高度冗余和非线性,导致传统方法预测准确率低,为提高预测精度,构建了一种改进的基因表达式编程算法的原发性高血压病靶器官损害预测模型,首先设计了适合靶器官损害预测的适应度函数,然后为克服早熟现象提出了以平均适应度形式自适应的改变重组算子概率和变异算子概率,通过数据训练得到靶器官损害的2年预测模型。采用对山东中医药大学第二附属医院原发性高血压病历史数据进行验证性仿真,结果表明,方法的预测精度更高,在原发性高血压病靶器官损害的预测中有着广泛的应用前景。     

Dysregulated CD38 Expression on Peripheral Blood Immune Cell Subsets in SLE

Given its uniformly high expression on plasma cells, CD38 has been considered as a therapeutic target in patients with systemic lupus erythematosus (SLE). Herein, we investigate the distribution of CD38 expression by peripheral blood leukocyte lineages to evaluate the potential therapeutic effect of CD38-targeting antibodies on these immune cell subsets and to delineate the use of CD38 as a biomarker in SLE. We analyzed the expression of CD38 on peripheral blood leukocyte subsets by flow and mass cytometry in two different cohorts, comprising a total of 56 SLE patients. The CD38 expression levels were subsequently correlated across immune cell lineages and subsets, and with clinical and serologic disease parameters of SLE. Compared to healthy controls (HC), CD38 expression levels in SLE were significantly increased on circulating plasmacytoid dendritic cells, CD14⁺⁺CD16⁺ monocytes, CD56⁺ CD16^dim natural killer cells, marginal zone-like IgD⁺CD27⁺ B cells, and on CD4⁺ and CD8⁺ memory T cells. Correlation analyses revealed coordinated CD38 expression between individual innate and memory T cell subsets in SLE but not HC. However, CD38 expression levels were heterogeneous across patients, and no correlation was found between CD38 expression on immune cell subsets and the disease activity index SLEDAI-2K or established serologic and immunological markers of disease activity. In conclusion, we identified widespread changes in CD38 expression on SLE immune cells that highly correlated over different leukocyte subsets within individual patients, but was heterogenous within the population of SLE patients, regardless of disease severity or clinical manifestations. As anti-CD38 treatment is being investigated in SLE, our results may have important implications for the personalized targeting of pathogenic leukocytes by anti-CD38 monoclonal antibodies.     

Functionalized Reduced Graphene Oxide as a Versatile Tool for Cancer Therapy

Cancer is one of the deadliest diseases in human history with extremely poor prognosis. Although many traditional therapeutic modalities—such as surgery, chemotherapy, and radiation therapy—have proved to be successful in inhibiting the growth of tumor cells, their side effects may vastly limited the actual benefits and patient acceptance. In this context, a nanomedicine approach for cancer therapy using functionalized nanomaterial has been gaining ground recently. Considering the ability to carry various anticancer drugs and to act as a photothermal agent, the use of carbon-based nanomaterials for cancer therapy has advanced rapidly. Within those nanomaterials, reduced graphene oxide (rGO), a graphene family 2D carbon nanomaterial, emerged as a good candidate for cancer photothermal therapy due to its excellent photothermal conversion in the near infrared range, large specific surface area for drug loading, as well as functional groups for functionalization with molecules such as photosensitizers, siRNA, ligands, etc. By unique design, multifunctional nanosystems could be designed based on rGO, which are endowed with promising temperature/pH-dependent drug/gene delivery abilities for multimodal cancer therapy. This could be further augmented by additional advantages offered by functionalized rGO, such as high biocompatibility, targeted delivery, and enhanced photothermal effects. Herewith, we first provide an overview of the most effective reducing agents for rGO synthesis via chemical reduction. This was followed by in-depth review of application of functionalized rGO in different cancer treatment modalities such as chemotherapy, photothermal therapy and/or photodynamic therapy, gene therapy, chemotherapy/phototherapy, and photothermal/immunotherapy.