Search

Characterizing the subtle changes of functional brain networks associated with the pathological cascade of Alzheimer’s disease (AD) is important for early diagnosis and prediction of disease progression prior to clinical symptoms. We developed a new deep learning method, termed multiple graph Gaussian embedding model (MG2G), which can learn highly informative network features by mapping high-dimensional resting-state brain networks into a low-dimensional latent space. These latent distribution-based embeddings enable a quantitative characterization of subtle and heterogeneous brain connectivity patterns at different regions, and can be used as input to traditional classifiers for various downstream graph analytic tasks, such as AD early stage prediction, and statistical evaluation of between-group significant alterations across brain regions. We used MG2G to detect the intrinsic latent dimensionality of MEG brain networks, predict the progression of patients with mild cognitive impairment (MCI) to AD, and identify brain regions with network alterations related to MCI.

Red blood cell (RBC) segmentation and classification from microscopic images is a crucial step for the diagnosis of sickle cell disease (SCD). In this work, we adopt a deep learning based semantic segmentation framework to solve the RBC classification task. A major challenge for robust segmentation and classification is the large variations on the size, shape and viewpoint of the cells, combining with the low image quality caused by noise and artifacts. To address these challenges, we apply deformable convolution layers to the classic U-Net structure and implement the deformable U-Net (dU-Net). U-Net architecture has been shown to offer accurate localization for image semantic segmentation. Moreover, deformable convolution enables free-form deformation of the feature learning process, thus making the network more robust to various cell morphologies and image settings. dU-Net is tested on microscopic red blood cell images from patients with sickle cell disease. Results show that dU-Net can achieve highest accuracy for both binary segmentation and multi-class semantic segmentation tasks, comparing with both unsupervised and state-of-the-art deep learning based supervised segmentation methods. Through detailed investigation of the segmentation results, we further conclude that the performance improvement is mainly caused by the deformable convolution layer, which has better ability to separate the touching cells, discriminate the background noise and predict correct cell shapes without any shape priors.

Primary malignant cardiac tumors (PMCTs) are extremely rare. The apparent immunity of the heart to invasive cancer has attracted considerable interest given the continuously rising incidence of cancer in other organs. This study aims to determine the conditions that could result in cardiac carcinoma and expand our understanding of cardiac tumor occurrence. We report two cases: a male (Patient-1) with primary cardiac malignant fibrous histiocytoma (MFH) and a female (Patient-2) with primary cardiac angiosarcoma. Merged genome-wide analyses of aCGH, Exome sequencing, and RNA-sequencing were performed on Patient-1 using peripheral blood, carcinoma tissue, and samples of adjacent normal tissue. Only whole-transcriptome analysis was carried out on Patient-2, due to insufficient quantities of sample from Patient-2. We identified a novel inherited loss of functional mutation of FH (Glu404Asp), a recurrent somatic hotspot mutation of PIK3CA (His1047Arg) and a somatic duplication in copy number of HIF1A. FH (E404D) severely compromised FH enzyme activity and lead to decreased protein expression in cardiac tumor tissues. We previously reported a functional mutation ACOX2 (R409H), which is potentially associated with decreased β-oxidation of fatty acids in the cardiac tumor tissue. Results of transcriptome analyses on two patients further revealed that the RNA expression of genes in the TCA cycle and beta-oxidation were uniformly downregulated. In this study, combined effects of FH (E404D) and ACOX2 (R409H) on metabolic switch from fatty acids to glucose were remarkably distinct, which might be an essential precondition to trigger the occurrence of PMCTs and mimic the Warburg effect, a hallmark of cancer metabolism.

The expansion of artificial constructs with the rapid economic development in China has led to ecological and environmental emergencies. The extent of the decline in natural resources and environmental conditions has recently been recognized. Identifying “ecological protection redlines”, i.e. ecological limits, to guarantee ecological baselines for natural resources and ecosystem service functions would therefore help to coordinate economic development and to protect ecological resources in the coming years. We used remotely sensed and climatic data to delimit the ecological protection redlines for Zibo, a typical and important city in Shandong province, as an example to illustrate the principles and methodology of ecological protection redlines. The area of the ecological protection redlines for Zibo encompassed 1132.26 km 2 , accounting for 18.98% of the total area of Zibo, were mainly distributed in the southern regions of the municipality, and consisted of extremely important areas of ecosystem service functions, including water conservation, both soil and water conservation, windbreaks and sand fixation, and the conservation of biodiversity. This area is extremely sensitive, and development is forbidden. Strict measures of management and control should be implemented to protect the long-term effectiveness of ecological protection redlines.

In this paper, we propose a new method of multiscale recurrence quantification analysis (MSRQA) to analyze the structure of order recurrence plots. The MSRQA is based on order patterns over a range of time scales. Compared with conventional recurrence quantification analysis (RQA), the MSRQA can show richer and more recognizable information on the local characteristics of diverse systems which successfully describes their recurrence properties. Both synthetic series and stock market indexes exhibit their properties of recurrence at large time scales that quite differ from those at a single time scale. Some systems present more accurate recurrence patterns under large time scales. It demonstrates that the new approach is effective for distinguishing three similar stock market systems and showing some inherent differences.

In order to improve the utilization of the residential electricity consumption data which contains the information on the user’s electricity consumption habits, a residential electricity consumption behaviors mining algorithm model is constructed. Firstly, according to the attribute, the collected data can be divided into the global data and the phase data, then the appropriate global variables are selected to mine the user’s electricity consumption patterns in the near future on the system clustering algorithm. Based on the theory of grey relational analysis, combing phase data with the power modes to analyze the potential characteristics of residential electricity consumption behaviors deeply that verify the ability of latest power mode to predict household electricity consumption situation in the coming few days and the effect of dominant phase variables on the peak load shifting. Finally, from the actual data of a certain family, the proposed data mining algorithm is testified that it can effectively explore the electricity consumption behavior habits and characteristics of the family.

Cell detection and cell type classification from biomedical images play an important role for high-throughput imaging and various clinical application. While classification of single cell sample can be performed with standard computer vision and machine learning methods, analysis of multi-label samples (region containing congregating cells) is more challenging, as separation of individual cells can be difficult (e.g. touching cells) or even impossible (e.g. overlapping cells). As multi-instance images are common in analyzing Red Blood Cell (RBC) for Sickle Cell Disease (SCD) diagnosis, we develop and implement a multi-instance cell detection and classification framework to address this challenge. The framework firstly trains a region proposal model based on Region-based Convolutional Network (RCNN) to obtain bounding-boxes of regions potentially containing single or multiple cells from input microscopic images, which are extracted as image patches. High-level image features are then calculated from image patches through a pre-trained Convolutional Neural Network (CNN) with ResNet-50 structure. Using these image features inputs, six networks are then trained to make multi-label prediction of whether a given patch contains cells belonging to a specific cell type. As the six networks are trained with image patches consisting of both individual cells and touching/overlapping cells, they can effectively recognize cell types that are presented in multi-instance image samples. Finally, for the purpose of SCD testing, we train another machine learning classifier to predict whether the given image patch contains abnormal cell type based on outputs from the six networks. Testing result of the proposed framework shows that it can achieve good performance in automatic cell detection and classification., Wei Qiu, Jiaming Guo and Xiang Li contributed equally

Empirical mode decomposition (EMD) is a data-driven signal analysis method for nonlinear and nonstationary data. Since it is intuitive, direct, posterior and adaptive, EMD is widely applied to various fields of study. In this paper, EMD and ensemble empirical mode decomposition (EEMD), a modified method of EMD, are applied to financial time series. Through analyzing the intrinsic mode functions (IMFs) of EMD and EEMD, we find EEMD method performs better on the orthogonality of IMFs than EMD. With clustering the ordered frequencies of IMFs, the IMFs obtained from EEMD method are grouped into high-, medium-, and low-frequency components, representing the short-, medium-, and long-term volatilities of the index sequences, respectively. With the cross-correlation analysis of DCCA cross-correlation coefficient, our findings allow us to gain further and detailed insight into the cross-correlations of stock markets.

A prefabricated strain-slice-style speckle pattern applied to high-temperature digital image correlation (DIC) method is proposed and studied in this paper. By virtue of spot welding process, this proposed speckle pattern can be prepared directly and easily on specimen surface. A heating experiment was performed to investigate the high-temperature performance of this proposed speckle pattern under air-oxidation condition. Besides, the ability of this proposed speckle for high-temperature DIC method was verified in a thermal expansion experiment, during which thermal deformation and coefficient of thermal expansion (CTE) values were measured and subsequently compared with the reference data. Results show the effectiveness of this prefabricated strain-slice-style speckle pattern for full-field in-plane thermal deformation measurement with high accuracy using DIC method under air-oxidation condition at high temperature up to 1000 °C.

Reliable cell segmentation and classification from biomedical images is a crucial step for both scientific research and clinical practice. A major challenge for more robust segmentation and classification methods is the large variations in the size, shape and viewpoint of the cells, combining with the low image quality caused by noise and artifacts. To address this issue, in this work we propose a learning-based, simultaneous cell segmentation and classification method based on the U-Net structure with deformable convolution layers. The U-Net architecture has been shown to offer a precise localization for image semantic segmentation. Moreover, deformable convolution enables the free form deformation of the feature learning process, thus making the whole network more robust to various cell morphologies and image settings. The proposed method is tested on microscopic red blood cell images from patients with sickle cell disease. The results show that U-Net with deformable convolution achieves the highest accuracy for both segmentation and classification tasks, compared with the original U-Net structure and unsupervised methods.

Complexity in time series is an intriguing feature of living dynamical systems such as traffic systems, with potential use for identification of system state. Here, we introduce a method for complex system analysis, called generalized sample entropy and surrogate data analysis, to measure the complexity in traffic signals. The behavior of the weekday series is quite different from that of the weekend series. In addition, we propose and discuss a new method of multifractality based on multifractal detrended fluctuation analysis (MF-DFA) and surrogate data. MF-DFA is the most popular method to detect multifractal characteristics of nonstationary time series. This new method is further applied to binomial multifractal series and traffic signals of Beijing, China. Results show that multifractal characteristics of the artificial data are regular and distinguishing the traffic signals of different patterns is significant. Besides, we find that there is a connection of multifractal characteristics between binomial multifractal series and traffic signals.

In this study, we develop a new algorithm based on fractional operators of variable-order in order to enhance image quality. First, three kinds of popular high-order discrete formulas are adopted to obtain the coefficients, and subsequently, a mask optimization method for selecting the fractional order adaptively is applied to construct a variable-order fractional differential mask along with the coefficients generated from the first step. We carry out experiments on OCT thoracic aorta images and some nature images with low contrast and noise, demonstrating that the high-order discrete method leads to significantly better performance in enhancing the edge information nonlinearly compared to the standard first-order discrete method. Moreover, the optimized mask with variable-order of the fractional derivative not only can preserve the edge information of the processed images adequately, but it also effectively suppresses the noise in the smooth area.

A thermo-mechanical interface element was developed to describe the effective penetration, the fillet weld size and the contact relationship between the skin plate and the stiffener during welding distortion simulation. Based on the interface element formulation, welding distortion in stiffened structures was predicted by means of thermal elastic–plastic finite element analysis. Comparisons between the traditional model and the interface element model were conducted. The results indicate that the calculated data of the interface element model are in better agreement with the experimental measurements. The influences of the welding sequence were discussed numerically based on these two models. The interface element model is more suitable for describing the effect of the welding sequence, which are of great importance on accurately predicting distortion in large scale structure welding. Simulations and experiments upon large scale structures were also carried out to verify the effectiveness of the developed numerical analysis procedure.

The morphological change of red blood cells(RBCs) plays an important role in revealing the biomechanical and biorheological characteristics of RBCs. Aiming to extract the shape indices for the sickle RBCs, an automated ex-vivo RBC shape quantification method is proposed. First, single RBC regions (ROIs) are extracted from raw microscopy image via an automatic hierarchical ROI extraction method. Second, an improved random walk method is used to detect the RBC outline. Finally, three types of RBC shape factors are calculated based on the elliptical fitting RBC contour. Experiments indicate that the proposed method can accurately segment the RBCs from the microscopy images with low contrast and prevent the disturbance of artifacts. Moreover, it can provide an efficient shape quantification means for diverse RBC shapes in a batch manner.

Urban expansion poses severe threats to ecosystems. It is therefore important we better understand the impact of different urban expansion level on ecosystems for developing regionally differentiated ecological protection policies. Here, we proposed a conceptual framework to describe the impacts of urbanization on bundles of ecosystem services. Referred to as the concept of land use degree by nighttime light data, we put forward and verified an urban expansion level model. According to this model, study area was divided into a slow increase zone, increase zone, and rapid increase zone. Then, taking Taihu Lake Basin in China as a case, we used Zonal-statistics and Pearson correlation coefficients to reveal the impact in different zones of urban expansion level on multiple ecosystem services: crop production, freshwater supply, aquatic production, net primary productivity, soil conservation, water retention, flood regulation, and forest recreation index. Our results revealed that urban expansion levels significantly impacted all ecosystem services. In either increase zone or rapid increase zone, we found lowered values of crop production, net primary productivity, soil conservation, water retention, and flood regulation, while both aquatic production and forest recreation index increased in all zones from 1990 to 2010. Across the levels of urbanization, urban expansion level was always negatively correlated with provisioning services. This result suggests local governors should improve crop production per unit area and increase the cultivated land area to guarantee food security. In addition, urban expansion level had positive correlations with the trade-offs between flood regulation and forest recreation index, and those among crop production, freshwater supply, and net primary productivity. Therefore, policy-makers should effectively maintain the land use balance among ecological protection, agriculture development, and urban expansion to better coordinate relationships between development and protection. In acquiring quantitative knowledge of how urban expansion level drives ecosystem changes, our findings may help guide future sustainable urban planning with respect to ecosystem services, urban development, and human welfare benefits.

Social work professional training in China is newly developed in the past decade. How to design and deliver specific training to meet the specific needs of social service professionals has been a crucial issue to social work educators in China. In 2004, China Women's University, in collaboration with the Faculty of Social Work, University of Manitoba, started the project ‘Building human capacity: Social work with rural women in China’ funded by the Canadian International Development Agency (CIDA). One of the main goals of project is to build social work professional capacity of the staff in the All-China Women's Federation (Women's Federation), so that they can provide service of better quality to women. In this paper, based on the data collected in the project sites, the authors will concentrate on what kind of training the Women's Federation staff received, and how they integrated that knowledge into their daily work. The authors will also suggest some implications for social work professional training ...

Region duplication forgery is one of the tampering techniques that are frequently used, where a part of an image is copied and pasted into another part of the same image. In this paper, a phase correlation method based on polar expansion and adaptive band limitation is proposed for region duplication forgery detection. Our method starts by calculating the Fourier transform of the polar expansion on overlapping windows pair, and then an adaptive band limitation procedure is implemented to obtain a correlation matrix in which the peak is effectively enhanced. After estimating the rotation angle of the forgery region, a searching algorithm in the sense of seed filling is executed to display the whole duplicated region. Experimental results show that the proposed approach can detect duplicated region with high accuracy and robustness to rotation, illumination adjustment, blur and JPEG compression while rotation angle is estimated precisely for further calculation.

Traditional Hessian multiscale filter consider only the local geometric feature but not the global grayscale infor- mation. In medical image analysis, Hessian filter is usually used to enhance the blood vessels. However, it also produces some pseudo vascular structures or some isolate noise points, such as the nasal soft tissues that have the similar shape with the vessels in MRA data, which will increase the difficulty of cerebrovascular segmentation. To resolve this issue, an im- proved Hessian multiscale filter is proposed in this paper. An image grayscale factor is added to the vascular similarity func- tion computed by Hessian matrix eigenvalue. This method is experimented on brain MRA data and lung CTA data. Experi- mental results show that this method can enhance vascular structures, and simultaneously reduce the appearance of the pseu- do vascular structures and the isolated noise points.

For the purpose of taking good use of the diagnosis obtained from medical experts and improving the accuracy of chest X-ray images retrieval, the lung fields are segmented and interested regions are marked off on the basis of chest X-ray images having been processed previously; the Gray Difference Statistics is used to indicate the texture feature of each region. Using the K-nearest neighbor classifier, the texture features are mapped respectively to the standard image classes pre-described by the experts, thus it realizes the semantic annotation of regions in the whole image. This method can not only narrow the semantic gap between the low-level features and the high-level semantics of images effectively, but also has an active effect on improving the efficiency of medical diagnosis.