Residual learning based CNN for breast cancer histopathological image classification

Biopsy is one of the most commonly used modality to identify breast cancer in women, where tissue is removed and studied by the pathologist under the microscope to look for abnormalities in tissue. This technique can be time-consuming, error-prone, and provides variable results depending on the expertise level of the pathologist. An automated and efficient approach not only aids in the diagnosis of breast cancer but also reduces human effort. In this paper, we develop an automated approach for the diagnosis of breast cancer tumors using histopathological images. In the proposed approach, we design a residual learning-based 152-layered convolutional neural network, named as ResHist for breast cancer histopathological image classification. ResHist model learns rich and discriminative features from the histopathological images and classifies histopathological images into benign and malignant classes. In addition, to enhance the performance of the developed model, we design a data augmentation technique, which is based on stain normalization, image patches generation, and affine transformation. The performance of the proposed approach is evaluated on publicly available BreaKHis dataset. The proposed ResHist model achieves an accuracy of 84.34% and an F1-score of 90.49% for the classification of histopathological images. Also, this approach achieves an accuracy of 92.52% and F1-score of 93.45% when data augmentation is employed. The proposed approach outperforms the existing methodologies in the classification of benign and malignant histopathological images. Furthermore, our experimental results demonstrate the superiority of our approach over the pre-trained networks, namely AlexNet, VGG16, VGG19, GoogleNet, Inception-v3, ResNet50, and ResNet152 for the classification of histopathological images.     

Document image mosaicing: A novel approach

There are situations where it is not possible to capture large documents with the given imaging media such as scanners or copying machines in a single stretch because of their inherent limitations. This results in capture of a large document in terms of split components of a document image. Hence, the need is to mosaic the split components into the original and put together the document image. In this paper, we present a novel and simple approach to mosaic two split images of a large document based on pixel value matching. The method compares the values of pixels in the columns of split images to identify the common or overlapping region (OR) in them, which helps in mosaicing of split images of a large document     

A novel scheme for image sharpness using inflection points

It is useful to increase the sharpness in medical images. This improvement can help medical diagnoses and treatment outcomes for patients. Noise overshoot and oversharpening effects are common artifacts of conventional sharpening algorithms. In this work, we propose an adaptive sharpness algorithm that achieves better sharpening effects than standard methods. As the pixel value changes abruptly at an edge, a method that adequately emphasized sudden variations in pixel values is effective for use in the sharpening process. Measurement of the gradient norm value of each pixel is calculated and compared to a threshold value to produce a curve. The curve quickly drops at the initial stage and then decreased more slowly for higher norm values. To distinguish the edge, an inflection point is determined by taking the second derivative of the curve and identifying the turnover point (where the curvature changes sign). Norm values higher than the inflection point are identified as belonging to the edge, and a simple sharpness filter is adaptively applied to these points. The proposed approach yielded better results than global filtering and a conventional unsharp masking approach when evaluated using Pratt's figure of merit.     

A novel level set method for image segmentation by combining local and global information

A novel level set method integrating local and global statistical information is proposed in this paper. In our method, a new signed pressure force (SPF) function is constructed by two parts. One is the global average intensity of the image, which can accelerate the evolution of the curve when the contour far away from the object boundaries. The other is the intensity average of difference image between the averaging convolution image and the original image, which can guide the evolving curve to catch the boundaries of the objects. In addition, an adaptive weighting function is utilized to adjust the ratio between the global and local terms, which can eliminate the inconvenient selection of weighting parameter. By substituting the new SPF function for the edge stopping function of the geodesic active contour model, we obtain a novel adaptive hybrid segmentation model, which is capable of segmenting the images with intensity inhomogeneity. What is more, in our method, the level set function is initialized with a binary function, which reduces the computational cost for the re-initialization step. The experimental results and comparisons with several popular models on synthetic and real images indicate that our method achieves superior performance in segmenting images with noise, low contrast and intensity inhomogeneity.     

基于支持向量机的彩色图像恢复新方法

引起图像退化的因素众多，因而难以用一个统一的方法来进行恢复处理。鉴于图像的像素和各颜色分量通道间本质上存在某种相关性，以及以支持向量机为核心的统计学习理论具有较好地解决小样本、非线性、高维数问题的能力，提出了一种新的空域图像恢复方法，并通过对来自于待处理图像本身的训练样本的学习，构造自适应的回归插值函数；然后基于该函数对图像作有选择的修改，从而达到图像恢复的目的。实验表明，该方法是有效的，并且具有较好的泛化性能。     

基于非线性最小二乘的电容层析成像图像重建算法

针对电容层析成像技术中的软场效应和病态问题,提出了基于非线性最小二乘算法的新电容层析成像(ECT)算法。在分析非线性最小二乘问题残量原理的基础上,给出了目标函数中二阶信息项的割线近似的校正公式,并利用Lipschitz空间连续的性质对非线性最小二乘算法的收敛性进行了证明,在此基础上探讨了ECT应用该算法的可行性。该算法满足收敛条件且重建图像误差小。仿真和实验结果表明,与LBP、Landweber和共轭梯度算法相比,对于简单流型该算法兼备成像质量高、边界均匀稳定等优点,该算法的提出为ECT图像重建算法的研究提供了一个新的思路。     

A novel framework for spatiotemporal monitoring and post-signal diagnosis of processes with image data

Advances in digital equipment and organizations' interest in having comprehensive and real-time information about products have increased the use of machine vision systems in organizations. In this paper, to monitor a sensory quality characteristic of a product based on images, the residual matrix of the intensity values of the nominal and captured images is divided into specific regions; then, the equality of the means of the regions is tested based on one-way ANOVA. To do so, a P-value–based control chart is applied to detect the out-of-control state as soon as possible. If an out-of-control alarm is received, Dunnett's test is used to identify region(s) with significant differences in the means of residuals (defective location[s] in the image) compared with other regions. After the locations of defective regions are identified, the change point of the process is estimated by using the maximum likelihood estimation approach. The performance of the proposed procedure is compared with some of the previous approaches in the literature. Then, the proposed procedure is implemented in a real-world case. The simulation study demonstrates the merits of the proposed procedure: It is not limited to specific geometric types of faults; it has the ability to detect multiple faults in the image; it takes less computational time; and it performs well in estimating the real time of change, as well as the location(s) and dimension(s) of the fault(s).     

Distance-Oriented Cuckoo Search enabled optimal histogram for underwater image enhancement: a novel quality metric analysis

The major problem in underwater image processing involves light attenuation. In order to overcome such light incorporation and dispersion issues, this paper proposes a new underwater image enhancement model that includes certain phases like contrast correction and colour correction. Here, two processes are worked out in contrast correction namely (i) global contrast correction and (ii) local contrast correction. Furthermore, as the histogram evaluation is the major part of any image enhancement, it is very important to choose the optimal limits of the histogram. To solve this optimization problem, this paper introduces a new optimization model namely Distance Oriented Cuckoo Search (DOCS) algorithm that also satisfies the objective function defined in this paper. The major process of this model focusses on Integrated Global and Local Contrast Correction (IGLCC). Finally, the proposed model compares its performance over other methods.     

A novel automated magnetic resonance image segmentation approach based on elliptical gamma mixture model for breast lumps detection

This article introduces a novel semisupervised automated segmentation approach for breast magnetic resonance (MR) image on multicore CPU-GPU systems. The basic idea of the proposed method is clustering-based semisupervised classifier devised by elliptical gamma mixture model (EGMM). Parameters of EGMM are identified by the iterative log-expectation maximization (EM) algorithm. The suggested classifier labels the groups of voxels in an input image first and then classifies the image slices using the EGMM. Two different implementations of the proposed algorithm have been developed based on two different types of high-performance computing architectures such as graphics processing units (GPUs) and multicore processors. To realize the real-time segmentation performance of our algorithm with two distinctive architecture, we have tested a set of breast MR images collected from MedPix. Comparison between two architectures in terms of segmentation performance and computational cost is assessed by the analysis of simulation and experimental results.     

壳聚糖涂层聚乳酸细胞微载体的制备和性能

采用氨解技术在聚乳酸微球表面引入自由氨基,再利用戊二醛将氨基转化为醛基,最后采用接枝涂层技术将壳聚糖固定到聚乳酸微球表面,制备了壳聚糖表面改性的聚乳酸细胞微载体.分别采用茚三酮法和乙酰丙酮-对二甲氨基苯甲醛法测定了聚乳酸微球表面的氨基和壳聚糖含量.发现氨基的量初始随氨解时间的延长而增大,达到最大(2.94×10-7mol/mg)后保持不变.与空白聚乳酸微球相比,软骨细胞在壳聚糖改性聚乳酸微球表面能够更有效地粘附和生长,分布更为均匀.     

A novel collagen gel-based measurement technique for quantitation of cell contraction force

Cell contraction force plays an important role in wound healing, inflammation, angiogenesis and metastasis. This study describes a novel method to quantify single cell contraction force in vitro using human aortic adventitial fibroblasts embedded in a collagen gel. The technique is based on a depth sensing nano-indentation tester to measure the thickness and elasticity of collagen gels containing stimulated fibroblasts and a microscopy imaging system to estimate the gel area. In parallel, a simple theoretical model has been developed to calculate cell contraction force based on the measured parameters. Histamine (100 µM) was used to stimulate fibroblast contraction while the myosin light chain kinase inhibitor ML-7 (25 µM) was used to inhibit cell contraction. The collagen matrix used in the model provides a physiological environment for fibroblast contraction studies. Measurement of changes in collagen gel elasticity and thickness arising from histamine treatments provides a novel convenient technique to measure cell contraction force within a collagen matrix. This study demonstrates that histamine can elicit a significant increase in contraction force of fibroblasts embedded in collagen, while the Young''s modulus of the gel decreases due to the gel degradation.     

Fabrication of a thermoresponsive cell culture dish: a key technology for cell sheet tissue engineering

Abstract

This article reviews the properties and characterization of an intelligent thermoresponsive surface, which is a key technology for cell sheet-based tissue engineering. Intelligent thermoresponsive surfaces grafted with poly(N-isopropylacrylamide) exhibit hydrophilic/hydrophobic alteration in response to temperature change. Cultured cells are harvested on thermoresponsive cell culture dishes by decreasing the temperature without the use of digestive enzymes or chelating agents. Our group has developed cell sheet-based tissue engineering for therapeutic uses with single layer or multilayered cell sheets, which were recovered from the thermoresponsive cell culture dish. Using surface derivation techniques, we developed a new generation of thermoresponsive cell culture dishes to improve culture conditions. We also designed a new methodology for constructing well-defined organs using microfabrication techniques.     

A novel multilayer aqueous tape casting method for anode-supported planar solid oxide fuel cell

Anode supported planar solid oxide fuel cell (SOFC) with large dimension (100 × 100 × 0.5 mm³) has been successfully prepared by using multilayer aqueous tape casting technique and co-firing process which allow for the creation of a thin layer of dense YSZ electrolyte on NiO/YSZ porous anode support. Then the performance of a single cell with YSZ electrolyte film and La_0.8Sr_0.2MnO₃ (LSM) + SSZ cathode is evaluated between 700 °C and 800 °C with hydrogen as the fuel and air as the oxidant. The open circuit voltage (OCV) of 1.08 V is close to the theoretical value at 800 °C. The area specific resistance value is 0.79 Ω cm² and the maximum power density is over 0.8 W cm^− 2 at 800 °C. In terms of the electrochemical performance, single cells fabricated from the large dimension half cell show excellent uniformity.     

Fabrication of a thermoresponsive cell culture dish: a key technology for cell sheet tissue engineering

This article reviews the properties and characterization of an intelligent thermoresponsive surface, which is a key technology for cell sheet-based tissue engineering. Intelligent thermoresponsive surfaces grafted with poly(N-isopropylacrylamide) exhibit hydrophilic/hydrophobic alteration in response to temperature change. Cultured cells are harvested on thermoresponsive cell culture dishes by decreasing the temperature without the use of digestive enzymes or chelating agents. Our group has developed cell sheet-based tissue engineering for therapeutic uses with single layer or multilayered cell sheets, which were recovered from the thermoresponsive cell culture dish. Using surface derivation techniques, we developed a new generation of thermoresponsive cell culture dishes to improve culture conditions. We also designed a new methodology for constructing well-defined organs using microfabrication techniques.     

A fast BNM (Best Neighborhood Matching): Algorithm and parallel processing for image restoration

Best Neighborhood Matching (BNM) algorithm is a good approach of error concealment in terms of restored image quality. However, this kind of error concealment algorithm is commonly computation‐intensive, which restricts their real applications on large‐scale image or video sequence restoration. In this article, we propose a fast method, named Jump and look around Best Neighborhood Matching (JBNM), which reduces computing time to one sixth of that by BNM, while the quality of the restored images remains almost the same. To further reduce processing time and meet large‐scale image restorations, a parallel JBNM working on a cluster of workstations is proposed. Several critical techniques, including reading policy, overlap stripe data distribution, and communication strategies, have been developed to obtain high performance. Both theoretical analysis and experiment results indicate that our parallel JBNM provides an efficient technique for image restoration applications. © 2003 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 13, 189–200, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ima.10057     

Gold nanoparticles loaded into niosomes: A novel approach for enhanced antitumor activity against human ovarian cancer

In the current study, gold nanoparticles (AuNPs) were prepared using the green synthesis method using Artemisia annua extract, loaded into niosomes, and investigated their cytotoxicity and apoptotic effects. To optimize the niosomal formulation containing AuNPs, the effects of surfactants: cholesterol molar ratio, Span 60: Tween 60 M ratio, and AuNP concentration (µg/mL) were investigated. After examining the drug release profile, mathematical models were assessed to predict release kinetic. The cytotoxicity of noisome encapsulated AuNPs and free AuNPs was evaluated against human ovarian cancer cell line (A2780) by MTT assay. The apoptotic/necrosis ratio was studied using flow cytometry as well as Real-Time PCR was used for the Bax and Bcl-2 apoptosis gene expression. The results showed that the entrapment efficiency and particle size of optimized niosomal formulation encapsulated AuNPs were 34.49%±0.84 and 153.6 ± 4.62 nm with a regular spherical shape, respectively. The release profile of AuNPs from niosomal formulation was 59%±1.0 after 8 h suggesting the controlled release profile. This formulation exerted dose-dependent cytotoxicity against the A2780 cells via induction of apoptosis and significant upregulation of mRNA expression of Bax gene; while expression of anti-apoptotic gene Bcl-2 was down-regulated. Thus, the findings suggest that AuNP-loaded niosomal formulation is considered a promising and suitable targeted system for improving anti-tumor activity against A2780 cells.     

A domain coupling method for finite element digital image correlation with mechanical regularization: Application to multiscale measurements and parallel computing

A general method is proposed to couple two subregions analyzed with finite element digital image correlation even when using a mechanical regularization (regularized digital image correlation). A Lagrange multiplier is introduced to stitch both displacements fields in order to recover continuity over the full region of interest. Another interface unknown is introduced to ensure, additionally, the equilibrium of the mechanical models used for regularization. As a first application, the method is used to perform a single measurement from images at two different resolutions. Secondly, the method is also extended to parallel computing in regularized digital image correlation. The problem is formulated at the interface and solved with a Krylov‐type algorithm. A dedicated preconditioner is proposed to significantly accelerate convergence. The resulting method is a good candidate for the analysis of large data sets. Copyright © 2016 John Wiley & Sons, Ltd.