Deep learning has become an important tool for image classification and natural language processing. However, the effectiveness of deep is highly dependent on quality training data as well net model learning. The set normally fairly large, pretty complex. It necessary to validate framework including model, executing environment, before it used any applications. In this paper, we propose approach validating accuracy a that includes convolutional neural network, massive set. first validated...

Diffraction imaging of cells allows rapid phenotyping by the response intracellular molecules to coherent illumination. However, its ability distinguish numerous types human leukocytes remains be investigated. Here, we show that accurate classification three lymphocyte subtypes can achieved with features extracted from cross-polarized diffraction image (p-DI) pairs. A deep neural network (DNN) DINet-PS has been developed for feature extraction and filtering of, in angular frequency domain,...

Big data validation and system verification are crucial for ensuring the quality of big applications. However, a rigorous technique such tasks is yet to emerge. During past decade, we have developed called CMA investigating classification biological cells based on cell morphology which captured in diffraction images. includes collection scientific software tools, machine learning algorithms, large-scale image repository. In order ensure CMA, framework rigorously validating massive scale as...

Deep learning has become an important tool for image classification and natural language processing. However, the effectiveness of deep is highly dependent on quality training data as well net model learning. The set normally fairly large, pretty complex. It necessary to validate framework including model, executing environment, before it used any applications. In this paper, we propose approach validating accuracy a that includes convolutional neural network, massive set. first validated...

This study aimed to explore the contralesional homotopic functional plasticity in brain of patients with unilateral temporal glioma.

Abstract Glioma can cause variable alterations to the structure and function of brain. However, there is a paucity studies on gray matter (GM) volume in brain region opposite temporal glioma before after surgery. Therefore, present study was initiated investigate alternation contralateral homotopic GM patients with unilateral lobe further, assess relationship between alternations cognition. Eight left (LTPs), nine right (RTPs), 28 demographically matched healthy controls (HCs) were included....

With the rapid development of Internet technology, number global users is rapidly increasing, and scale also expanding. The huge system has accelerated spread bad information, including images. Bad images reflect vulgar culture Internet. They will not only pollute environment impact core society but endanger physical mental health young people. In addition, some criminals use to induce download software containing computer viruses, which greatly security cyberspace. Cyberspace governance...

Morphological changes in apoptotic cells provide essential markers for defining and detection of apoptosis as a fundamental mechanism cell death. Among these changes, the nuclear fragmentation condensation have been regarded important but quantitative characterization is yet to be achieved. We acquired confocal image stacks 206 viable MCF-7 stained by three fluorescent dyes. Three-dimensional (3D) parameters were extracted quantify compare their differences morphology. To analyze...

Abstract Neurotrophin 3 (NT3) induces mouse cortical stem cells to an asymmetric division from a symmetric division, suggesting that NT3 may work as early differentiative signal for neural (NSCs). Here, using cultured post‐natal hippocampal model, we demonstrated NT3‐stimulation causes NSCs differentiate into oligodendrocyte precursors (OLPs) through extracellular signal‐related kinase1/2 (Erk1/2)‐dependent pathway. Following the treatment of 24 h, differentiated more OLPs and fewer neurons,...

During past several years, we have built an online big data service called CMA that includes a group of scientific modeling and analysis tools, machine learning algorithms large scale image database for biological cell classification phenotyping study. Due to the complexity "nontestable" software algorithms, adequately verifying validating services is grand challenge. In this paper, introduce framework ensuring quality services. The iterative metamorphic testing technique "non-testable"...

Tumor infiltration of central nervous system (CNS) malignant tumors may extend beyond visible contrast enhancement. This study explored tumor habitat characteristics in the intratumoral and peritumoral regions to distinguish common brain such as glioblastoma, primary lymphoma, metastases. The preoperative MRI data 200 patients with solitary were included from two datasets for training. Quantitative radiomic features extracted model performance was evaluated using (n = 50) third clinical...

The two major subtypes of human T cells, CD4+ and CD8+, play important roles in adaptive immune response by their diverse functions. To understand the structure–function relation at single cell level, we isolated 2483 2450 CD8+ cells from fresh splenocytes immunofluorescent sorting investigated morphologic relations to surface CD markers acquisition analysis cross-polarized diffraction image (p-DI) pairs. A deep neural network DINet-R has been built extract 2560 features across multiple...

Multiparameter spectrophotometry (MPS) provides a powerful and environmental-friendly tool for characterization of turbid materials including foods without additives. Existing methods MPS, however, lack either modeling accuracy or capacity translation into ease-of-use instrument. We have developed novel method MPS to rapidly retrieve the spectral data absorption coefficient, scattering coefficient anisotropy factor within framework radiative transfer (RT) theory using integrating spheres. To...

Coherent light scattering presents complex spatial patterns that depend on morphological and molecular features of biological cells. We present a numerical approach to establish realistic optical cell models for generating virtual cells accurate simulation diffraction images are comparable measured data prostate With contourlet transform algorithm, it has been shown the simulated extracted parameters can be used distinguish different nuclear volumes refractive indices against orientation...

SignificanceMultiparameter spectrophotometry (MPS) provides a powerful tool for accurate characterization of turbid materials in applications such as analysis material compositions, assay biological tissues clinical diagnosis and food safety monitoring.AimThis work is aimed at development validation rapid inverse solver based on particle swarm optimization (PSO) algorithm to retrieve the radiative transfer (RT) parameters absorption coefficient, scattering coefficient anisotropy factor...

Measurement of nuclear-to-cytoplasm (N:C) ratios plays an important role in detection atypical and tumor cells. Yet, current clinical methods rely heavily on immunofluroescent staining manual reading. To achieve the goal rapid label-free cell classification, realistic optical models (OCMs) have been developed for simulation diffraction imaging by single A total 1892 OCMs were obtained with varied nuclear volumes orientations to calculate cross-polarized image (p-DI) pairs divided into three...

A new and noncontact approach of multispectral reflectance imaging has been developed to inversely determine the absorption coefficient μ a , scattering s anisotropy factor g turbid target from one measured image.The incident beam was profiled with diffuse standard for deriving both calculated images.A GPU implemented Monte Carlo code parameters conjugate gradient descent algorithm existence unique solutions shown.We noninvasively determined embedded region thickness in heterogeneous targets...

Methods for rapid and label‐free cell assay are highly desired in life science. Single‐shot diffraction imaging presents strong potentials to achieve this goal as evidenced by past experimental results using methods such polarization flow cytometry. We present here a platform of toward solving these problems optical model (OCM) evaluations calculations analysis cross‐polarized image (p‐DI) pairs. Four types realistic OCMs have been developed with two prostate structures adjustable refractive...

Development of label-free methods for accurate classification cells with high throughput can yield powerful tools biological research and clinical applications. We have developed a deep neural network DINet extracting features from cross-polarized diffraction image (p-DI) pairs on multiple pixel scales to accurately classify in five types. A total 6185 were measured by polarization imaging flow cytometry (p-DIFC) method followed cell p-DI data. The averaged value SD accuracy found be 98.9% ±...