Detecting Lumbar Implant and Diagnosing Scoliosis from Vietnamese X-Ray Imaging Using the Pre-Trained API Models and Transfer Learning

With the rapid growth of the autonomous system, deep learning has become integral parts to enumerate applications especially in the case of healthcare systems. Human body vertebrae are the longest and complex parts of the human body. There are numerous kinds of conditions such as scoliosis, vertebra degeneration, and vertebrate disc spacing that are related to the human body vertebrae or spine or backbone. Early detection of these problems is very important otherwise patients will suffer from a disease for a lifetime. In this proposed system, we developed an autonomous system that detects lumbar implants and diagnoses scoliosis from the modified Vietnamese x-ray imaging. We applied two different approaches including pre-trained APIs and transfer learning with their pre-trained models due to the unavailability of sufficient x-ray medical imaging. The results show that transfer learning is suitable for the modified Vietnamese x-ray imaging data as compared to the pre-trained API models. Moreover, we also explored and analyzed four transfer learning models and two pre-trained API models with our datasets in terms of accuracy, sensitivity, and specificity.     

Multi-Slice Modeling in Circuit-Field Coupled Systems Using Finite-Element and Modified Nodal Analyses

We propose a systematic method for interfacing different 2-D low-frequency electromagnetic field slices and nonidentical multi-port circuit networks using finite-element (FE) and modified nodal analyses. Each independent field slice has its own length, geometry, materials, boundary conditions, and number of filamentary and massive conductors. Similarly, each multi-port circuit network can have different circuit elements, topology, and number of ports. As a result, field slices are not required to be successively connected, since multi-port circuit networks can exist between them. Thus, solutions for virtually any possible configuration of 2-D circuit-field coupled systems are possible. The number of slices and multi-port networks does not affect the general form of the final system of equations, which is symmetric for most practical cases. Although we developed the methodology for 2-D time-harmonic magnetic problems, it can be readily extended to the transient case. We analyzed an induction motor to test the proposed multi-slice FE approach and validated the results using the total short-circuit leakage reactance and commercial software (Flux2D).     

Automated Patient Discomfort Detection Using Deep Learning

The Internet of Things (IoT) has been transformed almost all fields of life, but its impact on the healthcare sector has been notable. Various IoT-based sensors are used in the healthcare sector and offer quality and safe care to patients. This work presents a deep learning-based automated patient discomfort detection system in which patients’ discomfort is non-invasively detected. To do this, the overhead view patients’ data set has been recorded. For testing and evaluation purposes, we investigate the power of deep learning by choosing a Convolution Neural Network (CNN) based model. The model uses confidence maps and detects 18 different key points at various locations of the body of the patient. Applying association rules and part affinity fields, the detected key points are later converted into six main body organs. Furthermore, the distance of subsequent key points is measured using coordinates information. Finally, distance and the time-based threshold are used for the classification of movements associated with discomfort or normal conditions. The accuracy of the proposed system is assessed on various test sequences. The experimental outcomes reveal the worth of the proposed system’ by obtaining a True Positive Rate of 98% with a 2% False Positive Rate.     

三维人体扫描技术在服装产品设计中的应用

目的为适应数字经济下服装产品设计领域的转型发展需求,探讨三维人体扫描技术在现代服装产品设计开发中的应用,为新型服装产品设计模式提供思维方向。方法分析目前三维人体扫描技术的现状,实验和筛选三维扫描的方法并快速获取人体模型数据样本。在此基础上,依据服装产品设计过程展开分析,结合实际应用案例,探讨基于三维人体扫描技术获取的人体模型数据在服装产品设计中的应用与价值。结论通过三维人体扫描技术可以快速获取丰富准确的三维人体模型数据,根据设计需求将数据接入服装产品设计和展演系统中,辅助企业与设计师提高服装产品的合体性与舒适度,提升以消费者为中心的个性化服装产品定制服务的满意度,驱动服装产品设计领域的智能化、数据化和信息化发展。     

An analytical modelling technique for predicting the stiffness of 3-D orthotropic laminated fabric composites

A new analytical modelling approach for the prediction of the stiffness of 3-D orthotropic laminated composites is given. The composite, which consists of stacked orthotropic layers which are in turn composed of a number of parallel unidirectional stripes, is assumed to be homogeneous and orthotropic macroscopically. The technique introduced is to discretise the representative unit cell of the composite into slices (layers) and then stripes (elements). The stiffness of each slice can then be obtained under the condition of isostrain or isostress. The final stiffness of the composite is formulated analytically by combining these slices. The model eliminates the inconsistency between macro- and micro-level strains and gives more realistic distributions of strain for the representative unit cell. The results demonstrate that the present model, which is both simple and computationally efficient, can give a very accurate prediction compared with data from experiments and some existing models.     

A Recommendation System Based on Fusing Boosting Model and DNN Model

In recent years, the models combining traditional machine learning with the deep learning are applied in many commodity recommendation practices. It has been proved better performance by the means of the neural network. Feature engineering has been the key to the success of many click rate estimation model. As we know, neural networks are able to extract high-order features automatically, and traditional linear models are able to extract low-order features. However, they are not necessarily efficient in learning all types of features. In traditional machine learning, gradient boosting decision tree is a typical representative of the tree model, which can construct new features related before and after tree. Convolutional neural networks have a better perception of local features. In this paper, we take advantage of convolutional networks to capture the local features. The features are constructed by the node leaf of gradient boosting decision tree. This paper employs the tree leaf node to mine the user behavior path features, and uses the deep model to extract the user abstract features. Based on a Kaggle competition, our model performs better in the test data than any other model.     

Vectorial modal analysis of dielectric waveguides based on a coupled transverse-mode integral equation. I. Mathematical formulation

We propose a rigorous full-vector integral-equation formulation for analyzing modal characteristics of the complex, two-dimensional, rectangular-like dielectric waveguide that is divisible into vertical slices of one-dimensional layered structures. The entire electromagnetic mode field is completely determined by the y-component electric and magnetic field functions on the interfaces between slices. These interfacial functions are governed by a system of vector-coupled transverse-mode integral equations (VCTMIE) whose kernels are made of orthonormal sets of both TE-to-y and TM-to-y modes from each slice. To solve for the unknown functions, we construct sets of suitable expansion functions and turn VCTMIE into a nonlinear matrix equation via orthogonal projection. The eigenvectors of the matrix provide the mode field solutions of the complex dielectric waveguide.     

Automatic coarse registration of three-dimensional surfaces by information theoretic selection of salient points

We describe a new method to register surface data measured by optical three-dimensional (3-D) sensors from various views of an object. With our method, complete 3-D models of objects can be generated without user interaction. Circumferential acquisition of 3-D objects is done by taking several views from different directions. To generate a complete 3-D-model, the views must be aligned with each other. This process is called registration and is commonly done interactively by searching for so-called corresponding points in the different views and by use of these points to calculate the appropriate rotation and translation. Our approach is based on automatically finding points that are eye catching or salient compared with other surface points. We derive a quantitative measure of point salience and a feature definition for free-form surfaces by introducing a concept to measure pragmatic information. Experiments confirm that our salient points can be robustly located on general free-form surfaces, even if there are no corners or edges. Furthermore, the neighborhoods of the salient points are highly distinguishable from each other. This results in a large reduction of the complexity of the subsequent geometric matching. The computing time is only a few seconds. We present results from various fields of application.     

A Multilevel Deep Feature Selection Framework for Diabetic Retinopathy Image Classification

Diabetes or Diabetes Mellitus (DM) is the upset that happens due to high glucose level within the body. With the passage of time, this polygenic disease creates eye deficiency referred to as Diabetic Retinopathy (DR) which can cause a major loss of vision. The symptoms typically originate within the retinal space square in the form of enlarged veins, liquid dribble, exudates, haemorrhages and small scale aneurysms. In current therapeutic science, pictures are the key device for an exact finding of patients’ illness. Meanwhile, an assessment of new medicinal symbolisms stays complex. Recently, Computer Vision (CV) with deep neural networks can train models with high accuracy. The thought behind this paper is to propose a computerized learning model to distinguish the key precursors of Dimensionality Reduction (DR). The proposed deep learning framework utilizes the strength of selected models (VGG and Inception V3) by fusing the extracated features. To select the most discriminant features from a pool of features, an entropy concept is employed before the classification step. The deep learning models are fit for measuring the highlights as veins, liquid dribble, exudates, haemorrhages and miniaturized scale aneurysms into various classes. The model will ascertain the loads, which give the seriousness level of the patient’s eye. The model will be useful to distinguish the correct class of seriousness of diabetic retinopathy pictures.     

Neural network approach to segmentation of magnetic resonance head images

A novel image segmentation scheme based on a neural network has been implemented to segment magnetic resonance head images. A three-layer perceptron-type neural network, trained with backward error propagation algorithm was used. The scheme utilizes first-echo intensity and computed T₂ values to construct a two-parameter space for classification. After training on a selected slice, the method successfully segments all slices for a given subject without any further human interaction.     

Sampling 3-D seismic data

A 3-D seismic survey is usually achieved by recording a parallel profile network. The 3-D data thus obtained are sampled and processed in a cubic grid for which the sampling requirements are generally derived from the usual 1-D viewpoint. The spectrum of 3-D seismic data has a support (the region of the Fourier space in which the spectrum is not zero) that can be approximated by a domain bounded by two cones. Considering the particular shape of this support, we use a 3-D sampling theory to obtain results applicable to the recording and processing of 3-D seismic data. This naturally leads to weaker sampling requirements than the 1-D viewpoint does. We define the hexagonal non-cubic sampling grid and the triangular non-cubic sampling grid and show that fewer sample points are needed to represent 3-D seismic data with the same degree of accuracy. Thus, using the hexagonal non-cubic sampling grid we point out that the maximum value of the spatial sampling interval along the profiles is larger by 15.6% than the one of the cubic sampling grid. We also point out that the triangular non-cubic sampling grid requires a number of sample points equal to half the number required by a cubic sampling grid.     

Using a 3-D multicellular simulation of spinal cord injury with live cell imaging to study the neural immune barrier to nanoparticle uptake

Development of nanoparticle (NP) based therapies to promote regeneration in sites of central nervous system (CNS;i.e.brain and spinal cord) pathology relies critically on the availability of experimental models that offer biologically valid predictions of NP fate in vivo.However,there is a major lack of biological models that mimic the pathological complexity of target neural sites in vivo,particularly the responses of resident neural immune cells to NPs.Here,we have utilised a previously developed in vitro model of traumatic spinal cord injury (based on 3-D organotypic slice arrays) with dynamic time lapse imaging to reveal in real-time the acute cellular fate of NPs within injury foci.We demonstrate the utility of our model in revealing the well documented phenomenon of avid NP sequestration by the intrinsic immune cells of the CNS (the microglia).Such immune sequestration is a known translational barrier to the use of NP-based therapeutics for neurological injury.Accordingly,we suggest that the utility of our model in mimicking microglial sequestration behaviours offers a valuable investigative tool to evaluate strategies to overcome this cellular response within a simple and biologically relevant experimental system,whilst reducing the use of live animal neurological injury models for such studies.     

采用仿射变换的红外与可见光图像配准方法

图像配准是图像融合技术的基本环节和首要问题。针对红外和可见光图像的配准问题,分析了两者配准的特点。引入6参数的仿射变换模型来描述3D空间中的目标,分析了通常采用的4参数模型的不完备性。利用模型计算配准参数,并用最小二乘法达到用多个配准点拟合出最优参数解的目的。实验结果表明,此方法可以很好地消除红外和可见光图像之间的差异。     

Classification and Research of Skin Lesions Based on Machine Learning

Classification of skin lesions is a complex identification challenge. Due to the wide variety of skin lesions, doctors need to spend a lot of time and effort to judge the lesion image which zoomed through the dermatoscopy. The diagnosis which the algorithm of identifying pathological images assists doctors gets more and more attention. With the development of deep learning, the field of image recognition has made longterm progress. The effect of recognizing images through convolutional neural network models is better than traditional image recognition technology. In this work, we try to classify seven kinds of lesion images by various models and methods of deep learning, common models of convolutional neural network in the field of image classification include ResNet, DenseNet and SENet, etc. We use a fine-tuning model with a multi-layer perceptron, by training the skin lesion model, in the validation set and test set we use data expansion based on multiple cropping, and use five models’ ensemble as the final results. The experimental results show that the program has good results in improving the sensitivity of skin lesion diagnosis.     

Acquisition of customer’s tailor measurements for 3D clothing design using virtual reality devices

Over recent years, various virtual prototyping technologies have been developed to innovate apparel industry. For each step of the garment design process one can find dedicated tools (from body acquisition to garment modelling and simulation) with the aim of making the process easier and faster. However, most of them are based on expensive solutions both for hardware and software systems. In this paper, we focus the attention on the first step of the made-to-measure garment design, i.e. customer’s measures acquisition. We present a plug-in, named Tailor Tracking, which permits to get the measurements by interacting with the customer’s avatar using hands as in the traditional way. Tailor Tracking has been developed using low cost devices, such as Microsoft Kinect sensor, Leap motion device and Oculus Rift, and open source libraries, such as Visualisation Toolkit (VTK) and Qt. The proposed approach is based on the use of multiple Kinect v2 to simultaneously acquire both customer’s body and motion. This permits to emulate the customer’s postures required to take the correct measurements. In addition, a virtual measuring tape is made available to replicate the one commonly used by the tailor. A men shirt has been considered as case study and a tailor and 14 people with no skills in garment design and different levels of experience in virtual reality technology have been involved to preliminary test Tailor tracking. Finally, tests as well as results reached so far are presented and discussed. Results have been considered quite good; however, some critical measures have been identified as well as future developments. Anyway, Tailor Tracking can represent an alternative solution to the existing approaches that automatically extract anthropometric measures from the customer’s avatar.     

基坑开挖数值模拟中土体本构模型的选取

基坑开挖数值分析中的一个关键问题是选取一个合适的土体本构模型。该文通过对基坑开挖过程中土体的主要应力变化路径进行分析, 指出开挖条件下的土体本构模型应能合理考虑土体变形特性的应力路径相关性和压硬性。在介绍与分析几种常用土体本构模型特点的基础之上, 通过模拟应力路径三轴试验、基坑工程算例与工程实例的对比分析, 探讨了常见土体本构模型的适用性。分析表明, Hardening Soil Model采用了不同的加荷与卸荷模量, 能够反映土体应力路径的影响, 且考虑了土体模量的应力水平相关性, 能预测得到较合理的坑壁侧移、地表沉降以及支护结构的内力, 因而建议采用Hardening Soil Model进行基坑开挖数值模拟分析。     

Image Recognition of Citrus Diseases Based on Deep Learning

In recent years, with the development of machine learning and deep learning, it is possible to identify and even control crop diseases by using electronic devices instead of manual observation. In this paper, an image recognition method of citrus diseases based on deep learning is proposed. We built a citrus image dataset including six common citrus diseases. The deep learning network is used to train and learn these images, which can effectively identify and classify crop diseases. In the experiment, we use MobileNetV2 model as the primary network and compare it with other network models in the aspect of speed, model size, accuracy. Results show that our method reduces the prediction time consumption and model size while keeping a good classification accuracy. Finally, we discuss the significance of using MobileNetV2 to identify and classify agricultural diseases in mobile terminal, and put forward relevant suggestions.     

服装样片二、三维转换算法

服装样片二、三维之间进行转换的算法，使得服装样片二、三维转换得到统一实现。服装样片二维到三维和三维到二维的转换，满足转换前后，二维样片和三维样片的面积应近似相等，边界线相互对应且长度保持相等，关键点的位置应相互对应的条件。此过程应用Delaunay三角化技术、散乱数据插值技术，目的就是求出二、三维服装样片之间相互对应的中间点。最后利用可视化技术，给出了转换结果的直观效果，验证了该方法的可实施性。     

Semi-three-dimensional algorithm for time-resolved diffuse optical tomography by use of the generalized pulse spectrum technique

Although a foil three-dimensional (3-D) reconstruction with both 3-D forward and inverse models provide, the optimal solution for diffuse optical tomography (DOT), because of the 3-D nature of photon diffusion in tissue, it is computationally costly for both memory requirement and execution time in a conventional computing environment. Thus in practice there is motivation to develop an image reconstruction algorithm with dimensional reduction based on some modeling approximations. Here we have implemented a semi-3-D modified generalized pulse spectrum technique for time-resolved DOT, where a two-dimensional (2-D) distribution of optical properties is approximately assumed, while we retain 3-D distribution of photon migration in tissue. We have validated the proposed algorithm by reconstructing 3-D structural test objects from both numerically simulated and experimental date. We demonstrate our algorithm by comparing it with the calibrated 2-D reconstruction that is in widespread use as a shortcut to 3-D imaging and proving that the semi-3-D algorithm outperforms the calibrated 2-D algorithm.     

A Linear Algorithm for Tracing Magnet Position and Orientation by Using Three-Axis Magnetic Sensors

For medical diagnoses and treatments, it is often desirable to wirelessly trace an object that moves inside the human body. A magnetic tracing technique suggested for such applications uses a small magnet as the excitation source, which does not require the power supply and connection wire. It provides good tracing accuracy and can be easily implemented. As the magnet moves, it establishes around the human body a static magnetic field, whose intensity is related to the magnet's 3-D position and 2-D orientation parameters. With magnetic sensors, these magnetic intensities can be detected in some predetermined spatial points, and the position and orientation parameters can be computed. Typically, a nonlinear optimization algorithm is applied to such a problem, but a linear algorithm is preferable for faster, more reliable computation, and lower complexity. In this paper, we propose a linear algorithm to determine the 5-D magnet's position and orientation parameters. With the data from five (or more) three-axis magnetic sensors, this algorithm results in a solution by the matrix and algebra computations. We applied this linear algorithm on the real localization system, and the results of simulations and real experiments show that satisfactory tracing accuracy can be achieved by using a sensor array with enough three-axis magnetic sensors.