Contour Points Research Articles

Một kỹ thuật xác định điểm bao phục vụ đánh giá sự thay đổi của vùng bất thường trên ảnh y tế

Assessing the growth or reduction of abnormal areas in medical imaging in general and tomography in particular is a greatly concerned and important research issue in recent years in Vietnam. The article presents a technique of identifying contour points of an abnormal area as a basis for later assessment of the development of abnormal area on medical imaging. The technique is based on the analysis of local structures in the vicinity of abnormal area boundaries combined with the use of convolutional neural networks and has been tested and evaluated based on 3D-IRCADb-01 sample data set with liver tumors.

Toward a Machine Learning Based Framework for Improved Automatic Segmentation of Head and Neck: A Single Institution Implementation

Manual delineation of organs at risk (OAR) is a time-consuming step in the radiation therapy treatment planning workflow. Automatic contouring is used in some clinics to streamline this process. Deformable atlas algorithms are presently the most prevalent method of automatic contouring, but the results often require significant revision. We aimed to show that, for head and neck cancer patients, deep learning methods can train a convolutional neural network to generate more accurate contours when compared to manually drawn expert contours than atlas methods, reducing the manual correction required. A total of 143 cases of head and neck CT scans and manually drawn expert contours from a single institution’s database were utilized in this study: 128 cases to train a CNN based on a modified U-Net architecture, and 15 cases for validation. Data augmentation and dropout layers were used to prevent overfitting. The model trained over 200 epochs for the following OARs: brainstem, cochlea, and the submandibular and parotid glands. We then used the trained model to generate contour data for these OARs on the 15 test cases. All generated contours were written to DICOM format. A commercial atlas-based program, which has been used clinically at the same institution, was also used to generate contour sets for the test cases. Both the atlas-based contours and deep learning-based contours (DLBC) were measured against the manually drawn expert contours as the “ground truth”. The accuracy performance was validated using three metrics: Dice Similarity Coefficient (DSC), which measures area overlap, Mean Surface Distance (MSD), and 95% Hausdorff Distance (HD), defined as the separation distance from ground truth that encompasses 95% of contour points in the automatic contour. For both the parotid and submandibular glands, the DLBC were more closely aligned with the ground truth contours than the atlas contours for all evaluation metrics (p-value < 0.005). For the parotid gland, the mean DLBC improvement over atlas for DSC, MSD, and HD was 17% ± 1.9%, 44% ± 13%, and 37% ± 15%, respectively. For the submandibular, the mean improvement was 58% ± 15%, 54% ± 19%, and 45% ± 15%, respectively. For the brainstem, the DLBC scored comparably to the atlas contours on all three metrics with no significant improvement. For the cochlea, the DSC scores for both methods were comparable, while the MSD and HD scores of the DLBC were superior to the atlas contours (p-value < 0.015), with mean improvement of 38% ± 13% and 33% ± 10% respectively. The measurably more accurate contours generated for the parotid gland, submandibular gland, and cochlea by the neural network would require less correction by a reviewer than those generated by the atlas method used in the clinic. The contours generated by the neural network for the brainstem were comparable. Expanding the neural network to define every OAR could offer a superior method of automatic contour generation for use in the clinic.

\u201cTopographic Shift\u201d: a new digital approach to evaluating topographic changes of the female breast

PurposeTo assess precise topographic changes of the breast, objective documentation and evaluation of pre- and postoperative results are crucial. New technologies for mapping the body using digital, three-dimensional surface measurements have offered novel ways to numerically assess the female breast. Due to the lack of clear demarcation points of the breast contour, the selection of landmarks on the breast is highly dependent on the examiner, and, therefore, is prone to error when conducting before-after comparisons of the same breast. This study describes an alternative to volumetric measurements, focusing on topographic changes of the female breast, based on three-dimensional scans.MethodThe study was designed as an interventional prospective study of 10 female volunteers who had planned on having aesthetic breast augmentation with anatomical, textured implants. Three dimensional scans of the breasts were performed intraoperatively, first without and then with breast implants. The topographic change was determined as the mean distance between two three-dimensional layers before and after augmentation. This mean distance is defined as the Topographic Shift.ResultsThe mean implant volume was 283 cc (SD = 68.6 cc, range = 210–395 cc). The mean Topographic Shift was 7.4 mm (SD = 1.9 mm, range = 4.8–10.7 mm). The mean Topographic Shifts per quadrant were: I: 8.0 mm (SD = 3.3 mm); II: 9.2 mm (SD = 3.1 mm); III: 6.9 mm (SD = 3.5 mm); IV: 1.9 mm (SD = 4.3 mm).ConclusionThe Topographic Shift, describing the mean distance between two three-dimensional layers (for example before and after a volume changing therapy), is a new approach that can be used for assessing topographic changes of a body area. It was found that anatomical, textured breast implants cause a topographic change, particularly on the upper breast, in quadrant II, the décolleté.

Boundary loss for highly unbalanced segmentation.

Widely used loss functions for CNN segmentation, e.g., Dice or cross-entropy, are based on integrals over the segmentation regions. Unfortunately, for highly unbalanced segmentations, such regional summations have values that differ by several orders of magnitude across classes, which affects training performance and stability. We propose a boundary loss, which takes the form of a distance metric on the space of contours, not regions. This can mitigate the difficulties of highly unbalanced problems because it uses integrals over the interface between regions instead of unbalanced integrals over the regions. Furthermore, a boundary loss complements regional information. Inspired by graph-based optimization techniques for computing active-contour flows, we express a non-symmetric L2 distance on the space of contours as a regional integral, which avoids completely local differential computations involving contour points. This yields a boundary loss expressed with the regional softmax probability outputs of the network, which can be easily combined with standard regional losses and implemented with any existing deep network architecture for N-D segmentation. We report comprehensive evaluations and comparisons on different unbalanced problems, showing that our boundary loss can yield significant increases in performances while improving training stability. Our code is publicly available1.

An Occlusion‐aware Edge‐Based Method for Monocular 3D Object Tracking using Edge Confidence

AbstractWe propose an edge‐based method for 6DOF pose tracking of rigid objects using a monocular RGB camera. One of the critical problem for edge‐based methods is to search the object contour points in the image corresponding to the known 3D model points. However, previous methods often produce false object contour points in case of cluttered backgrounds and partial occlusions. In this paper, we propose a novel edge‐based 3D objects tracking method to tackle this problem. To search the object contour points, foreground and background clutter points are first filtered out using edge color cue, then object contour points are searched by maximizing their edge confidence which combines edge color and distance cues. Furthermore, the edge confidence is integrated into the edge‐based energy function to reduce the influence of false contour points caused by cluttered backgrounds and partial occlusions. We also extend our method to multi‐object tracking which can handle mutual occlusions. We compare our method with the recent state‐of‐art methods on challenging public datasets. Experiments demonstrate that our method improves robustness and accuracy against cluttered backgrounds and partial occlusions.

Monitoring ambient vibration pollution based on visual information perception and neural network analysis

As a type of pollution, low-frequency vibrations in the environment have a negative effect on human life. To monitor pollution without extra complications, we propose a novel method based on visual information perception and neural network analysis. Image frame sequences can capture the state of objects, and changes in pixel values at salient vibration points may indirectly reflect ambient vibrations. In this paper, a test environment is established to simulate the influence of vibrations on daily life. A CCD camera is used to continuously sample objects via image frame sequences to monitor potential vibration pollution in the current environment through image processing and neural network analysis. During the image processing stage, a combination of image filtering and edge extraction can accurately locate the test objects’ contours. During the low-frequency vibration monitoring stage, the neural network analyzes the changes in pixel values at the contour points to monitor ambient vibration pollution. The network combines local and global features for vibration frequency classification and prediction. The proposed method's superiority is verified by the test results and a comparison of four different networks. The results demonstrate that this method accurately locates salient vibration points to monitor ambient vibration pollution.

Chinese Sign Language Recognition Based on DTW-Distance-Mapping Features

Sign language is an important communication tool between the deaf and the external world. As the number of the Chinese deaf accounts for 15% of the world, it is highly urgent to develop a Chinese sign language recognition (CSLR) system. Recently, a novel phonology- and radical-coded CSL, taking advantages of a limited and constant number of coded gestures, has been preliminarily verified to be feasible for practical CSLR systems. The keynote of this version of CSL is that the same coded gesture performed in different orientations has different meanings. In this paper, we mainly propose a novel two-stage feature representation method to effectively characterize the CSL gestures. First, an orientation-sensitive feature is extracted regarding the distances between the palm center and the key points of the hand contour. Second, the extracted features are transformed by a dynamic time warping- (DTW-) based feature mapping approach for better representation. Experimental results demonstrate the effectiveness of the proposed feature extraction and mapping approaches. The averaged classification accuracy of all the 39 types of CSL gestures acquired from 11 subjects exceeds 93% for all the adopted classifiers, achieving significant improvement compared to the scheme without DTW-distance-mapping.

Comparison of geomagnetic aided navigation algorithms for hypersonic vehicles

In this paper, we simulate, verify, and compare the performance of three classical geomagnetic matching aided navigation algorithms to assess their applicability to hypersonic vehicle navigation. Firstly, we introduce the various sources of the geomagnetic field. Secondly, we describe the principles and processes of the geomagnetic contour matching (MAGCOM) algorithm, iterative closest contour point (ICCP) algorithm, and Sandia inertial magnetic aided navigation (SIMAN) algorithm. Thirdly, we discuss the principles of inertial/geomagnetic integrated navigation, and propose the state and observation equations of integrated navigation. Finally, we perform a simulation of inertial/geomagnetic integrated navigation on the hypersonic boost-glide vehicle trajectory. The simulation results indicate that the real-time performance of the SIMAN algorithm can be optimized such that the matching accuracy is higher than that of the other two algorithms. At the same time, the SIMAN algorithm can achieve better stability, and though the amount of measurement noise can be larger, it can still achieve good positioning accuracy.

3D-MRI Brain Tumor Detection Model Using Modified Version of Level Set Segmentation Based on Dragonfly Algorithm

Accurate brain tumor segmentation from 3D Magnetic Resonance Imaging (3D-MRI) is an important method for obtaining information required for diagnosis and disease therapy planning. Variation in the brain tumor’s size, structure, and form is one of the main challenges in tumor segmentation, and selecting the initial contour plays a significant role in reducing the segmentation error and the number of iterations in the level set method. To overcome this issue, this paper suggests a two-step dragonfly algorithm (DA) clustering technique to extract initial contour points accurately. The brain is extracted from the head in the preprocessing step, then tumor edges are extracted using the two-step DA, and these extracted edges are used as an initial contour for the MRI sequence. Lastly, the tumor region is extracted from all volume slices using a level set segmentation method. The results of applying the proposed technique on 3D-MRI images from the multimodal brain tumor segmentation challenge (BRATS) 2017 dataset show that the proposed method for brain tumor segmentation is comparable to the state-of-the-art methods.

A Contour Angle Orientation for Power Equipment Infrared and Visible Image Registration

Automatic registration for infrared, and visible images of power equipment has become a challenging work in intelligent diagnosis system of the power grid. Existing registration methods usually fail in accurately aligning power equipment infrared, and visible images because of resolutions, spectrums, and viewpoints differences. To solve this problem, we propose a novel main orientation of feature points named contour angle orientation (CAO), and describe an automatic infrared, and visible image registration method named CAO-Coarse to Fine (CAO-C2F). CAO is based on the contour feature of images, and invariant to images viewpoints, and scales differences. C2F is a feature matching method to obtain correct matches. Our proposed CAO-C2F method includes four steps. First, feature points in contours are extracted by the curvature scale space (CSS) corner detector based on local, and global curvature. Second, the CAO of each feature point is computed as the main orientation. Third, modified scale-invariant feature transform (SIFT) descriptors on the main orientations are extracted, and matched by bilateral matching. Finally, accurate matches are obtained by applying the C2F method. Registration experiments on a self-established images dataset show our proposed CAO-C2F method accurately aligns images, and outperforms other state-of-arts in terms of precision, recall, and root-mean-square error.

SOME ASPECTS OF CREATING GEODETIC SUPPORT FOR A THREE-DIMENSIONAL REAL ESTATE CADASTRE

The paper considers the problems that arise when creating a geodesic justification of a three-dimensional real estate cadastre and, the lack of justification for the necessary accuracy of determining the third coordinate of characteristic points of contours of buildings and structures. An information and analytical review of the current legal framework in the construction sector, used in our country at the stage of design, construction and operation of buildings and structures, which regulates the main requirements for the creation of project documentation used in the future when forming technical plans for objects of capital construction (OCC) as the basis for information entered in the Unified State Register of Real Estate (USRRE). At the same time, more attention is paid to ensuring the accuracy of geometric parameters in construction.

Regularizing algorithm with an adaptive stabilizer for the image-restoration problem

This paper proposes and appraises what we believe to be a new design for a two-dimensional stabilizer that makes it possible to use iterations to refine the geometrical parameters of segments on an image. The algorithm makes it possible to compute the contour points of the boundaries of the segments at the final stages of the calculations and implements a simple effective procedure for accomplishing this. The results of using the algorithm are demonstrated: simultaneous restoration of an image with the segments that exist on the image and images of the restored contours of the segments on the image.

High-quality vascular modeling and modification with implicit extrusion surfaces for blood flow computations

Background and ObjectiveHigh-quality vascular modeling is crucial for blood flow simulations, i.e., computational fluid dynamics (CFD). As without an accurate geometric representation of the smooth vascular surface, it is impossible to make meaningful blood flow simulations. The purpose of this work is to develop high-quality vascular modeling and modification method for blood flow computations. MethodsWe develop a new technique for the accurate geometric modeling and modification of vasculatures using implicit extrusion surfaces (IES). In the proposed method, the skeleton of the vascular structure is subdivided into short curve segments, each of which is then represented implicitly locally as the intersection of two mutually orthogonal implicit surfaces defined by distance functions. A set of contour points is extracted and fitted with an implicit curve for accurately specifying the vessel cross-section profile, which is then extruded locally along the skeleton to fill the gaps between two vascular tube cross sections. We also present a new implicit geometric editing technique to modify the constructed vascular model with pathology for virtual stenting. ResultsExperimental results and validations show that accurate vascular models with highly smooth surfaces can be generated by the proposed method. In addition, we conduct some blood flow simulations to indicate the effectiveness of proposed method for hemodynamic simulations. ConclusionsThe proposed technique can achieve precise geometric models of vasculatures with any required degree of smoothness for reliable blood flow simulations.

A numerical method for calculating round slabs based on generalized equations of the finite difference method

Round slabs are widely used in construction equipment. Currently, many different special structures are being built that have a circular shape in the plan. The article provides an analysis of thin isotropic circular slab in pure bending, the action of concentrated forces and slab loaded by a distributed load, supported at the centre. Different boundary conditions are considered. The initial differential equations of slab bending in polar coordinates are approximated by generalized equations of the finite difference method (MCR). The algorithm allows you to take into account the final discontinuities of the desired function, the right part of the original differential equations, as well as discontinuities of derivatives of these functions without thickening the grid and using contour points. It should be noted that for the Central point of the plate, differential bending equations and their numerical approximation in Cartesian coordinates are used, which makes it possible to determine the forces and movements at this point at any load.The calculation algorithm is reduced to solving a system of differential-algebraic equations composed for field and contour points. The results of the calculation of slabs for these effects are presented. The reliability of the solution is confirmed by the convergence of results on some grids when the step is reduced, by comparing some solutions with existing ones, and by performing static checks. The advantage of this method is the ability to implement solutions with a small number of partitions.

Construction of an axonometric outline of a curved surface of a part

The article considers the geometric construction of the orthogonal isometry of a curved surface. To solve this problem, the authors propose to perform additional construction of the contour of the section of the surface in the orthogonal drawing and use the coordinates of the points of this section to construct an axonometric outline of the surface. The mismatch between the axonometric contour of the rotation surface and the corresponding contour of this surface in the projection drawing is illustrated. As a result, the proposed solution allows geometrically accurate construction of support and additional points of the axonometric contours of the revolution surfaces.

Crane Hook Detection Based on Mask R-CNN in Steel-making Plant

This paper proposes a solution based on machine vision and deep learning for the hidden safety problems that the hook does not match the ladle trunnion correctly, which is one of the hidden dangers during the crane lifting the ladle. Mask Region Convolutional Neural Network (Mask-RCNN) was introduced to segment the crane hook and find the bottom point of the hook contour. The trunnion center point can be directly located in image by painting a special color on it. Then determine whether the hook and trunnion match correctly by calculating the angle between the horizontal line and the line connecting the bottom point of the hook and the trunnion center point. According to the experimental rsulrs of 100 test images, that the average accuracy (AP) of our methods for hook segmentation can reach 92%. And the accuary of the safety judgment algorithm has reached 96%.

Process to assess rheological characteristics of rocks by underground geodesic data in terms of a viscoelastic model

Under consideration is the process for evaluation of rheological properties of construction components in the underground chamber-and-pillar system to mine solid mineral reserves. The process is based on solving the inverse mixed-type problems in terms of a 2D viscoelastic geomedium model. Feasibility is demonstrated to evaluate in-situ state parameters characterizing rheological deformation of rocks. Numerical experiments with the use of synthetic input data (viz. relative displacements of the worked-out space contour points recorded by underground geodesic monitoring) for the prescribed absolute precision of the instruments made it possible to establish a required population of measurement data to provide an inverse problem-solving procedure.

Morphological reconstruction method of irregular shaped ballast particles and application in numerical simulation of ballasted track

Abstract To quantitatively investigate the morphological features of railway ballast, a new statistical index, Curvature Index (CI), was presented. With a set of 584 digitized railway ballast particles obtained through 3D scanning, the statistical distribution functions of CI and existing global shape indices (i.e., long axis, middle axis, short axis and sphericity index) were obtained. Then, based on Proper Orthogonal Decomposition (POD) and Radial Basis Function (RBF) Neural Network, a new shape reconstruction method was proposed to generate an arbitrary number of ballast particles that met the desired probability density distribution of morphological indices. And based on local curvature distribution on particle surface, a new simplification algorithm was implemented to reduce the number of contour points of each particle to no more than 31 on the premise of preserving global and local morphological features, which improved the efficiency of numerical simulations. These methods were used to reconstruct and simplify the arbitrary shapes of gravel ballast particles. Then, a DEM-FEM coupling model of ballasted track-subgrade was established by applying the regenerated ballast particles to analyze the contact stress at the ballast-soil interface. The numerical simulations can effectively reflect the contact stress distributions resulted from laboratory tests, which further confirmed that the reconstruction of the ballast particles reflected the morphological characteristics of real ballast.

Quantitative 3-Dimensional Photographic Assessment of Breast Cosmesis After Whole Breast Irradiation for Early Stage Breast Cancer: A Secondary Analysis of a Randomized Clinical Trial

PurposeOur purpose was to use 3-dimensional (3D) surface photography to quantitatively measure breast cosmesis within the framework of a randomized clinical trial of conventionally fractionated (CF) and hypofractionated (HF) whole breast irradiation (WBI); to identify how 3D measurements are associated with patient- and physician-reported cosmesis; and to determine whether objective measures of breast symmetry varied by WBI treatment arm or transforming growth factor β 1 (TGFβ1) status. Methods and MaterialsFrom 2011 to 2014, 287 women age ≥40 with ductal carcinoma in situ or early-stage invasive breast cancer were enrolled in a multicenter trial and randomized to HF-WBI or CF-WBI with a boost. Three-dimensional surface photography was performed at 3 years posttreatment. Patient-reported cosmetic outcomes were recorded with the Breast Cancer Treatment Outcome Scale. Physician-reported cosmetic outcomes were assessed by the Radiation Therapy Oncology Group scale. Volume ratios and 6 quantitative measures of breast symmetry, termed F1-6C, were calculated using the breast contour and fiducial points assessed on 3D surface images. Associations between all metrics, patient- and physician-reported cosmesis, treatment arm, and TGFβ1 genotype were performed using the Kruskal-Wallis test and multivariable logistic regression models. ResultsAmong 77 (39 CF-WBI and 38 HF-WBI) evaluable patients, both patient- and physician-reported cosmetic outcomes were significantly associated with the F1C vertical symmetry measure (both P < .05). Higher dichotomized F1C and volumetric symmetry measures were associated with improved patient- and physician-reported cosmesis on multivariable logistic regression (both P ≤ .05). There were no statistically significant differences in vertical symmetry or volume measures between treatment arms. Increased F6C horizontal symmetry was observed in the CF-WBI arm (P = .05). Patients with the TGFβ1 C-509T variant allele had lower F2C vertical symmetry measures (P = .02). ConclusionsQuantitative 3D image-derived measures revealed comparable cosmetic outcomes with HF-WBI compared with CF-WBI. Our findings suggest that 3D surface imaging may be a more sensitive method for measuring subtle cosmetic changes than global patient- or physician-reported assessments.

Revisiting spectral clustering for near-convex decomposition of 2D shape

We present a novel 2D shape decomposition algorithm via a recursive partitioning process. Starting with the contour points of a shape, we repeatedly separate the points into two parts by spectral clustering, until the stopping condition is met. Motivated by the fact that the points in a convex part are mutually visible, we regard the visibility matrix of points as the affinity matrix of spectral clustering to obtain a near-convex decomposition. Additionally, we present an efficient stopping rule to avoid over-segmentation on the shape branches. The stopping criterion is based on a novel shape signature called visible protrusion strength which can be used to measure the segmentability of a sub-shape. Finally, we demonstrate the efficiency of our algorithm on a variety of publicly available shapes, and provide qualitative and quantitative comparisons with state-of-art approaches.