Three-dimensional Registration Research Articles

Longitudinal bone microarchitectural changes are best detected using image registration.

The purpose of this study was to determine which longitudinal analysis technique (no registration (NR), slice-match (SM) registration, or three-dimensional registration (3DR)) produced the most realistic longitudinal changes in a 3-year study of bone density and structure using high-resolution peripheral quantitative computed tomography (HR-pQCT). We assessed HR-pQCT scans of the distal radius and tibia for men and women (N = 40) aged 55-70years at baseline and 6, 12, 24, and 36months. To evaluate which longitudinal analysis technique (NR, SM, or 3DR) best captured physiologically reasonable 3-year changes, we calculated the standard deviation of the absolute rate of change in each bone parameter. The data were compared between longitudinal analysis techniques using repeated measures ANOVA and post hoc analysis. As expected, both SM and 3DR better captured physiological longitudinal changes than NR. At the tibia, there were no differences between SM and 3DR; however, at the radius where precision was lower, 3DR produced better results for total bone mineral density. At least SM or 3DR should be implemented in longitudinal studies using HR-pQCT. 3DR is preferable, particularly at the radius, to ensure that physiological changes in bone density are observed.

Toward a robust and fast real-time point cloud registration with factor analysis and Student’s-t mixture model

Three-dimensional (3D) point cloud registration generally involves in unsatisfied situations like Gaussian white noise, data missing and disorder in affine. This paper proposes a robust and real-time point cloud registration, which combines the Student’s-t mixture model (SMM) with factor analysis. The proposed method extending the point cloud mathematical model to the orthogonal factor model and employs the SMM to fit the point cloud data, because the degree of freedom of Student’s t-distribution makes it more flexible in fitting the probability distribution of data. Since the Expectation Maximization (EM) algorithm has a stable estimation ability for the mixture model, the EM algorithm is used to estimate the factor load matrix. The filed data and experimental results show that the proposed algorithm can achieve accurate registration and fast convergence even in the case of point cloud disorder, data occlusion, incomplete loss and noise.

Robust three-dimensional registration on optical coherence tomography angiography for speckle reduction and visualization.

In the clinical applications of optical coherence tomography angiography (OCTA), the repeated scanning and averaging method can provide better contrast with reduced speckle noises in the final results, which are useful for visualizing and quantifying vascular components with high accuracy, reproducibility, and reliability. However, the inevitable patient motion presents a challenge to this method. The objective of this study is to meet this challenge by introducing a 3D registration method to register optical coherence tomography (OCT)/OCTA scans for precise volume averaging of multiple scans to improve the signal-to-noise ratio (SNR) and increase quantification accuracy. The proposed method utilized both rigid affine transformation and non-rigid B-spline transformation in which their parameters were optimized and calculated by the average stochastic gradient descent on OCT structural images. In addition, we also introduced a multi-level resolution approach to further improve the robustness and computational speed of our proposed method. The imaging performance was tested on in vivo imaging of human skin and eye and assessed by SNR, peak signal-to-noise ratio (PSNR) and normalized correlation coefficient (NCC). Five subjects were enrolled in this study for obtaining in vivo images of skin and retina. The proposed registration and averaging method provided substantial improvements of the imaging performance in terms of vessel connectivity and signal to noise ratio. The increase of repeated volume numbers in the averaging improves all the metrics assessed, i.e., SNR, PSNR and NCC. An improvement of the SNR from 10 to 40 dB after 10 repeated volumetric averaging was achieved. The proposed 3D registration and averaging method is effective in reducing speckle noises and suppressing motion artifacts, thereby improving SNR, PSNR and NCC metrics for final averaged images. It is expected that the proposed algorithm would be practically useful in better visualization and more reliable quantification of in vivo OCT and OCTA data, which would be beneficial to OCT clinical applications.

Research on the Method of 3D Registration Technology

Three-dimensional registration technology is the foundation, key and difficult point of constructing augmented reality system. This paper first reviews the three-dimensional registration technology in augmented reality, and compares the advantages and disadvantages of each method. Then, from the improvement of feature descriptor, the combination of registration technology and tracking technology, online learning-based registration technology and deep learning-based registration technology, the improvement strategy of registration method is analysed and summarized, which is the three-dimensional registration technology. In-depth research provides ideas. Finally, the problems existing in the research of 3D registration technology are pointed out, and the future research directions are prospected.

Research on Augmented Reality Method Based on Improved ORB Algorithm

Aiming at the problems of time-consuming three-dimensional registration algorithm in augmented reality system, low accuracy of feature point matching and poor registration effect, an improved ORB (S-ORB) algorithm is proposed and applied to the Majiayao painted pottery augmented reality system. Firstly, the scale space is established by Gaussian filtering and downsampling, and the feature points are extracted by FAST corner detection. Secondly, the descriptors of feature points are calculated by rBFEF algorithm. Finally, the feature matching set is obtained by hamming distance and the mismatching point is eliminated by RANSAC algorithm to complete the three-dimensional registration. Make the virtual information stack correctly added to the real scene. The experimental results show that the S-ORB algorithm is better than the ORB, SURF and BRISK algorithms in time and accuracy when the target position is rotated and scaled. It has better real-time and robustness in the AR system of the painted pottery.

两阶段变尺度三维点云配准算法研究

两阶段变尺度三维点云配准算法研究

Improved Demons Three-Dimensional Registration Algorithm and the Application in Image-Guided Radiotherapy

Improved Demons Three-Dimensional Registration Algorithm and the Application in Image-Guided Radiotherapy

Three-dimensional point cloud registration technique for self-designed LiDAR

Three-dimensional point cloud registration technique for self-designed LiDAR

Validation of Portable Mobile Mapping System for Inspection Tasks in Thermal and Fluid–Mechanical Facilities

The three-dimensional registration of industrial facilities has a great importance for maintenance, inspection, and safety tasks and it is a starting point for new improvements and expansions in the industrial facilities context. In this paper, a comparison between the results obtained using a novel portable mobile mapping system (PMMS) and a static terrestrial laser scanner (TLS), widely used for 3D reconstruction in civil and industrial scenarios, is carried out. This comparison is performed in the context of industrial inspection tasks, specifically in the thermal and fluid-mechanics facilities in a hospital. The comparison addresses the general reconstruction of a machine room, focusing on the quantitative and qualitative analysis of different elements (e.g., valves, regulation systems, burner systems and tanks, etc.). The validation of the PMMS is provided considering the TLS as ground truth and applying a robust statistical analysis. Results come to confirm the suitability of the PMMS to perform inspection tasks in industrial facilities.

In Vivo Kinematics of Bicruciate-Retaining Total Knee Arthroplasty with Anatomical Articular Surface under High-Flexion Conditions.

Bicruciate-retaining total knee arthroplasty (BCR-TKA) recreates normal knee movement by preserving the anterior cruciate and posterior cruciate ligaments. However, in vivo kinematics of BCR-TKA with the anatomical articular surface remains unknown. The objective of this study was to evaluate in vivo kinematics of BCR-TKA with the anatomical articular surface during high-flexion activities. For this purpose, 17 knees after BCR-TKA with an anatomical articular surface were examined. Under fluoroscopy, each patient performed squatting and cross-legged sitting motions. To estimate the spatial position and orientation of the knee, a two-dimensional or three-dimensional registration technique was used. Rotation, varus-valgus angle, and anteroposterior translation of medial and lateral contact points of the femoral component relative to the tibial component were evaluated in each flexion angle. The results showed that from 80 to 110° of flexion, the femoral external rotation during squatting was significantly larger than that during cross-legged sitting. At maximum flexion, the knees during sitting indicated significantly more varus alignment than during squatting. During squatting, a medial pivot pattern was observed from minimum flexion to 10° flexion, with no significant movement beyond 10° of flexion. Conversely, during cross-legged sitting, no significant movement was detected from minimum flexion to 60° of flexion, with a medial pivot beyond 60° of flexion. Therefore, the knees showed relatively normal kinematics after BCR-TKA with an anatomical articular surface; however, it varied during high-flexion activities depending on the activity.

Three-dimensional registration and shape reconstruction from depth data without matching: A PDE approach

The widespread availability of depth sensors like the Kinect camera makes it easy to gather three-dimensional (3D) data. However, accurately and efficiently merging large datasets collected from di¤erent views is still a core problem in computer vision. This question is particularly challenging if the relative positions of the views are not known if there are few or no overlapping points, or if there are multiple objects. Here, we develop a method to reconstruct the 3D shapes of objects from depth data taken from different views whose relative positions are not known. Our method does not assume that common points in the views exist nor that the number of objects is known a priori. To reconstruct the shapes, we use partial differential equations (PDE) to compute upper and lower bounds for distance functions, which are solutions of the Eikonal equation constrained by the depth data. To combine various views, we minimize a function that measures the compatibility of relative positions. As we illustrate in several examples, we can reconstruct complex objects, even in the case where multiple views do not overlap, and, therefore, do not have points in common. We present several simulations to illustrate our method including multiple objects, non-convex objects, and complex shapes. Moreover, we present an application of our PDE approach to object classification from depth data.

Two-dimensional-three-dimensional registration for fusion imaging is noninferior to three-dimensional- three-dimensional registration in infrarenal endovascular aneurysm repair

BackgroundFusion imaging is a tool for intraoperative three-dimensional (3D) guidance in endovascular aneurysm repair (EVAR). In many aortic centers, the registration for location is based on an intraoperative 3D dataset acquired by means of cone-beam computed tomography (3D-3D registration). Another registration method is based on two two-dimensional (2D) images (lateral and posteroanterior) acquired with the use of intraoperative fluoroscopy for registration with a computed tomographic angiogram (2D-3D registration). The aim of the present study was to compare 2D-3D registration with 3D-3D registration regarding noninferiority in accuracy and to describe radiation exposure and ease of use of both modalities. MethodsFrom December 2014 to September 2015, 50 sequentially enrolled patients received EVAR with the use of fusion imaging using 2D-3D registration. No adjustments were made until the first angiography with inserted stent graft. The deviation of fusion imaging to the actual position of the lower renal artery compared with digital subtraction angiography was measured. A historic cohort of 101 patients treated with EVAR and fusion imaging with 3D-3D registration (3D-3D cohort) served as the control group for this study. ResultsCraniocaudal deviation did not differ significantly (4.6 ± 4.4 mm in the 2D-3D cohort vs 3.6 ± 3.9 mm in the 3D-3D cohort; P = .17). The difference of the means was 1.05 mm with a 95% confidence interval of −2.45 to 0.34 and a P value for the noninferiority test of .0249, indicating that 2D-3D registration was noninferior in terms of a margin of δ = 2.5 mm. 2D-3D registration was significantly faster with significantly less additional radiation necessary: 0.45 ± 0.28 vs 45.7 ± 9.1 Gy·cm2 in the 3D-3D cohort (P < .001); 2.3 ± 1.3 vs 5.3 ± 4.3 minutes in the 3D-3D cohort (P < .001). ConclusionsFusion imaging during EVAR with the use of 2D-3D registration is feasible in routine EVAR. Our findings of two consecutive cohorts with the same clinical, hardware, and software setup used for the procedures underscore that the accuracy of 2D-3D registration is noninferior to that of a 3D-3D registration workflow, with advantages in terms of radiation exposure, intraoperative time demand, and ease of use.

Comprehensive description of T2 value spatial variations in non-osteoarthritic femoral cartilage using three-dimensional registration of morphological and relaxometry data

PurposeThe aim of this study was to develop and assess a method of quantifying cartilage T2 relaxation times in a series of volumes of interest (VOIs) covering the entire cartilage of the femoral condyles. Subsequently, the method was used to test for T2 spatial variations in non-osteoarthritic (OA) knees. MethodsTen non-OA subjects (five female, average 30 years) were enrolled after informed consent. Three-dimensional bone and cartilage models were created by double echo steady state (DESS) morphological magnetic resonance image (MRI) segmentation, and the models were semi-manually registered with multi-slice, multi-echo (MSME) T2 MRI. Mean T2 values were calculated for 12 VOIs derived from cartilage thickness literature and their respective superficial and deep layers. ResultsAnalyses showed that intra- and inter-rater reliabilities of the presented method were “good” to “excellent” in more than 90% of the VOIs. Additionally, several spatial differences in T2 values were observed, including, for the medial condyle, higher T2 values in the anterior and central VOIs versus in the posterior VOI (p < .05). T2 values were also generally higher in the superficial versus deep layers (p < .05). ConclusionsThe presented MRI T2 analysis method is reliable and provides a comprehensive quantification of spatial heterogeneity of healthy cartilage compositional properties. This method can be further applied to better understand knee OA pathophysiology and potentially define clinically relevant diagnostic features of the disease.

Combination of radiofrequency ablation and transcatheter arterial chemoembolization to treat hepatocellular carcinoma: measurement of distance from needle tip to nodule for assessment of local tumor progression.

ObjectiveThis study was performed to determine the relationship between the minimum distance from the radiofrequency ablation (RFA) needle tip to the tumor and local tumor progression (LTP) of hepatocellular carcinoma (HCC) nodules and identify prognostic factors for LTP.MethodsWe reviewed 197 patients (197 nodules) who underwent RFA after transcatheter arterial chemoembolization for HCC from January 2010 to January 2015. Three-dimensional registration of images was used to calculate the minimum distance from the tip to the tumor. We then divided the minimum distance into two groups: <2 and ≥2 mm. Contrast-enhanced computed tomography was performed after treatment. The LTP rate was calculated 1 and 3 years after RFA. We performed multivariate analysis to identify independent prognostic factors for LTP.ResultsThe cumulative 1-year LTP rates in the <2- and ≥2-mm groups were 82.7% and 4.3%, respectively, and the cumulative 3-year LTP rates in the two groups were 94.8% and 10.8%, respectively. The minimum distance from the needle tip to the tumor was an independent prognostic factor for LTP.ConclusionsA minimum distance of 2 mm from the needle tip to the tumor should be completely ablated along with the tumor.

A hybrid photogrammetry approach for archaeological sites: Block alignment issues in a case study (the Roman camp of A Cidadela)

Abstract Photogrammetry is a cost-effective and versatile technique used for the three-dimensional (3D) registration of archaeological heritage sites. Managing datasets of heterogeneous images in terms of camera type, elevation platform, position or acquisition time can now be addressed by structure from motion (SfM) software via bundle adjustment in a single block based on collinearity principles. This development enables new possibilities with regard to data completeness assurance for 3D documentation, even for complex sites with occlusive elements and hidden areas. However, hybrid photogrammetry in large datasets often requires multiple photogrammetric blocks that must be processed individually and subsequently aligned to obtain a unified point cloud. In this paper, we discuss the steps required to homogenize the information and the methods used to perform block alignment in these cases. A case study of low-altitude aerial photogrammetry with several cameras and platforms is presented for the Roman camp of A Cidadela in NW Spain as a representative example of an archaeological site that is difficult to survey using a single photogrammetric platform. The relatively large expanse of the area and the fact that it is partially covered by a protective structure constitute an ideal framework for the fusion of multiplatform imagery. The most accurate digital surface model (DSM) was obtained via point-based method fusion, during which subsets are aligned based on automatically extracted tie points (TPs) between the dense point clouds; however, point-based method fusion is very time consuming. When hardware capabilities allow, conducting the process in a single block is preferable, which is a noticeably more accurate procedure than independent block fusion.

基于相关系数平方和最大的三维点云配准

基于相关系数平方和最大的三维点云配准

基于独立成分分析的三维点云配准算法

基于独立成分分析的三维点云配准算法

基于因子分析法的三维点云配准算法

基于因子分析法的三维点云配准算法

Construction and consideration of hospital services continuous improvement based on patients' experience

Patients' experience is the feeling of patients in the process of medical perception.This paper expounded the new concept based on the experience of patients with medical services, then reviewed the specific practices of improve services, which including three-dimensional registration service, process intensification, service public commitment, patients choose doctors in medico-technical departments, hospital president in outpatient.Finally, the article proposed that patients' experience should be based on patients' evaluation, have love heart, exceed customers' expectation and build patients' experience culture. Key words: Patient experience; Hospital services; Continuous improvement

Improving Registration of Augmented Reality by Incorporating DCNNS into Visual SLAM

Augmented reality (AR) by analyzing the characteristics of the scene, the computer-generated geometric information which can be added to the real environment in the way of visual fusion, reinforces the perception of the world. Three-dimensional (3D) registration is one of the core issues of in AR. The key issue is to estimate the visual sensor’s posture in the 3D environment and figure out the objects in the scene. Recently, computer vision has made significant progress, but the registration based on natural feature points in 3D space for AR system is still a severe problem. There is the difficulty of working out the mobile camera’s posture in the 3D scene precisely due to the unstable factors, such as the image noise, changing light and the complex background pattern. Therefore, to design a stable, reliable and efficient scene recognition algorithm is still very challenging work. In this paper, we propose an algorithm which combines Visual Simultaneous Localization and Mapping (SLAM) and Deep Convolutional Neural Networks (DCNNS) to boost the performance of AR registration. Semantic segmentation is a dense prediction task which aims to predict categories for each pixel in an image when applying to AR registration, and it will be able to narrow the searching range of the feature point between the two frames thus enhancing the stability of the system. Comparative experiments in this paper show that the semantic scene information will bring a revolutionary breakthrough to the AR interaction.