3D Reference Model Research Articles

Cross-Fertilization between Computational Morphogenesis and Digital Pathology

Introduction/ Background We believe that Computational Morphogenesis (CM) and Digital Pathology (DP) can be combined to provide valuable tools to study cancer, and improve patient diagnosis and prognosis. CM could play a research-oriented role in the quest to understand the biological and biophysical processes that underlie the different paths of cancer development. DP can play a key role for this task. In return, domain-specific validated CM Model (CMM) such as a stage I colon cancer model could assist routine DP. Our aim is to give an outline of what DP can bring to CM and how DP can be enhanced by incorporating CMMs. In order to have a CM platform that can be tailored to produce a specific CMM, the platform must allow specifying (i) cells lineages and their properties, (ii) their interactions via autocrine, paracrine and endocrine signals and (iii) their interactions with the extracellular matrix. We now describe how data from biology and DP can provide the necessary inputs to switch from a generic CM to specific CMM. What DP can bring to CM In cancer studies, cell lineages and their spatial organization can be obtained via tumor-associated (TA) biomarkers such as PD-L1 - PD1 [1] and CD86 - CTLA-4 [2]. Other biomarkers of particular interest include those that characterize TA and non-TA mesenchymal stem cells [3, 4], fibroblasts [5], macrophages [6, 7], and T cells [8]. Using 3D z-staked representations, the instantaneous snapshot of the spatial organization of specific cell lineages can instruct the modeling task, and different dynamic models can be built since TA and non-TA cells of a same general class (e.g. macrophages) can have distinct phenotypes such as promoting or inhibiting tumor progression [6, 7]. Once the 3D spatial organization of the different cell-types is encoded into a CMM, reverse engineering and simulation can help determine putative initial tissue organization that may have led to the current tumor. Similarly, possible future states of the tumor can be predicted and then validated by comparing the simulated 2D/3D features by those obtained from real data. Hence, DP data could constitute one of the cornerstones of the iterative process of CMM specification, calibration, selection and validation. Enhancing DP systems by integrating CMMs Validated CMM could enhance current DP systems for patient diagnosis and prognosis. Machine learning (ML) technics have been applied to analyze 2D WSI [8, 9] and it is very likely that extending such methods to 3D tumor representations will improve the classification task. A large number of CMM-derived 3D reference models (3DRM) of, say, stage 1 colon cancer, could be produced as a complement to 3D models obtained by z-staking. The output of ML technics applied to real and simulated data could then be benchmarked by pathologists in order to assess if, when and how 3DRM can be integrated into decision-support systems dedicated to DP. We have designed a prototype generic CM tool that has been tested to generate multi-scale models of vascularized kidney structures from virtual stem cells. It can be linked to a DP platform for further developments.

Examplar coherent 3D face reconstruction from forensic mugshot database

Reconstructing 3D face models from 2D face images is usually done by using a single reference 3D face model or some gender/ethnicity specific 3D face models. However, different persons, even those of the same gender or ethnicity, usually have significantly different faces in terms of their overall appearance, which forms the base of person recognition via faces. Consequently, existing 3D reference model based methods have limited capability of reconstructing precise 3D face models for a large variety of persons. In this paper, we propose to explore a reservoir of diverse reference models for 3D face reconstruction from forensic mugshot face images, where facial examplars coherent with the input determine the final shape estimation. Specifically, our 3D face reconstruction is formulated as an energy minimization problem with: 1) shading constraint from multiple input face images, 2) distortion and self-occlusion based color consistency between different views, and 3) depth uncertainty based smoothness constraint on adjacent pixels. The proposed energy is minimized in a coarse to fine way, where the shape refinement step is done by using a multi-label segmentation algorithm. Experimental results on challenging datasets demonstrate that the proposed algorithm is capable of recovering high quality 3D face models. We also show that our reconstructed models successfully boost face recognition accuracy.

A new reference model for 3D inversion of airborne magnetic data in hilly terrain — A case study from northern Sweden

The inherent nonuniqueness in modeling magnetic data can be partly reduced by adding prior information, either as mathematical constructs or simply as bounds on magnetization obtained from laboratory measurements. If a good prior model can be used as a reference model, then the quality of estimated models through an inverse approach can be greatly improved. But even though data on magnetic properties of rocks might exist, their distribution may often be quite irregular on local and regional scales, so that it is difficult to define representative classes of rock types suitable for constraining geophysical models of magnetization. We have developed a new way of constructing a reference model that varies only laterally and is confined to the part of the terrain that lies above the lowest topography in the area. To obtain this model, several estimated 2D magnetization distributions were constructed by data inversion as a function of the iteration number. Then, a suitable 2D model of the magnetization in the topography was chosen as a starting point for constructing a 3D reference model by modifying it with a vertical decay such that its average source depth was the same for all horizontal positions. The average source depth of the reference model was chosen to satisfy the average source depth obtained from analyzing the radial power spectrum of the area studied. Finally, the measured magnetic data were inverted in three dimensions using the given reference model. For a selected reference model, shallow structures indicated a better overall correlation with large remanent magnetizations measured on rock samples from the area. Throughout the entire model, the direction of magnetization was allowed to vary freely. We found that the Euclidean norm of the estimated model was reduced compared with the case where the magnetization direction was fixed.

Adjustment of Sonar and Laser Acquisition Data for Building the 3D Reference Model of a Canal Tunnel.

In this paper, we focus on the construction of a full 3D model of a canal tunnel by combining terrestrial laser (for its above-water part) and sonar (for its underwater part) scans collected from static acquisitions. The modeling of such a structure is challenging because the sonar device is used in a narrow environment that induces many artifacts. Moreover, the location and the orientation of the sonar device are unknown. In our approach, sonar data are first simultaneously denoised and meshed. Then, above- and under-water point clouds are co-registered to generate directly the full 3D model of the canal tunnel. Faced with the lack of overlap between both models, we introduce a robust algorithm that relies on geometrical entities and partially-immersed targets, which are visible in both the laser and sonar point clouds. A full 3D model, visually promising, of the entrance of a canal tunnel is obtained. The analysis of the method raises several improvement directions that will help with obtaining more accurate models, in a more automated way, in the limits of the involved technology.

Computational anatomy of the proximal humerus: An ex vivo high-resolution peripheral quantitative computed tomography study

SummaryBackground/ObjectiveSpatial knowledge of the anatomy of the proximal humerus is critical for effective treatment, particularly in patients affected by fragility fractures. High-resolution peripheral quantitative computed tomography (HR-pQCT) imaging with medical image processing techniques enable three dimensional (3D) analysis of volumetric bone mineral density (vBMD) of bones of different sizes and shapes.MethodsTo elucidate the bony anatomy and to create 3D reference data, we conducted a computerized HR-pQCT-based study in intact postmortem samples of the proximal humerus to highlight the anatomy with particular emphasis on the size, shape, and bone stock distribution pattern.Fifty-eight defrozen intact humerus samples from 28 female and 30 male donors, who were aged 61–98 years old (mean age ± standard deviation, 80.6 ± 9 years), were scanned in the proximal third using the extended standard HR-pQCT protocol. A 3D statistical bone and averaged bone density models with low, middle, and high total vBMDs were computed. We examined the 3D patterns of size and shape variations using principal component analysis, and the vBMD distributions and variabilities using volume-rendering and virtual bore probing.ResultsThe computer models revealed a highly variable bony anatomy in which size was the predominant variation in the first principal component (PC). In the second PC, we observed notable variabilities in the shape of the head and shaft inclination. A distinct 3D pattern of bone stock distribution was detected in which the lowest vBMD values were identified in the medullary cavity, middle values were identified in the central zone, and the highest values were identified in the cortex and humeral head—particularly in the subarticular zones. In the presence of bone loss, the vBMD values were ubiquitously decreased, but the pattern of 3D bone stock distribution was maintained.ConclusionThe new anatomical 3D data that we acquired will improve the understanding of the normal bony anatomy of the proximal humerus. The extended HR-pQCT protocol and computer models may be used for other skeletal sites and used as 3D reference models that can be applied to systematically improve implant design and anchorage.

Focus plus context visualization based on volume clipping for markerless on-patient medical data visualization

Focus plus context visualization can be used in augmented reality to improve the visual perception of the augmented scene. In the scope of in situ or on-patient medical data visualization, the focus plus context paradigm is used to improve depth perception for physicians showing the patient׳s anatomy as a focus region in the context of the patient׳s body. Volume clipping is one technique to realize focus plus context visualization. However, some of the existing methods for focus plus context visualization based on volume clipping do not run in full real time or are prone to artifacts. In this paper, we present an extension for two of these techniques to improve performance and image quality of the original approaches. We validate all the techniques in a markerless augmented reality environment. A 3D reference model is tracked by the application, and volumetric medical data are shown to the user at the position of the patient׳s anatomy. Our technique is able to handle multiple anatomic regions, although the main region of interest used in this paper is the face. Moreover, tracking accuracy is improved by the use of a hierarchical approach. From an evaluation of the proposed techniques, the results obtained highlight that all of them are free of artifacts, optimized for real-time performance, and improve the visual quality of the augmented scene.

BUILDING A 3D REFERENCE MODEL FOR CANAL TUNNEL SURVEYING USING SONAR AND LASER SCANNING

Abstract. Maintaining canal tunnels is not only a matter of cultural and historical preservation, but also a commercial necessity and a security issue. This contribution adresses the problem of building a full 3D reference model of a canal tunnel by merging SONAR (for underwater data recording) and LASER data (for the above-water parts). Although both scanning devices produce point clouds, their properties are rather different. In particular, SONAR data are very noisy and their processing raises several issues related to the device capacities, the acquisition setup and the tubular shape of the tunnel. The proposed methodology relies on a denoising step by meshing, followed by the registration of SONAR data with the geo-referenced LASER data. Since there is no overlap between point clouds, a 3-step procedure is proposed to robustly estimate the registration parameters. In this paper, we report a first experimental survey, which concerned the entrance of a canal tunnel. The obtained results are promising and the analysis of the method raises several improvement directions that will help obtaining more accurate models, in a more automated fashion, in the limits of the involved technology.

Reliability in wireless sensor networks: A survey and challenges ahead

Ensuring energy efficient and reliable transport of data in resource constrained Wireless Sensor Networks (WSNs) is one of the primary concerns to achieve a high degree of efficiency in monitoring and control systems. The two techniques typically used in WSNs to achieve reliability are either retransmission or redundancy.Most of the existing research focuses on traditional retransmission-based reliability, where reliable transmission of data packets is ensured in terms of recovering the lost packets by retransmitting them. This might result in additional transmission overhead that not only wastes sensors’ limited energy resources but also makes the network congested and in turn affects the reliable transmission of data. On the other hand, employing redundancy to achieve reliability in WSNs has received comparatively lesser emphasis by the research community [35] and this area warrants further investigation. In redundancy-based reliability mechanisms, a bit loss within a packet can be recovered by utilizing some form of coding schemes. This ability to correct the lost or corrupted bits within a packet would significantly reduce the transmission overhead caused by the retransmission of the entire packet.Both retransmission and redundancy can either be performed on a hop-by-hop or an end-to-end basis. Hop-by-hop method allows the intermediate nodes to perform retransmission or redundancy. On the other hand, in the end-to-end approach, retransmission or redundancy is performed only at the source and the destination nodes. However, a hybrid mechanism with an efficient combination of these retransmission and redundancy techniques in order to achieve reliability has so far been neglected by the existing research. Depending on the nature of the application, it is also important to define the amount of data required to ensure reliability. This introduces the concept of packet or event level reliability. Packet reliability requires all the packets from all the relevant sensor nodes to reach the sink, whereas event reliability ensures that the sink only gets enough information about a certain event happening. Thus retransmission or redundancy techniques using hop-by-hop or end-to-end mechanisms aim to achieve either packet or event level reliability.This paper presents a survey on reliability protocols in WSNs. We review several reliability schemes based on retransmission and redundancy techniques using different combinations of packet or event reliability in terms of recovering the lost data using hop-by-hop or end-to-end mechanisms. We further analyze these schemes by investigating the most suitable combination of these techniques, methods and required reliability level in order to provide energy efficient reliability mechanism for resource constrained WSNs. This paper also proposes a 3D reference model for classifying research in WSN reliability, which will be used to perform in-depth analysis of the unexplored areas.

Face Recognition Across Poses Using a Single 3D Reference Model

Face recognition across pose is generally tackled by either 2D based or 3D based approaches. The 2D-based often require a training set from which the cross-pose multi-view relationship can be learned and applied for recognition. The 3D based are mostly composed of 3D surface reconstruction of each gallery face, synthesis of 2D images of novel views using the reconstructed model, and match of the synthesized images to the probes. The depth information provides crucial information for arbitrary poses but more methods are yet to be developed. Extended from a latest face reconstruction method using a single 3D reference model and a frontal registered face, this study focuses on using the reconstructed 3D face for recognition. The recognition performance varies with poses, the closer to the front, the better. Several ways to improve the performance are attempted, including different numbers of fiducial points for alignment, multiple reference models considered in the reconstruction phase, and both frontal and profile poses available in the gallery. These attempts make this approach competitive to the state-of-the-art methods.

High-Quality On-Patient Medical Data Visualization in a Markerless Augmented Reality Environment

To provide on-patient medical data visualization, a medical augmented reality environment must support volume rendering, accurate tracking, real-time performance and high visual quality in the final rendering. Another interesting feature is markerless registration, to solve the intrusiveness introduced by the use of fiducial markers for tracking. In this paper we address the problem of on-patient medical data visualization in a real-time high-quality markerless augmented reality environment. The medical data consists of a volume reconstructed from 3D computed tomography image data. Markerless registration is done by generating a 3D reference model of the region of interest in the patient and tracking it from the depth stream of an RGB-D sensor. From the estimated camera pose, the volumetric medical data and the reference model are combined allowing a visualization of the patient as well as part of his anatomy. To improve the visual perception of the scene, focus+context visualization is used in the augmented reality scene to dynamically define which parts of the medical volume will be visualized in the context of the patient’s image. Moreover, context-preserving volume rendering is employed to dynamically control which parts of the volume will be rendered. The results obtained show that the markerless environment runs in real-time and the techniques applied greatly improve the visual quality of the final rendering.

Heat transfer of spent ion exchange resin in iron ore sintering process

Spent ion exchange resin (SIER) is a kind of solid waste which derived from water treatment process in iron and steel industry. Utilization of SIER as a replacement of fuel in sintering process is a promising method of SIER treatment. In order to investigate the feasibility of this method, heat transfer effects of SIER in iron ore sintering process were studied in this paper via numerical simulation. A 3D unsteady numerical reference model was developed on the basis of the porous media model and local non-equilibrium thermodynamics model and verified by the data form measurement. Heat transfer effects of different SIER mass fractions (0–8%) in sintering material during the sintering process were studied on the numerical model established in this paper. The results showed that with the increasing mass fraction of SIER, the maximum combustion zone thickness and flame front speed are both increased, the heating-up point, the moment when solid temperature reached the top and the maximum temperatures were all became earlier in different location of sintering bed. When SIER content is 8%, the maximum sintering temperature exceeds the maximum limit of best sintering temperature.

On the modeling of flexure hinge mechanisms with finite beam elements of variable cross section

In this paper, a modeling technique for flexure hinge mechanisms is studied. Beam elements of variable cross sections are deployed within a finite element procedure to model a circular flexure hinge. The resulting finite element model has very few degrees of freedom and is accurate in both static and dynamic analysis. Furthermore the modeling approach is applied to an amplifier mechanism. Comparing the results of the proposed model with a 3D finite element reference model, high accuracy for a broad spectrum of hinge parameters is reported while reducing the number of degrees of freedom immensely.

Performance Analysis of a Two-Hop MIMO Mobile-to-Mobile via Stratospheric-Relay Link Employing Hierarchical Modulation

Next generation wireless communication networks intend to take advantage of the integration of terrestrial and aerospace infrastructures. Besides, multiple-input multiple-output (MIMO) architecture is the key technology, which has brought the wireless gigabit vision closer to reality. In this direction, high-altitude platforms (HAPs) could act as relay stations in the stratosphere transferring information from an uplink to a downlink MIMO channel. This paper investigates the performance of a novel transmission scheme for the delivery of mobile-to-mobile (M-to-M) services via a stratospheric relay. It is assumed that the source, relay, and destination nodes are equipped with multiple antennas and that amplify-and-forward (AF) relaying is adopted. The performance is analyzed through rigorous simulations in terms of the bit-error rate (BER) by using a recently proposed 3D geometry-based reference model in spatially correlated flat-fading MIMO channels, employing a hierarchical broadcast technique and minimum mean square error (MMSE) receivers.

A new P-velocity model for the Tethyan margin from a scaled S-velocity model and the inversion of P- and PKP-delay times

We estimate a 3D P-velocity model for the Tethyan margin by inverting P- and PKP-delay times. The inversion is relative to a 3D reference model, which is a scaled S-velocity model for the same region. This S-velocity model was derived by jointly inverting regional S and Rayleigh waveform fits, teleseismic arrival times, Rayleigh-wave group velocities, and independent Moho constraints. Thus, our 3D reference model includes structures resolved over a larger depth range as well as more information on aseismic regions with few stations than is typically obtained from traditional teleseismic delay time inversions. We then inverted P- and PKP-delay times to obtain perturbations relative to the scaled 3D reference model. Comparing our P-velocity model (EAPV11) with P-velocity models derived from P data only, we find a model with more uniform and better depth resolution, including velocity anomalies for aseismic regions with few stations such as North Africa, southeastern Arabia, and the East European platform. Using EAPV11 to predict arrival times for relatively accurately located events that were not used in the inversion shows that our model produces significant variance reductions for these data as well. Therefore, our approach to build P-velocity models based on 3D reference S-velocity models may provide a practical way to better estimate P-velocity anomalies in the uppermost mantle and beneath aseismic regions with few stations.

Use of MCAM in creating 3D neutronics model for ITER building

The three dimensional (3D) neutronics reference model of International Thermonuclear Experimental Reactor (ITER) only defines the tokamak machine and extends to the bio-shield. In order to meet further 3D neutronics analysis needs, it is necessary to create a 3D reference model of the ITER building. Monte Carlo Automatic Modeling Program for Radiation Transport Simulation (MCAM) was developed as a computer aided design (CAD) based bi-directional interface program between general CAD systems and Monte Carlo radiation transport simulation codes. With the help of MCAM version 4.8, the 3D neutronics model of ITER building was created based on the engineering CAD model. The calculation of the neutron flux map in ITER building during operation showed the correctness and usability of the model. This model is the first detailed ITER building 3D neutronics model and it will be made available to all international organization collaborators as a reference model.

A Three-Dimensional Geometry-Based Statistical Model of2×2Dual-Polarized MIMO Mobile-to-Mobile Wideband Channels

A three-dimensional (3D) model for wide-band dual-polarized (DP) multiple-input-multiple-output (MIMO) mobile-to-mobile (M2M) channels is proposed. Using geometrical scattering based on concentric spheres at the transmitter (Tx) and at the receiver (Rx), a 3D parametric reference model for2×2M2M DP multipath fading channels is developed. The channel model assumes the use of colocated half-wavelength dipole antennas for vertical and horizontal polarizations at both transmit and receive stations. Model parameters include the velocities of the Tx and Rx nodes, the distance between the nodes, the 3D antenna pattern gains, the azimuth and elevation angles of arrival and departure, the geometrical distribution of the scatterers, the Rician K-factors defining the fading envelope distributions, the maximum Doppler frequency, the scattering loss factors, the cross-polar power discrimination ratio (XPD), and the copolarization power ratio (CPR). Using the proposed model, expressions for joint time-frequency correlation functions (TFCFs) are derived which are used to investigate system behavior over different wide sense stationary uncorrelated scattering (WSSUS) channel realizations. The numerical results illustrate the sensitivities of the TFCF to simultaneous time and frequency offsets for the2×2DP-MIMO architectures.

3D Face Reconstruction from a Single Image Using a Single Reference Face Shape

Human faces are remarkably similar in global properties, including size, aspect ratio, and location of main features, but can vary considerably in details across individuals, gender, race, or due to facial expression. We propose a novel method for 3D shape recovery of faces that exploits the similarity of faces. Our method obtains as input a single image and uses a mere single 3D reference model of a different person's face. Classical reconstruction methods from single images, i.e., shape-from-shading, require knowledge of the reflectance properties and lighting as well as depth values for boundary conditions. Recent methods circumvent these requirements by representing input faces as combinations (of hundreds) of stored 3D models. We propose instead to use the input image as a guide to "mold" a single reference model to reach a reconstruction of the sought 3D shape. Our method assumes Lambertian reflectance and uses harmonic representations of lighting. It has been tested on images taken under controlled viewing conditions as well as on uncontrolled images downloaded from the Internet, demonstrating its accuracy and robustness under a variety of imaging conditions and overcoming significant differences in shape between the input and reference individuals including differences in facial expressions, gender, and race.

More than a decade of developmental gene expression atlases: where are we now?

To unravel regulatory networks of genes functioning during embryonic development, information on in situ gene expression is required. Enormous amounts of such data are available in literature, where each paper reports on a limited number of genes and developmental stages. The best way to make these data accessible is via spatio-temporal gene expression atlases. Eleven atlases, describing developing vertebrates and covering at least 100 genes, were reviewed. This review focuses on: (i) the used anatomical framework, (ii) the handling of input data and (iii) the retrieval of information. Our aim is to provide insights into both the possibilities of the atlases, as well as to describe what more than a decade of developmental gene expression atlases can teach us about the requirements of the design of the ‘ideal atlas’. This review shows that most ingredients needed to develop the ideal atlas are already applied to some extent in at least one of the discussed atlases. A review of these atlases shows that the ideal atlas should be based on a spatial framework, i.e. a series of 3D reference models, which is anatomically annotated using an ontology with sufficient resolution, both for relations as well as for anatomical terms.

Scanner à rayons X et paléoanthropologie crânienne

Since its invention in 1972, computed tomography (C.T.) has significantly evolved. With the advent of multi-slice detectors (500 times more sensitive than conventional radiography) and high-powered computer programs, medical applications have also improved. CT is now contributing to paleoanthropological research. Its non-destructive nature is the biggest advantage for studying fossil skulls. The second advantage is the possibility of image analysis, storage, and transmission. Potential disadvantages include the possible loss of files and the need to keep up with rapid technological advances. Our experience since the late 1970s, and a recent PhD thesis, led us to describe routine applications of this method. The main contributions of CT to cranial paleoanthropology are five-fold: --Numerical anatomy with rapid acquisition and high spatial resolution (helicoidal and multidetector CT) offering digital storage and stereolithography (3D printing). --Numerical biometry (2D and 3D) can be used to create "normograms" such as the 3D craniofacial reference model used in maxillofacial surgery. --Numerical analysis offers thorough characterization of the specimen and its state of conservation and/or restoration. --From "surrealism" to virtual imaging, anatomical structures can be reconstructed, providing access to hidden or dangerous zones. --The time dimension (4D imaging) confers movement and the possibility for endoscopic simulation and internal navigation (see Iconography). New technical developments will focus on data processing and networking. It remains our duty to deal respectfully with human fossils.

Finite‐frequency seismic tomography

The use of 3‐D sensitivity kernels in tomographic inversions has received much attention in the seismological community. Conventional tomographic studies are based on ray theory, which assumes that the arrival time of a body wave depends only on the velocity along the ray path between the source and receiver. Observed seismic waves are finite frequency signals, and their travel times are sensitive to a 3‐D volume around the geometrical ray and subjected to wavefront healing, scattering, and other diffractive effects. As a result, seismic waves passing through velocity heterogeneities with dimensions smaller than the width of the Fresnel zone undergo significant wave‐front healing, which results in reduced travel‐time shifts compared to the values predicted by ray theory. 3‐D sensitivity kernels take into account the effects of wave diffraction and scattering. This presentation will provide a brief review of finite‐frequency tomography, which is in a stage of rapid development. With the advancement of computer technology, it has become increasingly practical to construct 3D sensitivity kernels based on 3D reference models using full waveform simulations. This holds the promise of a new level of the integration and joint interpretation of finite‐frequency waves propagating in complex media.