Three-dimensional visualization of X-ray micro-CT with large-scale datasets: Efficiency and accuracy for real-time interaction

This study evaluated the transformative importance of dynamic SEM in offering a more thorough understanding of real-time consumer behaviours and thus transcending the limitations of traditional SEM approaches that typically rely on static data. The study analysed the recent advancements in the dynamic SEM and its capability to strengthen marketing strategies by accurately capturing evolving consumer interactions. The study evaluated the published peer-reviewed literature ranging from 2010 to 2024 to assess the advancement, comparisons, applications, accuracy and methodological complexities of both dynamic and traditional SEM approaches in the domain of consumer behaviours and interactions marketing analytics. The inclusion criteria were studies focusing on consumer behaviour, research articles published within 14 years, studies employing dynamic SEM methods and datasets that include time-series data. The findings for objective one show that dynamic SEM analyses complex, temporal and real-time data because it has been integrated with advanced modern methods and approaches such as Ecological Momentary Assessment and Experience Sampling Method, Bayesian methods for estimation, machine learning algorithms and cloud computing platforms. The findings for objective two indicate that dynamic SEM is practically and accurately capable of analysing temporal and real-time high-frequency, complex, and large-scale datasets from digital platforms like social media and e-commerce. The results obtained from the comparative analysis for objective three show that dynamic SEM provides significant improvements by offering a more accurate reflection of evolving consumer interactions and preferences than traditional SEM. Dynamic SEM integrates temporal elements and therefore allows for adeptly modelling consumer choices, moods, attitudes, and emotional states over time. Performance metrics such as MAE, RMS, and CFI confirm that dynamic SEM enhances fit and predictive precision. The findings show that dynamic SEM substantially and significantly outperforms traditional SEM since it has been integrated with advanced methods that enhance the understanding of real-time consumer behaviour and interactions by effectively capturing temporal variations in consumer behaviour and interactions. Thus, organisations should adopt and implement the dynamic SEM to optimise and improve their marketing strategies. The study contributes to knowledge that the dynamic SEM is superior in capturing real-time consumer behaviours, which results in enhancing marketing analytics and strategies.

This paper explores how recent convolutional neural network (CNN)-based techniques can be used to interpolate images inside scientific image databases. These databases are frequently used for the interactive visualization of large-scale simulations, where images correspond to samples of the parameter space (e.g., timesteps, isovalues, thresholds, etc.) and the visualization space (e.g., camera locations, clipping planes, etc.). These databases can be browsed post hoc along the sampling axis to emulate real-time interaction with large-scale datasets. However, the resulting databases are limited to their contained images, i.e., the sampling points. In this paper, we explore how efficiently and accurately CNN-based techniques can derive new images by interpolating database elements. We demonstrate on several real-world examples that the size of databases can be further reduced by dropping samples that can be interpolated post hoc with an acceptable error, which we measure qualitatively and quantitatively.

The V3D system provides three-dimensional (3D) visualization of gigabyte-sized microscopy image stacks in real time on current laptops and desktops. Combined with highly ergonomic features for selecting an X, Y, Z location of an image directly in 3D space and for visualizing overlays of a variety of surface objects, V3D streamlines the on-line analysis, measurement, and proofreading of complicated image patterns. V3D is cross-platform and can be enhanced by plug-ins. We built V3D-Neuron on top of V3D to reconstruct complex 3D neuronal structures from large brain images. V3D-Neuron enables us to precisely digitize the morphology of a single neuron in a fruit fly brain in minutes, with about 17-fold improvement in reliability and 10-fold savings in time compared to other neuron reconstruction tools. Using V3D-Neuron, we demonstrated the feasibility of building a 3D digital atlas of neurite tracts in the fruit fly brain.

An extensible approach for real-time bidding with model-free reinforcement learning

Mobile Augmented Reality merges the virtual objects with real world on mobile devices, while video retrieval brings out the similar looking videos from the large-scale video dataset. Since mobile augmented reality application demands the real-time interaction and operation, we need to process and interact in real-time. Furthermore, augmented reality based virtual objects can be poorly textured. In order to resolve the above mentioned issues, in this research, we propose a novel, fast and robust approach for retrieving videos on the mobile augmented reality environment using an image and video queries. In the beginning, Top-K key-frames are extracted from the videos which significantly increases the efficiency. Secondly, we introduce a novel frame based feature extraction method, namely Pyramid Ternary Histogram of Oriented Gradient (PTHOG) to extract the shape feature from the virtual objects in an effective and efficient manner. Thirdly, we utilize the Double-Bit Quantization (DBQ) based hashing to accomplish the nearest neighbor search efficiently, which produce the candidate list of videos. Lastly, the similarity measure is performed to re-rank the videos which are obtained from the candidate list. An extensive experimental analysis is performed in order to verify our claims.

With the in-depth development of the application of virtual reality technology, people have higher and higher requirements for the complexity and realism of virtual scene, which far exceeds the real-time processing ability of computer graphics hardware. Back to this, there is an urgent need to solve the contradiction between the complexity of the scene and the real-time interaction. In this paper, the real-time visualization of real-time distributed virtual reality is studied in many aspects. Starting with the analysis of the characteristics of real-time visualization technology of distributed virtual reality, this paper studies the system structure of real-time distributed virtual reality, puts forward the idea of realizing real-time distributed virtual reality visualization, introduces grid scientific computing model, and constructs a three-dimensional vertical visualization platform of distributed virtual reality. The visual analysis of virtual reality is carried out for the corresponding slope and corresponding geological conditions; the comprehensive analysis of spatial distribution is carried out by using three-dimensional visualization, the relevant contents of real-time display, and the generation and real-time display of realistic graphics. In order to increase the realism of 3D image, visibility judgment, and blanking technology, the level of detail model and texture mapping is used. The simulation results show that the grid scientific computing model is effective and can support the three-dimensional visual analysis of distributed virtual reality.

Traditional medical image visualization software is difficult to run in cross-platform without any coordination mechanism, while the capacity of storage and the volume rendering are limited. We propose a cloud-based medical image visualization platform. To use the large storage and the powerful graphics computing capabilities in the cloud, it can codec medical image and implement segmentation and visualize volume. Then we propose an adaptive medical meta-data and image data transmission protocol, which also supports collaborative visualization. Through real-time interaction between cloud and browsers and peer transmission between browsers and browsers, we achieve a variety of two-dimensional and three-dimensional visualization of images and collaborative interaction features. Experimental results show that our platform not only support browsing large amounts of two-dimensional data and pseudo-color mapping in real time, but also support advanced segmentation and three-dimensional visualization with coordination mechanism, and has better real-time interactivity.

X-ray micro-CT has become a very powerful and common tool for non-destructive three-dimensional (3D) visualization and analysis of objects. Many systems are commercially available, but they are typically limited in terms of operational freedom both from a mechanical point of view as well as for acquisition routines. HECTOR is the latest system developed by the Ghent University Centre for X-ray Tomography (http://www.ugct.ugent.be) in collaboration with X-Ray Engineering (XRE bvba, Ghent, Belgium). It consists of a mechanical setup with nine motorized axes and a modular acquisition software package and combines a microfocus directional target X-ray source up to 240 kV with a large flat-panel detector. Provisions are made to install a line-detector for a maximal operational range. The system can accommodate samples up to 80 kg, 1 m long and 80 cm in diameter while it is also suited for high resolution (down to 4 μm) tomography. The bi-directional detector tiling is suited for large samples while the variable source-detector distance optimizes the signal to noise ratio (SNR) for every type of sample, even with peripheral equipment such as compression stages or climate chambers. The large vertical travel of 1 m can be used for helical scanning and a vertical detector rotation axis allows laminography experiments.The setup is installed in a large concrete bunker to allow accommodation of peripheral equipment such as pumps, chillers, etc., which can be integrated in the modular acquisition software to obtain a maximal correlation between the environmental control and the CT data taken. The acquisition software does not only allow good coupling with the peripheral equipment but its scripting feature is also particularly interesting for testing new and exotic acquisition routines.

Although computer capabilities have been improved significantly, a large-scale virtual reality (VR) system demands much more in terms of memory and computation than the current computer systems can offer. This paper discusses two important issues related to VR performance and applications in building navigation. These are dynamic loading of models based on cell segmentation for the optimal VR operation, and the route optimization based on path planning for easy navigation. The VR model of engineering and information technology complex (EITC) building at the University of Manitoba is built as an example to show the feasibility of the proposed methods. The reality, enhanced by three-dimensional (3D) real-time interactivity and visualization, leads navigators into a state of the virtual building immersion.

In order to reduce the risk of themoseeds hyperthemia clinical operation, an intuitive 3-D virtual patient needs to be reconstructed by the preoperative treatment planning system. The reconstruct results can supply the geometric physiological information of lesions and their surrounding tissues to form an optimal operation planning. In this study, the construction of 3D image intuitive environment of diagnosis and treatment in the system was investigated. On the basis of 2-D DICOM medical images, the 3-D data field was established and visually rendered, and the methods of volume rendering and surface rendering were correspondingly investigated. Results show that these two algorithms can provide specific rendering models for different organs. Volume rendering method can show good effects for 3-D body visualization and real-time interaction, and surface rendering method can extract iso-surface models to support tissue segmentation display and build thermal dose field.

Neurons transmit active potentials through axons, which are essential for the brain to function. In this study, the axonal networks of the murine brain were visualized with X-ray tomographic microscopy, also known as X-ray microtomography or micro-CT. Murine brain samples were freeze-dried to reconstitute the intrinsic contrast of tissue constituents and subjected to X-ray visualization. A whole brain hemisphere visualized by absorption contrast illustrated three-dimensional structures including those of the striatum, corpus callosum, and anterior commissure. Axonal tracts observed in the striatum start from the basal surface of the cerebral cortex and end at various positions in the basal ganglia. The distribution of X-ray attenuation coefficients indicated that differences in water and phospholipid content between the myelin sheath and surrounding tissue constituents account for the observed contrast. A rod-shaped cutout of brain tissue was also analyzed with a phase retrieval method, wherein tissue microstructures could be resolved with up to 2.7 μm resolution. Structures of axonal networks of the striatum were reconstructed by tracing axonal tracts. Such an analysis should be able to delineate the functional relationships of the brain regions involved in the observed network.

Three-dimensional visualization is one of the most important researches in the field of medical images processing. Volume rendering is a powerful technique for visualizing meaningful information extracted from volumetric data in three-dimensional visualization. Although it does not need to generate intermediate geometric primitives and its rendering result is high quality, computational complexity of volume rendering is too high to achieve real-time interaction. Therefore, a fast GPU based high-quality three-dimensional visualization volume rendering method is proposed. A GPU based method is proposed to improve the run-time efficiency of volume rendering using volumetric texture. Because of limitation of video memory storage, large-scale volumetric data cannot be reconstructed or an expensive GPU is required. Therefore a slice texture based volume rendering method is proposed, which merely cost 4MB video memory by making use of a real-time dynamic switch mechanism. On the other hand, in order to eliminate annular distortion of volume rendering, a GPU based pre-integration classification method is presented. Although our methods are all implemented at general PCs, they can not only guarantee image quality but also improve rendering speed.

Wetland plants, such as rice, develop aerenchyma, which is formed due to cell collapse, in their roots as pathways for diffusion of oxygen. Understanding of aerenchyma development in their plants is essential to elucidate the resistance mechanism of these plants to waterlogging stress. Synchrotron X-ray micro-computed tomography (micro-CT) is powerful for in-situ three-dimensional visualization of aerenchyma [1]. In this study, we have employed refraction contrast X-ray micro-CT was performed at a bending magnet beamline BL20B2 of SPring-8 to observe when and where aerenchyma starts to form and how they expand. Imbibed rice (Oryza sativa L. ssp. japonica cv. Nipponbare) caryopses were placed and roots were grown in darkness for 2 to 4 days in plastic tubes filled with agar containing Hoagland medium. The effective pixel size of the detector was 2.33 or 4.86 μm/pixel. Dark areas in a tomographic slice of a root were confirmed to correspond to air spaces formed due to cell collapse. Frequency distribution of location of collapsed cells showed it was the highest in the 5th and the 6th tiers from the outside of the endodermis. Air spaces formed due to collapse of a single cell were frequently observed, indicating that aerenchyma initiates at independent locations and then expand. Frequency of location of an adjacent collapsed cell to an existing collapsed cell tended to be higher in the radial and longitudinal direction. Volume of air space including aerenchyma, which were estimated from isosurface models created using tomograms, was confirmed to increase over a period of time. [1] I Karahara et al, Ann Bot 110 (2012) 503-509. C2-O-04 doi:10.1093/jmicro/dfv186

The use of geological models for mine production scheduling, planning, and design is a common aspect of current digital mine construction. Establishing a mapping relationship from digital geological resources to mining process simulation and then to risk early warning, enabling real-time interaction between digital models and physical mines, is an essential component of mining digital twins and an important direction for future development. This study is based on a non-ferrous metal mine and involves the development of data interaction functionality between 3Dmine (enterprise edition) and 3DEC7.0 software. This enables data mapping between geological models and numerical models, as well as real-time 3D visualization of risk points in the geological model. The main research findings are as follows: (1) Based on UAV photogrammetry and geological exploration data, a refined 3D geological model incorporating the surface, subsidence zones, goaf groups, and roadway systems was constructed using 3Dmine. The mine numerical model was then generated through 3Dmine-3DEC coupling technology. (2) A 3DEC-3Dmine data interaction interface based on Python was developed. Intelligent extraction and format conversion of mechanical parameters, such as stress and displacement, were achieved through secondary development, and a multi-software collaboration platform was built using an SQL database. A three-dimensional visual characterization script for risk points was developed. (3) Based on the strength–stress ratio and the nearest distance attribute assignment method, the three-dimensional visualization of blocks with different risk levels in 3Dmine is realized. (4) When the adjacent mine rooms are excavated in turn, the range of grade II risk area will be obviously expanded and a more serious grade III risk area will appear. The research findings offer a direction for the future development of mining digital twin technology, as well as technical support and theoretical guidance for analyzing and predicting safety risks during the mining process.

Customizable 6 degrees of freedom grasping dataset and an interactive training method for graph convolutional network