Ascribe New Dimensions to Scientific Data Visualization with VR
Daniela Ushizima, Guilherme Melo dos Santos, Zineb Sordo, Ronald Pandolfi, Jeffrey Donatelli · 2025
Read Ascribe New Dimensions to Scientific Data Visualization with VR by Daniela Ushizima, Guilherme Melo dos Santos, Zineb Sordo, Ronald Pandolfi, Jeffrey Donatelli (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-13448.txt.
Introduction
Daniela Ushizima and four co-authors present a virtual reality platform for exploring scientific images, in a 2025 paper. For fifty years the mouse has been the main tool for handling data, yet flat screens cramp images that are truly three-dimensional. Their platform, ASCRIBE-VR, puts researchers inside the data. It merges AI-made segmentations with immersive views of scans such as X-ray tomography and magnetic resonance images.
The tools run on common headsets and let users inspect structures at many scales, with the AI's labels close at hand. Feedback from the human explorer can loop back into the analysis. The authors aim the system at materials research, where samples hide complex inner geometry. They argue that joining machine output with human intuition inside virtual space can speed discovery, linking people with digital twins of their specimens.
Catalog record
| CREATOR | Daniela Ushizima, Guilherme Melo dos Santos, Zineb Sordo, Ronald Pandolfi, Jeffrey Donatelli |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-13448
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Ascribe New Dimensions to Scientific Data Visualization with VR Thanks: Citation : Ushizima et al. Ascribe new dimensions to scientific data visualization and interactive exploration with VR. DOI:000000/11111. Daniela Ushizima, Guilherme Melo dos Santos, Zineb Sordo, Ronald Pandolfi, Jeffrey Donatelli Abstract For over half a century, the computer mouse has been the primary tool for interacting with digital data, yet it remains a limiting factor in exploring complex, multi-scale scientific images. Traditional 2D visualization methods hinder intuitive analysis of inherently 3D structures. Virtual Reality (VR) offers a transformative alternative, providing immersive, interactive environments that enhance data comprehension. This article introduces ASCRIBE-VR, a VR platform of Autonomous Solutions for Computational Research with Immersive Browsing & Exploration, which integrates AI-driven algorithms with scientific images. ASCRIBE-VR enables multimodal analysis, structural assessments, and immersive visualization, supporting scientific visualization of advanced datasets such as X-ray CT, Magnetic Resonance, and synthetic 3D imaging. Our VR tools, compatible with Meta Quest, can consume the output of our AI-based segmentation and iterative feedback processes to enable seamless exploration of large-scale 3D images. By merging AI-generated results with VR visualization, ASCRIBE-VR enhances scientific discovery, bridging the gap between computational analysis and human intuition in materials research, connecting human-in-the-loop with digital twins. K eywords Virtual Reality, AI in Materials Science, Digital Twins, Scientific Visualization 1 Introduction For over half a century, the computer mouse has been the primary tool for interacting with digital data, including complex 3D models. Originally invented in the 1960s, it revolutionized human-computer interaction, yet it remains fundamentally a two-dimensional device: an imperfect bridge to the inherently three-dimensional world of materials science, biological research and more. Researchers working with intricate datasets, which encompass micro structures across multiple scales and conditions, often rely on traditional 2D screens and input devices that can limit intuitive exploration, understanding, and learning [ 5 ] . Virtual reality (VR) offers a long-overdue evolution in how we engage with data, providing immersive, interactive environments that feel more natural and dynamic. Using VR, scientists dependent on image records can manipulate, analyze, and visualize complex structures in ways that were previously impractical, fostering deeper insights and more effective collaboration. This paper explores the current landscape of VR applications in materials science, and biology, evaluating how these technologies enhance data interaction, visualization, and analysis (Sec. 2 ). Current scientific visualization methods frequently fail to provide adequate interactivity or depth, limiting scientists’ ability to intuitively comprehend complex data structures. To overcome these limitations, we developed an innovative VR platform called ASCRIBE-VR (Sec. 3 ), a platform for Autonomous Solutions for Computational Research with Immersive Browsing & Exploration in VR. ASCRIBE-VR is based on Unreal Engine, enhancing visual perception using the Meta Quest 3 and enabling effective human-in-the-loop decision-making for experimental validation. 2 Background: VR Applications for Scientific Data Several VR applications and platforms have emerged that cater to the needs of scientific data visualization and interaction, some of which have direct or potential applications in materials science. We list them in alphabetical order below, highlighting their main applications, the game engines under consideration, and the devices that they were tested on. Crystal Structure VRLE A Virtual Reality Learning Environment (VRLE) specifically designed for teaching crystal structures in Materials Science and Engineering has been developed using Unity [ 14 ] . This application addresses the limitations of traditional two-dimensional representations by providing immersive three-dimensional visualizations of atomic lattices. Users can explore various types of crystal structures from multiple viewpoints, including a floor view where they can walk inside the structures, a hover view for a top-down perspective, and a center view to observe the interior. The VRLE allows for manipulation of unit cells, identification of their types, and creation of custom crystal structures by positioning atoms [ 14 ] . Importantly, it includes features for analytical measurement, enabling users to measure the distance between atoms using VR controllers. This project demonstrates how VR can enhance the understanding of fundamental materials science concepts through interactive exploration. Immersive ParaView ParaView is a widely used open-source, multi-platform application for data analysis and visualization across various scientific domains [ 24 ] . Immersive ParaView enhances the base software with VR capabilities, allowing for more natural and intuitive exploration of 3D data using technologies such as CAVEs and head-mounted displays [ 20 ] . Tomviz is a specific version of ParaView tailored for the visualization and analysis of electron tomography data, a technique in materials science for reconstructing 3D volumes of materials at the nanoscale and mesoscale. Tomviz provides tools for generating shaded contours and volumetric projections, performing detailed data analytics through histograms, statistics, filters, and user-customized Python scripts, and combining multiple datasets for rich analysis [ 13 ] . Both ParaView and Tomviz are open-source and free to use, making them valuable resources for materials science researchers. ParaView supports a wide range of scientific data formats, including image formats, volume data formats, simulation software-specific formats, and also supports OpenVR and Oculus systems [ 20 ] . Nanome Nanome is a virtual reality software platform targeted primarily for drug design and molecular visualization, but its capabilities extend to materials science as well [ 10 ] . It allows users to visualize, build, and simulate chemical compounds, proteins, and nucleic acids in an immersive 3D environment [ 8 ] . Researchers can collaborate remotely in real-time, using VR headsets and hand controllers to manipulate molecules [ 9 ] . Nanome supports loading molecules from public databases such as RCSB, PubChem, and Drugbank, as well as from local computer files [ 8 ] . It enables users to study the geometries of molecules in 3D, take measurements of distances and angles, and even stand inside protein binding pockets to investigate their structure [ 7 ] . The software also supports features like electron density maps, which are essential for advanced crystallography research [ 8 ] . While initially focused on biomolecules, Nanome can be adapted for materials science by converting atoms to different types if needed, as demonstrated in research on amorphous carbon [ 21 ] . Nanome offers various licensing options, including academic and research licenses, making it accessible to a wide range of users [ 9 ] . Nanome has announced its availability on the Meta Quest Pro [ 11 ] . vLUME vLUME is a virtual reality software package specifically designed for visualizing and analyzing large three-dimensional single-molecule localization microscopy (3D-SMLM) datasets, but its core functionalities could be relevant to certain materials science applications. It allows researchers to ’walk’ inside individual cells and study everything from individual proteins to entire cellular structures. The software can load multiple datasets with millions of data points and uses in-built clustering algorithms to find patterns [ 6 ] . vLUME enables robust visualization, segmentation, annotation, and quantification of millions of fluorescence points from various 3D-SMLM techniques. Users can segment regions of interest using VR controllers, perform custom analyses using user-defined scripts (in C#), and create fly-through videos for presentations [ 25 ] . vLUME is free for academic use and is compatible with major commercial VR hardware like the Oculus Rift and HTC Vive [ 6 ] . While primarily used with biological datasets, authors argue about its ability to handle large point-cloud data and its