Instant3D: A User-Friendly GUI Integrating TotalSegmentator for Immediate Medical Image Segmentation and 3D Reconstruction

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Background: Automatic segmentation is indispensable in medical imaging, yet advanced tools often remain confined to experts due to command-line complexity. TotalSegmentator delivers accurate multi-organ segmentation, but lacks accessibility for clinicians and educators. To overcome this barrier, we developed Instant3D, an open-source graphical user interface (GUI) that makes high-quality 3D reconstruction straightforward and intuitive. Objective: The purpose of this study was to introduce the Instant3D GUI and validate its performance. Methods: We developed Instant3D in Python using PyQt6. It accepts DICOM, NIfTI, or NRRD input. Users select regions of interest through a suggestion-enabled interface, and the tool automatically runs TotalSegmentator. Outputs include STL meshes for 3D visualization, CSV files with volumetric data, and per-slice SVG masks. Crucially, these SVG files are interoperable with SegRef3D, enabling interactive correction and refinement of automated results—combining the strengths of artificial intelligence segmentation and user-driven adjustment. Batch processing is supported for large datasets. We validated Instant3D by testing it on representative computed tomography (CT) and magnetic resonance imaging (MRI) datasets, including publicly available example data from the 3D Slicer Sample Data module. Results: Instant3D reliably produced 3D models from CT and MRI scans. STL meshes preserved anatomical fidelity, SVG masks facilitated slice-level review and editing in SegRef3D, and CSV outputs provided quantitative volume data. The GUI eliminated the need for command-line knowledge, lowering the entry barrier for diverse users across research and education. Conclusions: We developed Instant3D, an open-source and user-friendly platform that democratizes advanced segmentation by combining automation with practical usability. It provides a seamless GUI, bridging advanced automatic segmentation with practical applications in research and development. Clinical Impact: Instant3D 3D reconstructions extend beyond visualization, offering immediate value for radiomics-driven quantitative research, virtual reality-based surgical simulation, 3D printing for surgical planning and patient education, and glasses-free 3D display teaching tools.

Article activity feed