Related Experiment Video
Updated: May 24, 2025
12:49
A Method for 3D Reconstruction and Virtual Reality Analysis of Glial and Neuronal Cells
Published on: September 28, 2019
12.7K
Seg & Ref: a newly developed toolset for artificial intelligence-powered segmentation and interactive refinement for
Satoru Muro1, Takuya Ibara2, Akimoto Nimura2
1Department of Clinical Anatomy, Graduate School of Medical and Dental Sciences, Institute of Science Tokyo, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, Japan.
Microscopy (Oxford, England)
|March 4, 2025
Summary
We developed "Seg & Ref," an AI-powered toolset for automated 3D reconstruction. This user-friendly workflow simplifies segmentation and mask editing, enhancing morphological research without programming knowledge.
Area of Science:
- Developmental Biology
- Computational Biology
- Morphology
Background:
- Traditional 3D reconstruction is labor-intensive due to manual segmentation.
- Existing automated segmentation tools lack user-friendly interfaces for morphologists.
Purpose of the Study:
- To present an AI-powered workflow for 3D reconstruction using an intuitive toolset.
- To address the limitations of manual segmentation and complex existing tools.
Main Methods:
- Developed "Seg & Ref," an interactive toolset for AI-powered segmentation and mask editing.
- Utilized "SAM2 GUI for Img Seq" with Segment Anything Model 2 for automated segmentation via a web-based GUI.
- Employed "Segment Editor PP" (PowerPoint-based) for interactive mask refinement.
- Reconstructed 3D models using 3D Slicer.
Main Results:
- Successfully reconstructed key structures of a Carnegie Stage 15 human embryo, including the spinal cord, heart, and liver.
- Achieved accurate spatial relationships and morphologies in the 3D models.
- Demonstrated a labor-saving approach for 3D reconstruction workflows.
Conclusions:
- The "Seg & Ref" toolset offers a versatile and accessible solution for 3D reconstruction.
- This workflow enhances morphological research by providing precise segmentation through user-friendly interfaces.
- Eliminates the need for programming expertise, making advanced 3D reconstruction more accessible.
