SuRVoS: Super-Region Volume Segmentation workbench.

Area of Science:

• Microscopy and Imaging
• Computational Biology
• Biomedical Image Analysis

Background:

• Accurate segmentation of biological volumes is essential for data analysis but often hindered by a lack of user-friendly tools.
• Current automatic segmentation methods require extensive manual training data, which is scarce due to sample complexity and labeling time.

Purpose of the Study:

• To develop a novel workbench, SuRVoS (Super-Region Volume Segmentation), that integrates automatic segmentation with user expertise.
• To provide a more efficient and accessible tool for segmenting challenging biological volumes.

Main Methods:

• SuRVoS partitions volumes into hierarchical Super-Regions.
• It utilizes user-provided training annotations to guide and extend segmentation.
• The software leverages Super-Regions for faster segmentation compared to voxel-based methods.

Main Results:

• SuRVoS effectively combines automated segmentation with expert user input.
• The Super-Region approach facilitates quicker and easier segmentation, particularly on noisy, low-dose biological datasets.
• This method addresses the scarcity of manual training data for complex biological samples.

Conclusions:

• SuRVoS offers a practical solution for segmenting biological volumes, overcoming limitations of fully automatic or manual approaches.
• The workbench enhances the utilization of biological volume data by improving segmentation efficiency and accessibility.
• It is particularly beneficial for analyzing complex and noisy biological imaging data.