Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2511.22236 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911291532115968 |
|---|---|
| author | Püttmann, Simon Contreras, Jonathan Jair Sànchez Kowitz, Lennart Lampen, Peter Gupta, Saumya Panzeri, Davide Hagemann, Nina Xiong, Qiaojie Hermann, Dirk M. Chen, Cao Chen, Jianxu |
| author_facet | Püttmann, Simon Contreras, Jonathan Jair Sànchez Kowitz, Lennart Lampen, Peter Gupta, Saumya Panzeri, Davide Hagemann, Nina Xiong, Qiaojie Hermann, Dirk M. Chen, Cao Chen, Jianxu |
| contents | Accurate 3D microscopy image segmentation is critical for quantitative bioimage analysis but even state-of-the-art foundation models yield error-prone results. Therefore, manual curation is still widely used for either preparing high-quality training data or fixing errors before analysis. We present VessQC, an open-source tool for uncertainty-guided curation of large 3D microscopy segmentations. By integrating uncertainty maps, VessQC directs user attention to regions most likely containing biologically meaningful errors. In a preliminary user study uncertainty-guided correction significantly improved error detection recall from 67% to 94.0% (p=0.007) without a significant increase in total curation time. VessQC thus enables efficient, human-in-the-loop refinement of volumetric segmentations and bridges a key gap in real-world applications between uncertainty estimation and practical human-computer interaction. The software is freely available at github.com/MMV-Lab/VessQC. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_22236 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Bridging 3D Deep Learning and Curation for Analysis and High-Quality Segmentation in Practice Püttmann, Simon Contreras, Jonathan Jair Sànchez Kowitz, Lennart Lampen, Peter Gupta, Saumya Panzeri, Davide Hagemann, Nina Xiong, Qiaojie Hermann, Dirk M. Chen, Cao Chen, Jianxu Computer Vision and Pattern Recognition Accurate 3D microscopy image segmentation is critical for quantitative bioimage analysis but even state-of-the-art foundation models yield error-prone results. Therefore, manual curation is still widely used for either preparing high-quality training data or fixing errors before analysis. We present VessQC, an open-source tool for uncertainty-guided curation of large 3D microscopy segmentations. By integrating uncertainty maps, VessQC directs user attention to regions most likely containing biologically meaningful errors. In a preliminary user study uncertainty-guided correction significantly improved error detection recall from 67% to 94.0% (p=0.007) without a significant increase in total curation time. VessQC thus enables efficient, human-in-the-loop refinement of volumetric segmentations and bridges a key gap in real-world applications between uncertainty estimation and practical human-computer interaction. The software is freely available at github.com/MMV-Lab/VessQC. |
| title | Bridging 3D Deep Learning and Curation for Analysis and High-Quality Segmentation in Practice |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.22236 |