PAVSAT: an automated blood vessel analysis tool using deep learning-based segmentation and image processing
Ryozo Ishida, Naoi Hosoe, Anna Shimizu, Kazuhiro Takara, Yumiko Hayashi, Lamri Lynda, Hiroyasu Kidoya · 2026
Read PAVSAT: an automated blood vessel analysis tool using deep learning-based segmentation and image processing by Ryozo Ishida, Naoi Hosoe, Anna Shimizu, Kazuhiro Takara, Yumiko Hayashi, Lamri Lynda, Hiroyasu Kidoya (2026) 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/PMC13543378.txt.
Description
Ryozo Ishida and colleagues published a 2026 paper introducing PAVSAT, an automated tool for measuring blood vessel networks in microscope images. Imaging now captures tangled vascular structures faster than analysts can trace them, and older software struggles with curved vessels of changing width. The system pairs a YOLOv8 deep-learning segmenter, trained on stained vessel images, with skeleton extraction and a new branch-point finder. It takes repeated width measurements along each vessel, then converts the network into a graph for analysis.
A boundary-aware tiling scheme, which processes overlapping image tiles and merges the results, cut missed vessels sharply in test samples while adding few false alarms. Against expert hand measurements, the tool detected 91 to 98 percent of vessels and measured widths with 89 to 93 percent accuracy. Freely automating this work could speed studies of how vessels develop and how disease distorts them.
Catalog record
| CREATOR | Ryozo Ishida, Naoi Hosoe, Anna Shimizu, Kazuhiro Takara, Yumiko Hayashi, Lamri Lynda, Hiroyasu Kidoya |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60670 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |