A public generalizable AI tool for automated segmentation of coronal brain tissue slabs for 3D neuropathology
Jonathan Williams Ramirez, Dina Zemlyanker, Lucas Deden-Binder, Rogeny Herisse, Erendira Garcia Pallares, Karthik Gopinath, Harshvardhan Gazula, Christopher Mount · 2026
Read A public generalizable AI tool for automated segmentation of coronal brain tissue slabs for 3D neuropathology by Jonathan Williams Ramirez, Dina Zemlyanker, Lucas Deden-Binder, Rogeny Herisse, Erendira Garcia Pallares, Karthik Gopinath, Harshvardhan Gazula, Christopher Mount (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/plos-pone-0355740.txt.
Description
Jonathan Williams Ramirez and seven colleagues release a free artificial intelligence tool that automatically outlines brain tissue in photographs of sliced postmortem brains, in this 2026 paper. Brain banks routinely photograph coronal slabs, and new methods can rebuild three-dimensional maps of disease from those photos, but someone first has to trace the tissue in every image by hand. Their model, built on a standard segmentation network, was trained on more than fourteen hundred hand-traced photographs from two sites, covering fresh and fixed tissue and varied diagnoses.
On unseen photographs the tool matched human tracings closely, with overlap and boundary errors approaching the agreement between human raters themselves. The authors have made the model and its training data public. They argue removing the tracing bottleneck should let far more laboratories adopt three-dimensional neuropathology and mine their existing photo archives.
Catalog record
| CREATOR | Jonathan Williams Ramirez, Dina Zemlyanker, Lucas Deden-Binder, Rogeny Herisse, Erendira Garcia Pallares, Karthik Gopinath, Harshvardhan Gazula, Christopher Mount |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34842 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |