Bo-Net: Deep learning-based model for automatic bone stromal cell segmentation of fluorescence microscopy images
PLOS One · 04.08.2026 · 21 (8) · e0353796
DOI: 10.1371/journal.pone.0353796 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 04.08.2026. Verfasst von Giorgio Allegri und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Electronics, Information and Bioengineering, Polytechnic University of Milan, Milan, Italy. Hauptthemen laut MEDLINE: Deep Learning, Image Processing, Computer-Assisted, Bone and Bones. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, Cancer Prevention and Research Institute of Texas, NIH.
Abstract (Originalsprache)
Background and objectives Bone is a multicellular organ that is the site of complex pathophysiological events (such as cancer). 3D confocal and multiphoton microscopy, followed by manual analysis and quantification, allow the identification of molecular, cellular, and tissue mechanisms in their original context, enabling the investigation of spatial biology at subcellular level. The automation of these analyses is becoming increasingly important due to their time-consuming nature, lack of standardization, and inter-subject variability.
Methods
Deep learning applied to image analysis could overcome current limitations and optimize preclinical research. This includes automatic semantic segmentation of bone cells (osteoblasts, osteoclasts, and blood vessels) and mineral component, followed by parameter extraction and quantification. Accordingly, fluorescence microscopy images were generated, pre-processed, and fed (total 21,395 images: 17,104 for training, 4291 for validation) into a neural network-based architecture named Bo-Net.
Results
The fully automatic Bo-Net achieved strong performance across the tested metrics and segmentation accuracy similar to experienced biologists (R = 0.81–0.99 for parameters tested) and improved analysis time from days to seconds in a variety of experimental conditions, confirming its relevance in different biologically relevant contexts.
Conclusions
As a result, a fully automated, yet reliable, tool to optimize the analysis of multiparametric fluorescence microscopy images available to the bone research community was generated.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Electronics, Information and Bioengineering, Polytechnic University of Milan, Milan, Italy.
- Center for Precision Surgery, Houston Methodist Research Institute, Houston, Texas, United States of America.
- David H. Koch Center for Applied Research of Genitourinary Cancers and Genitourinary Medical Oncology Department, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America.
- Rice University, Dept. of Bioengineering, Houston, Texas, United States of America.
- Department of Surgery, Houston Methodist Hospital, Houston, Texas, United States of America.
- LaSIE, UMR CNRS, La Rochelle Université, La Rochelle, France.
ORCID-Profile: Giulia Alessandrelli, Antonios Mikos, Stefano Casarin, Eleonora Dondossola
Förderung
- NCI NIH HHS
- Cancer Prevention and Research Institute of Texas
- NIH
- David H. Koch Center for Applied Research of Genitourinary Cancers
- Carolyn Bookout Presidential Distinguished Chair fund
Zitieren
APA
Allegri, G., Pavirani, L., Barrios, S., Sorice, S. R. C., Alessandrelli, G., Marsilio, L., Mikos, A., Cerveri, P., Casarin, S., & Dondossola, E. (2026). Bo-Net: Deep learning-based model for automatic bone stromal cell segmentation of fluorescence microscopy images. PLOS One, 21(8), e0353796. https://doi.org/10.1371/journal.pone.0353796
Vancouver
Allegri G, Pavirani L, Barrios S, Sorice SRC, Alessandrelli G, Marsilio L, et al. Bo-Net: Deep learning-based model for automatic bone stromal cell segmentation of fluorescence microscopy images. PLOS One. 2026;21(8):e0353796. doi:10.1371/journal.pone.0353796
BibTeX
@article{Allegri2026,
author = {Allegri, Giorgio and Pavirani, Luca and Barrios, Sergio and Sorice, Sofia Rosy Caterina and Alessandrelli, Giulia and Marsilio, Luca and Mikos, Antonios and Cerveri, Pietro and Casarin, Stefano and Dondossola, Eleonora},
title = {Bo-Net: Deep learning-based model for automatic bone stromal cell segmentation of fluorescence microscopy images},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {8},
pages = {e0353796},
doi = {10.1371/journal.pone.0353796}
}
EUVIMED