Ensembling Unets for Rare Chromosomal Aberration Detection in Metaphase Images, Uncertainty Quantification, and Ionizing Radiation Dose Estimation
Cytometry Part A · 2026-07 · 109 (7) · 485-499
DOI: 10.1002/cyto.a.70042 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cytometry Part A, online erschienen am 16.06.2026. Verfasst von Antonin Deschemps und 12 weiteren Autor:innen. Beteiligte Einrichtung: SAIRPICO Project-Team, Inria Center at University of Rennes, Institut Curie, Cellular and Chemical Biology Unit (U1339 INSERM, UMR3666 CNRS), PSL Res…. Hauptthemen laut MEDLINE: Metaphase, Chromosome Aberrations, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 48 Literaturangaben. Gefördert von: Agence de l'innovation de Défense, Agence Nationale de la Recherche, Région Bretagne.
Abstract (Originalsprache)
In biological dosimetry a radiation dose is estimated using the average number of chromosomal aberrations per peripheral blood lymphocytes. This analysis is still manually performed on 2D metaphase images depicting the 23 pairs of chromosomes because the false discovery rate of current automated detection systems is too high and variable because of sensitivity to small variations in image quality (chromosome spread, illumination variations …). Therefore, the current systems are only used to assist human experts. Designing more performant automatic and reliable chromosomal aberration detection systems has become of paramount importance to improve diagnosis speed and reduce human expertise time. Here, we propose a novel deep‐learning method for automatic rare chromosomal aberration detection and uncertainty quantification. We formulate the problem as a unique regression problem requiring the minimization of a sparsity‐promoting loss to reduce the false alarm rate. Furthermore, we select checkpoints at the end of each epoch during training to form a model ensemble. The resulting artificial experts are further analyzed to derive a consensus voting, similar to an agreement of human annotator rating, to provide trustworthy aberration detections and confidence intervals. A radiation dose curve is finally derived from deep learning‐assisted counting of dicentrics and fragments in metaphase images, in high agreement with the reference hand‐crafted curve in biological dosimetry with a promising dose estimation validation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- SAIRPICO Project-Team, Inria Center at University of Rennes, Institut Curie, Cellular and Chemical Biology Unit (U1339 INSERM, UMR3666 CNRS), PSL Research University, Rennes, France.
- SERAMED/LRAcc, Radiobiology and Regenerative Medicine Research Service, Radiobiology of Accidental Exposure Laboratory, ASNR-French Authority for Nuclear Safety and Radiation Protection, Fontenay-aux-Roses, France.
- Department of Radiation Biology, IRBA-Armed Forces Biomedical Research Institute, Brétigny-sur-Orge, France.
ORCID-Profile: Charles Kervrann
Förderung
- Agence de l'innovation de Défense
- Agence Nationale de la Recherche
- Région Bretagne
Zitieren
APA
Deschemps, A., Grégoire, E., Martinez, J. S., Sanou, E. D., Vaurijoux, A., Fernandez, P., Dugue, D., Bobyk, L., Valente, M., Gruel, G., Moebel, E., Benadjaoud, M., & Kervrann, C. (2026). Ensembling Unets for Rare Chromosomal Aberration Detection in Metaphase Images, Uncertainty Quantification, and Ionizing Radiation Dose Estimation. Cytometry Part A, 109(7), 485-499. https://doi.org/10.1002/cyto.a.70042
Vancouver
Deschemps A, Grégoire E, Martinez JS, Sanou ED, Vaurijoux A, Fernandez P, et al. Ensembling Unets for Rare Chromosomal Aberration Detection in Metaphase Images, Uncertainty Quantification, and Ionizing Radiation Dose Estimation. Cytometry Part A. 2026;109(7):485-499. doi:10.1002/cyto.a.70042
BibTeX
@article{Deschemps2026,
author = {Deschemps, Antonin and Grégoire, Eric and Martinez, Juan S and Sanou, Edmond D and Vaurijoux, Aurélie and Fernandez, Pascale and Dugue, Delphine and Bobyk, Laure and Valente, Marco and Gruel, Gaëtan and Moebel, Emmanuel and Benadjaoud, Mohamedamine and Kervrann, Charles},
title = {Ensembling Unets for Rare Chromosomal Aberration Detection in Metaphase Images, Uncertainty Quantification, and Ionizing Radiation Dose Estimation},
journal = {Cytometry Part A},
year = {2026},
volume = {109},
number = {7},
pages = {485--499},
doi = {10.1002/cyto.a.70042}
}
EUVIMED