Dual-input deep learning system for microbial identification from blood agar plates
PLOS One · 27.07.2026 · 21 (7) · e0353761
DOI: 10.1371/journal.pone.0353761 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 27.07.2026. Verfasst von Takao Naito und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Clinical Microbiology Analysis Development Research, Juntendo University Graduate School of Medicine, Tokyo, Japan. Hauptthemen laut MEDLINE: Deep Learning, Image Processing, Computer-Assisted, Bacteria. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science, Shimadzu, Kurozumi Medical Foundation.
Abstract (Originalsprache)
Objective
The morphological classification of microbial cultures and colonies requires specialized knowledge and experience; therefore, automation in this field remains limited. This study examines the feasibility of automating the identification of pathogenic microbial species from colony images of cultured microorganisms.
Methods
Two distinct image datasets were constructed for 10 clinically relevant species: a colony image dataset, consisting of individual colony images cropped, and a tile image dataset, generated by dividing entire images of the culture plate into tiles. Separate ResNet-50–based models were trained in each dataset, and their outputs were integrated to evaluate the classification performance. Training was conducted using 10,048 colony images and 23,003 tile images collected from 418 strains, and performance was assessed with five-fold cross-validation (K = 5).
Results
The colony image model achieved a sensitivity of 0.934 and a specificity of 0.993, while the tile image model achieved a sensitivity of 0.918 and a specificity of 0.991. Integration of the two models into an ensemble model further improved performance, yielding a sensitivity of 0.955 and a specificity of 0.995 when tested on 76 independent strains.
Conclusion
The ensemble model approach provides high accuracy and robustness, suggesting its potential technical contributions to microbial identification in clinical microbiology laboratories.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Clinical Microbiology Analysis Development Research, Juntendo University Graduate School of Medicine, Tokyo, Japan.
- Department of Clinical Laboratory Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan.
- Department of Clinical Laboratory Technology, Faculty of Medical Science, Juntendo University, Chiba, Japan.
ORCID-Profile: Satomi Takei, Yoko Tabe
Förderung
- Japan Society for the Promotion of Science
- Shimadzu
- Kurozumi Medical Foundation
- Morinomiyako Medical Research Foundation
- Charitable Trust Laboratory Medical Research Promotion Fund
Zitieren
APA
Naito, T., Takei, S., Misawa, S., Kuribara, M., & Tabe, Y. (2026). Dual-input deep learning system for microbial identification from blood agar plates. PLOS One, 21(7), e0353761. https://doi.org/10.1371/journal.pone.0353761
Vancouver
Naito T, Takei S, Misawa S, Kuribara M, Tabe Y. Dual-input deep learning system for microbial identification from blood agar plates. PLOS One. 2026;21(7):e0353761. doi:10.1371/journal.pone.0353761
BibTeX
@article{Naito2026,
author = {Naito, Takao and Takei, Satomi and Misawa, Shigeki and Kuribara, Miyuki and Tabe, Yoko},
title = {Dual-input deep learning system for microbial identification from blood agar plates},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0353761},
doi = {10.1371/journal.pone.0353761}
}
EUVIMED