SC-MAPS: Image-based deep learning for bacterial identification and antibiotic-susceptibility testing
- 1University of Oxford
Abstract
Rapid bacterial identification could help guide antimicrobial susceptibility testing (AST), but image-based classification must generalise across bacterial strains. We developed SC-MAPS, a multichannel microscopy dataset comprising 916,435 single cells from 13 bacterial species, with independent strains reserved for evaluation. Combining phase-contrast and fluorescence information using deep learning achieved 65.4% single-cell species-classification accuracy on held-out strains. To assess whether imperfect species predictions retained useful information for downstream testing, we defined eight antimicrobial-resistance-informed AST-routing groups. Pooling species probabilities increased routing accuracy to 77.1%, rising to 91.1% when predictions from five cells were combined and 96.2% with 20 cells. Synthetic mixtures revealed limitations in minority-species recovery, while exploratory analysis of unseen organisms demonstrated the potential for out-of-distribution detection. SC-MAPS provides a framework for investigating transferable bacterial phenotypes and developing image-based identification and AST-routing strategies.
Related
- Performance of Direct Identification and EUCAST Rapid Antimicrobial Susceptibility Testing in Bloodstream Infections Using Sepsityper® and Disk Diffusion Methods
- Bacterial profile and antibiotic susceptibility of bloodstream infections among infants, children, and adolescents undergoing treatment for acute leukemia
- Application of NaCl-supplemented antimicrobial susceptibility testing conditions to marine bacterial isolates
- Rapid identification and antimicrobial susceptibility testing of pathogenic microorganisms from positive blood cultures
- Benchmarking four commercial antimicrobial susceptibility testing systems and disk diffusion to reference broth microdilution in the assessment of gram-negative bacterial isolates
BibTeX
@misc{tiwari2026scmaps,
title = {{SC-MAPS: Image-based deep learning for bacterial identification and antibiotic-susceptibility testing}},
author = {A. Tiwari and P. Turner and C. Feehily and C. Nellaker and N. Stoesser and A. Kapanidis},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.23.26366325},
url = {https://www.medrxiv.org/content/10.64898/2026.10.23.26366325v1},
note = {CC BY-NC},
}
RIS
TY - UNPB
TI - SC-MAPS: Image-based deep learning for bacterial identification and antibiotic-susceptibility testing
AU - A. Tiwari
AU - P. Turner
AU - C. Feehily
AU - C. Nellaker
AU - N. Stoesser
AU - A. Kapanidis
PY - 2026
DA - 2026/10/23
PB - medRxiv
DO - 10.64898/2026.10.23.26366325
UR - https://www.medrxiv.org/content/10.64898/2026.10.23.26366325v1
AB - Rapid bacterial identification could help guide antimicrobial susceptibility testing (AST), but image-based classification must generalise across bacterial strains. We developed SC-MAPS, a multichannel microscopy dataset comprising 916,435 single cells from 13 bacterial species, with independent strains reserved for evaluation. Combining phase-contrast and fluorescence information using deep learning achieved 65.4% single-cell species-classification accuracy on held-out strains. To assess whether imperfect species predictions retained useful information for downstream testing, we defined eight antimicrobial-resistance-informed AST-routing groups. Pooling species probabilities increased routing accuracy to 77.1%, rising to 91.1% when predictions from five cells were combined and 96.2% with 20 cells. Synthetic mixtures revealed limitations in minority-species recovery, while exploratory analysis of unseen organisms demonstrated the potential for out-of-distribution detection. SC-MAPS provides a framework for investigating transferable bacterial phenotypes and developing image-based identification and AST-routing strategies.
ER -