Identification of pathogens and detection of antibiotic susceptibility at single-cell resolution by Raman spectroscopy combined with machine learning
Problem
Rapid, label-free identification of pathogenic bacteria and their antibiotic resistance at single-cell resolution, avoiding slow culture and susceptibility assays.
Why it matters
Faster pathogen and resistance detection enables earlier diagnosis and better antibiotic use in clinics.
What the AI did
Machine learning classifies single-cell Raman spectra to identify bacterial species and detect antibiotic resistance without culture.
Results
Average identification accuracy of 90.73 ± 9.72% across 12 common pathogenic bacterial species.
Impact
Peer-reviewed study demonstrating ~90.7% average identification accuracy at single-cell resolution in a lab setting.
Record
| Venue | Frontiers in Microbiology |
|---|---|
| Published | 2023 |
| Work type | article |
| Authors | Weilai Lu, Haifei Li, Haoning Qiu, Lu Wang, Jie Feng, Yu Vincent Fu |
| Citations | 45 (OpenAlex, at import time) |
| References | 65 |
| Model or method | not stated in source |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3389/fmicb.2022.1076965 |
| Open access | gold |
| Keywords | Raman spectroscopy, machine learning, pathogen identification, antibiotic resistance, single-cell |
Importance rating
significant — thousands to millions affected
Rapid pathogen and antibiotic-resistance detection addresses a stated clinical need for early diagnosis and antibiotic optimization; abstract gives lab accuracies but no quantified patient or health-system scale.