Related Experiment Video
Updated: Jul 2, 2026
Multiplex Detection of Bacteria in Complex Clinical and Environmental Samples using Oligonucleotide-coupled Fluorescent Microspheres
Published on: October 23, 2011
BiFusionPathoNet: fusion network for drug-resistant bacteria identification via optical scattering patterns
Yichuan Wang1, Xu He2, Mubashir Hussain1,3
1Engineering Research Center of Intelligent Theranostics Technology and Instruments, Ministry of Education, School of Biomedical Engineering and Informatics, Nanjing Medical University, Nanjing, 211166, China. liubin@njmu.edu.cn.
This study introduces an AI-powered method using light and Raman scattering to rapidly detect drug-resistant bacteria like methicillin-resistant Staphylococcus aureus (MRSA). The multimodal approach achieved high accuracy, outperforming single-signal methods.
Area of Science:
- Biotechnology
- Medical Diagnostics
- Artificial Intelligence
Background:
- Antibiotic resistance is a growing global health threat, necessitating rapid diagnostic tools.
- Distinguishing between methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-sensitive Staphylococcus aureus (MSSA) is crucial for effective treatment.
Purpose of the Study:
- To develop and evaluate a rapid, accurate method for identifying drug-resistant bacteria, specifically MRSA.
- To compare the performance of single-modal (MDLS or Raman) and multimodal AI-driven detection models.
Main Methods:
- A microfluidic platform integrated with optical fibers was used to collect Multi-angle Dynamic Light Scattering (MDLS) and Raman scattering signals from bacteria.
- Three artificial intelligence models were developed: ResistNet (MDLS), SERB-CNN (Raman), and BiFusionPathoNet (multimodal fusion).
Main Results:
- ResistNet achieved 83.8% accuracy on MDLS data.
- SERB-CNN attained 91.84% accuracy on public Raman data and 93.5% on custom data.
- BiFusionPathoNet, the multimodal model, reached a superior accuracy of 96.8%, significantly outperforming single-modal approaches.
Conclusions:
- The multimodal strategy combining MDLS and Raman scattering signals with AI is highly effective for rapid and accurate detection of drug-resistant bacteria.
- This approach offers a promising solution for timely diagnosis and management of infections caused by resistant pathogens like MRSA.
