Current antimicrobial susceptibility testing methods rely on slow processes, hindering urgent treatment for life-threatening infections like sepsis, leading to potential overuse of antibiotics and increased antimicrobial resistance. Researchers have developed a novel and rapid test. The test can provide results within 30 minutes, using a combination of fluorescence microscopy and artificial intelligence (AI) to detect antimicrobial resistance (AMR).
How does it work?
- The method involves training deep-learning models to analyse bacterial cell images, identifying structural changes that occur when cells are treated with antibiotics.
- The researchers believe that the rapid nature of their method could facilitate targeted antibiotic treatments, reducing treatment times, minimising side effects, and slowing down the rise of AMR.
How successful is this test?
- The model demonstrated effectiveness across multiple antibiotics, achieving at least 80% accuracy on a per-cell basis.
- The AI-based approach was tested on clinical isolates of E. coli with varying levels of resistance to the antibiotic ciprofloxacin, detecting resistance at least 10 times faster than established clinical methods.
Future work:
- The researchers aim to further develop the method for faster and more scalable clinical use, adapting it for different bacteria and antibiotics.
Read more at: Novel AI-powered method detects antimicrobial resistance within 30 minutes