When penicillin was discovered, it was heralded as “the silver bullet” because of its extraordinary ability to eradicate germs that cause illness without endangering humans. Since then, several other antibiotics have been created that target a variety of bacteria; however, the more frequently these antibiotics are used, the higher the chance that strains of bacteria may evolve resistance to them.
Their findings were published in the journal Frontiers in Microbiology.
The global issue of antibiotic resistance poses a serious threat to public health as it reduces the number of effective treatments available for bacterial illnesses. It is crucial to promptly identify antibiotic-resistant bacteria in order to guarantee that patients receive appropriate therapy; nonetheless, the most accessible approach to accomplish this entails cultivating the bacteria in a lab for many days and administering medications to observe their reaction.
There is some evidence that antibiotic resistance reveals itself in other ways; for example, the morphology of Gram-negative rod-shaped bacteria changes when they are exposed to antibiotics,
We were interested in determining whether this feature could be used to detect antibiotic resistance without actually treating the bacteria with antibiotics.
Miki Ikebe
Also, Read| Fever enhances the mitochondrial damage in immune cells
In order to do this, the scientists gave Escherichia coli set amounts of several antibiotics, which caused the bacteria to become resistant to the drugs. They then withdrew the antibiotic treatment and utilised machine learning to analyse the forms, sizes, and other physical properties of the bacteria based on microscope photos.
The results were very clear,
The antibiotic-resistant strains were fatter or shorter than their parental strains, especially those that were resistant to quinolone and β-lactams.
Kunihiko Nishino
In order to determine whether bacterial morphology and antibiotic resistance were related, the researchers then looked at the genetic composition of the resistant bacteria. The findings demonstrated that the shape alterations seen in the antibiotic-resistant bacteria were, in fact, linked to genes involved in energy consumption and antibiotic resistance.
Our findings show that drug-resistant bacteria can be identified from microscope images, in the absence of antibiotics, using machine learning,
Miki Ikebe
Based on the comparable shapes and sizes of the bacteria resistant to quinolone, β-lactams, and chloramphenicol, it is plausible that the same genetic process causes antibiotic resistance in all of these strains. In the future, a machine learning tool may be able to quickly evaluate patient samples in order to recommend the best medication for treating their infection.
Source: Osaka University – News
Journal Reference: Ikebe, Miki, et al. “Bioinformatic Analysis Reveals the Association between Bacterial Morphology and Antibiotic Resistance Using Light Microscopy with Deep Learning.” Frontiers in Microbiology, vol. 15, 2024, p. 1450804, DOI: https://doi.org/10.3389/fmicb.2024.1450804.
Last Modified:
