Artificial intelligence to combat antibiotic-resistant bacteria

Although it generates skepticism in some sectors, the use of AI in medicine breaks new ground in the area, by allowing large volumes of data to be analyzed quickly, contributing to decision-making after combining new information with patterns detected by these systems, under a look that usually escapes human analysis, at least in the first instance.

A medical problem that has not yet been fully resolved is the existence of bacteria resistant to antibiotics. Currently, in the absence of a concrete solution, there are people who die in hospitals after contracting an infectious condition of this kind.

An AI solution to combat antibiotic resistance

Resources already exist to combat antibiotic-resistant bacteria, but the task of finding which antibiotics are still effective against a particular pathogen often takes two or more days. This delay is due to the fact that the samples are grown in laboratories and the drug tests are carried out in Petri dishes, to evaluate their effectiveness. The problem is that, in addition to being an expensive process, many critically ill patients cannot wait more than two days.

Scientists from the ETH Zurich University, from Switzerland, trained AI algorithms on the mass spectrometry data, to teach them to detect antibiotic resistance on their own. This mechanism should offer very fast results, since the scientists point out that in this way it is possible to detect signs of resistance to antibiotics in bacteria up to 24 hours earlier than current diagnostic tools. By detecting these signs early, doctors can design antibiotic therapy more precisely and start treatment sooner.

To train this AI system, the scientists used a dataset of more than 300,000 mass spectra of individual bacteria. The resulting database covers around 800 different bacteria and more than 40 different antibiotics, for which the algorithm was able to detect antibiotic resistance autonomously.

What is most important about this approach is that algorithms can answer these questions very quickly, which clears the way for rapid and tailored antibiotic therapies, especially in severe cases of infection.

In the near future, the ETH Zürich University team plans to launch a clinical trial, to assess how an AI-based approach would work under real-life conditions, as the research has so far only been carried out in a laboratory.

Source link

Lenny Li

I started to play with tech since middle school. Smart phones, laptops and gadgets are all about my life. Besides, I am also a big fan of Star War. May the force be with you!

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button