Researchers use clinical trials to evaluate a medical or surgical intervention. It’s a way to determine whether a new treatment such as a drug or diet is safe and effective for people.
Unfortunately, clinical trials don’t always catch the possible side effect that a drug may have. Instead, these reactions pop-up only after the drug has reached the patient, leading to morbidity or mortality.
As you can imagine, a system that can predict the frequencies of a specific drug’s side effects could be a game-changer.
In a statement to the press, Professor Alberto Paccanaro from Royal Holloway, University of London, said:
“It is extremely important to predict what the frequencies of the side effects of the drugs will be after the first stages of clinical trials. At the moment, we don’t have any systems that can do this.”
But that could change soon. Thanks to a machine learning approach developed by Paccanaro and a colleague, Dr. Diego Galeano.
Here’s how it works.
Using Artificial Intelligence to Predict the Frequencies of Drug Side Effects
The new algorithm relies on the same principle that Netflix uses to recommend movies to users. But, instead of suggesting a TV show to binge, the AI predicts the potential side effects of drugs before they hit the market.
It can also determine the percentage of people with side effects after the first stage of human trials. That way, researchers will know how to direct the trial in the future.
According to Professor Paccanaro, this is the first AI system to be capable of such a feat.
The accurate estimation of a specific side effect’s frequency is essential to patient care in clinical practice. Such an algorithm could also save pharmaceutical companies a lot of time and effort.
For example, it could help reduce the risk of drug withdrawal from the market. At the same time, the AI model can prevent costly reassessment of side effect frequencies through new clinical trials.
Dr. Shantao Li and Professor Mark Gerstein from Yale University also contributed to the paper.
Comments (0)
Most Recent