The algorithms have the potential to improve patient outcomes and reduce the need for new trials to reassess missed side effects during the first clinical trials.
Drug side effects have always been notoriously hard to predict over the long term, so applying artificial intelligence to this question is a natural fit. A new algorithm created by the Royal Holloway University of London is intended to do just that.
It has a unique proposition. It helps researchers understand and predict potential drug side effects much the same way algorithms provide recommendations for new movies. While this may seem trite, it promises a new way forward with drug research.
It uses a machine learning approach developed by Dr. Diego Galeano and Professor Alberto Paccanaro, stepping into the gap left between symptoms presented at clinical trials and what actually happens when the drug reaches the market.
It’s this gap that causes researchers and developers anguish and contributes to morbidity and mortality associated with all potential drugs. This algorithm helps better predict what side effects will appear and, more importantly, who will experience them.
The algorithm could significantly streamline the time to market process as drug developers gain a greater understanding of how side effects will appear without continued lengthy trials. The algorithm is essential because there are no systems in place to explore these potentialities.
Right after the first stages of trials, the algorithms will take over to help begin predicting the rate of side effects for these drugs, helping to simplify the process.
The algorithms have the potential to improve patient outcomes and to relieve the burden pharmaceutical companies face trying to bring a new solution to market. It reduces the potential that new trials will have to take place to reassess missed side effects during the first clinical trials.
Because numerous side effects are missed during all stages of clinical trials, opening pharmaceutical companies up to lawsuits and causing poor patient outcomes. Instead, the algorithm creates a better evaluation system to catch missed side effects through its unique prediction model.
As more drugs are tagged with side effect percentages, we could have faster times to market and more reliable drug solutions for some of our most persistent issues. This could be a new era for drug trials, one that involves less frustration and more consistent results.