There is often a serious drawback to clinical trials: the people who receive drugs in them are selected very strictly. Most clinical trials, for example, do not allow pregnant women. And most have age requirements. Also, most do not allow people with conditions other than those tested.
This screening process narrows the available pool of potential volunteers and also unnecessarily excludes many people who may benefit from therapy. Artificial intelligence could change that screen.
A team of researchers at Stanford University, in collaboration with biotech company Genentech, has developed an artificial intelligence-based system that can safely reunite clinical trial participants who may have been previously excluded.
The new system, called Trial Pathfinder, compares the survival results of clinical trial participants included in a large database.
As the system analyzes the data, it learns more about which patients are more or less likely to have problems during a clinical trial of a new drug, depending on various factors, such as the age, weight, whether they are pregnant and their history. doctor. The system can then be used to emulate a clinical trial with the inclusion of people who have previously had a leak. Finally, researchers can use the information from the system when defining criteria for their clinical trial in the real world.
Tests with real-world data on specific applications, such as certain types of cancer, have shown that it is capable of increasing the permitted populations of volunteers in such drug trials by approximately 53%.