A group of researchers has developed a machine learning method which processes massive amounts of data to help determine which existing drugs could improve outcomes in diseases for which they are not prescribed.
New cheaper uses
The drug reuse It is an interesting activity because it could reduce the risk associated with safety testing of new drugs and dramatically reduce the time it takes to bring a drug to market for clinical use.
But arriving at those new uses usually involves a combination of serendipity and expensive, time-consuming randomized clinical trials.
To combat this problem, researchers at Ohio State University created a model that combines huge data sets related to patient care with high-powered computing. to reach repurposed drug candidatess and the estimated effects of those existing medications on a defined set of outcomes.
The research team used insurance claims data in nearly 1.2 million heart disease patients, providing information on assigned treatment, disease outcomes, and other values.
The deep learning algorithm can also take into account the passage of time in each patient's experience, for each visit, prescription and diagnostic test. Model input for medications is based on their active ingredients.
Although this study focused on the proposed repurposing of medications to prevent heart failure and stroke in patients with coronary artery disease, the model is flexible and could be applied to most diseases.
–
The news
New machine learning algorithm is helping determine which drugs can be repurposed for other conditions
was originally published in
Xataka Science
by
Sergio Parra
.