Researchers at Ohio State University deployed machine learning to identify new applications of old, FDA-approved drugs.

Ping Zhang, Senior Author, Assistant Professor of Computer Science and Engineering and Biomedical Informatics, Ohio State University, led a team of medical researchers on a project that used AI to suggest new uses for old drugs. The team found that machine learning could process huge amounts of data to determine whether existing, FDA-approved medications could be repurposed for use in other diseases. (International Business Times)

Their procedure is a workable alternative to expensive and time-consuming clinical trials for determining the efficacy and safety of such repurposed drugs.

It also is a safer process and helps to quickly get a drug to market for use by patients.

How it worked

Dr. Zhang and his team were looking for repurposed drug candidates that could prevent heart failure and stroke in patients with coronary artery disease.

They ran their deep learning algorithm on a huge database of insurance claims data related to nearly 1.2 million heart disease patients.

The system found nine drugs that are currently not in use for the treatment of heart disease. However, they still might result in better outcomes for heart patients.

Of these nine drugs, three were already under testing for treating heart disease. The six new drugs however point to the potential for using the system on a more broad-based basis.

“We are the first team to introduce the use of the deep learning algorithm to handle the real-world data, control for multiple confounders, and emulate clinical trials,” Zhang said in this regard to News Medical.

Speeding up clinical trials

“This work shows how artificial intelligence can ‘test’ a drug on a patient, and speed up hypothesis generation and potentially speed up a clinical trial,” says Zhang. “But we will never replace the physician – only clinicians will make drug decisions.”

There’s already a startup doing this

In July last year, The San Diego Union-Tribune reported that Repurpose.AI, a San Diego start-up, was using artificial intelligence to find new uses for existing or abandoned drugs.

Repurpose.AI uses algorithms to perform billions of calculations to analyze 10,000 features of a molecular structure. These molecules are drawn from 20,000 drug compounds in initial clinical trials. The system also covers 4,000 compounds that the FDA has already approved.

According to George Nicola, founder, the company’s high-power algorithms shorten the process of getting a drug into clinical trials from years to months.

Related Story:  Baricitinib, An AI-flagged Rheumatoid Drug, Could Treat COVID-19