Can AI Drive More Diversity in Drug Development?

Nov. 29, 2022 – Artificial intelligence could help improve diversity, equity and inclusion in clinical trials and drug development by overcoming some traditional human biases in these areas, but we’re not there yet, experts say. The technology could also help doctors with data insights to make diagnosis and treatment more accurate.

That starts with the quality. Artificial intelligence (AI) relies on large amounts of data to create algorithms — or computer instructions — to develop best practices and predictions. But the guides are only as good as the data used to create them. And humans are the ones creating the data.

“The basis for the development of AI technologies is people, and these people have their own biases,” said Naheed Kurji, executive chair of the Alliance for Artificial Intelligence in Healthcare. “As a result, the algorithms will have their own biases.”

An example is technology that uses language to diagnose diseases.

“There are many cases, examples where companies have not recognized the language differences between different cultures,” says Kurji. If the technology is based on speech patterns from a limited demographic, “then if this model is applied in the real world to another demographic with a different accent, this model fails.”

“As a result, it’s not representative.”

Another example is genetic and genomic data.

“Give or take, over 90 percent of genetic and genomic data comes from people of European descent. It’s not from people of Africa, Southeast Asia, Asia, or South America,” says Kurji, who is also the president and CEO of Cyclica Inc., a Toronto-based data-driven drug discovery company.

Therefore, “a lot of the research that’s been done at this level of data is inherently biased,” he says.

To be fair

Creating data that respects the diversity, equity and inclusion of people and cultures around the world is not a hopeless challenge. But experts say it will take time. Once this is achieved, AI should be closer to being free of human and systemic bias.

Greater awareness is essential.

“The solution to the problem comes from people who inherently understand that there is bias,” says Kurji, and then only include fair and balanced data that passes a diversity test.

Choose smarter?

Another promising avenue for AI is to streamline the drug development process, narrow down potential drug candidates, and make clinical trials more cost-effective.

“If the source data has challenges and limitations, then AI will just propagate those limitations,” agrees Sastry Chilukuri, co-CEO of data-driven clinical trials company Medidata and founder and president of Acorn AI. “The source data needs to become more representative and fairer so that the AI ​​can reflect what’s happening.”

When it comes to human or systemic biases in drug development, “it would be an oversimplification to say that AI or machine learning can fix them,” says Angeli Moeller, PhD, Head of Data and Integrations Generating Insights at Roche in Berlin. “But using AI and machine learning responsibly can help us identify bias and find ways to mitigate potential negative impacts.”

Silent Partners?

At the same time, as AI aims to streamline drug development, the technology can also help all doctors do their jobs better, experts say. For example, AI would help spread knowledge and expertise far and wide, and share best practices from experienced doctors with more complex patients. This would help those who only treat a few such patients each year.

The surgical volume in New York City or in Delhi could be as high as hundreds of patients a year, says Chilukuri. “But when you go into US interiors like Nebraska, the surgeon just doesn’t see that much volume.”

AI could help doctors “by providing the kind of tools that would allow them to deliver the same world-class care to their entire population much more quickly,” he says.

increase efficiency

AI could help make therapy more targeted by using data to identify patients at highest risk. The technology could also improve some bottlenecks in medicine, such as the time it takes to interpret X-ray images, says Kurji.

There’s one AI company “whose entire business model isn’t about replacing your radiologist, it’s about making radiologists better,” he notes. One of the company’s goals is to “prevent death or serious illness from missed X-ray scans or being left in the stack and simply not being treated quickly enough for that patient.”

Radiologists are so busy they may only have 30 seconds or less to interpret each scan, Chilukuri says. AI can flag a potentially worrying lesion, but also compare an image to previous scans of the same patient. This view of AI applies not only to radiology, but to all data-driven areas of medicine.

Advancing personalized medicine

AI could also guide a personal approach to surgery “because it’s not like people come in small, medium and large,” says Chilukuri. The technology could help surgeons determine exactly where to operate on an individual patient.

Moeller agrees that AI has the potential to advance personalized medicine.

“AI can help diagnose and predict risk, which earlier interventions may mean,” says Moeller, who is also vice chair of the board of directors of the Alliance for Artificial Intelligence in Healthcare. “For example, if you look at a diabetic, what is that? Likelihood that he or she will develop eye problems due to diabetic macular edema?”

The technology could also help to see the big picture.

“Machine learning can look for patterns in a population that might not be in your medical textbook,” says Moeller.

Beyond diagnosis and treatment, AI could also help with recovery by tailoring rehabilitation for each patient, Chilukuri predicts.

“It’s not like every person goes into rehab the same way. So you have highly customized AI plans that allow you to actually stay on track and predict where you’re going.”