AI to help find patients for clinical trials
PASADENA: Of the many inefficiencies throughout the long, expensive drug-discovery process, patient recruitment for clinical trials can be a particularly brutal bottleneck. For example, various studies have shown that as many as 40% of trials fail to meet their enrollment goals.
Enter Deep 6, a Pasadena-based startup announcing $17 million in fresh funding at a $50 million valuation for its artificial intelligence-powered technology that can suggest candidates for clinical trials in “minutes instead of months.” It has raised $22 million total.
One of the reasons that clinical trial recruitment takes so long is that the patient health information necessary to determine if someone is a good match for a given trial is spread across electronic medical records (EMRs), physicians notes, pathology reports and other forms of documentation. Deep 6 uses natural language processing—an AI technique that allows software to “understand” language—to pull together information from those systems, parse it and let researchers filter for specific conditions and traits. It has also trained its software to make inferences about what condition a person may have based on their symptoms, so that it can flag people for whom that condition may not be explicitly written out.
Bill McKeon, CEO of Texas Medical Center, a collection of 61 different health institutions, says that TMC is rolling out Deep 6 software across its network to replace the typical process of finding clinical trial candidates, which requires associates to manually flip through thick folders of medical records.
“It’s just a long slog to find patients for clinical trials,” he says.
When TMC researchers compared one recruitment scenario to results driven by Deep 6’s software, however, the difference was dramatic, McKeon says: In a given case study, it took six months to find 12 eligible patients for a trial, while the same matching process through Deep 6’s software found 80 potential people in minutes.
“It’s just completely transforming,” McKeon says.
Deep 6’s Wout Brusselaers says that beyond Texas Medical Center the company has signed on roughly 20 customers, including other health centers as well as research organizations and medical device companies.
He highlighted another instance in which a researcher at Cedars-Sinai Medical Center took six months to find two suitable patients for a cardiology study. Deep 6 software identified 16 possible patients in less than an hour.
“Because 90% of clinical trials are delayed, there’s a huge impact on the speed of innovation in healthcare,” he says, “let alone the impact on the lives of patients who are waiting desperately for a new cure that could save their lives.”
Deep 6 itself doesn’t have access to any sensitive patient information, because its software sits on top of a health system’s data layer. It makes money through licensing its software while also working to create a marketplace model where pharmaceutical or medical device companies use the platform to connect with medical centers. Deep 6 will take a transaction fee on any resulting partnerships.
The startup is just one of a handful of companies applying artificial intelligence to clinical trials. For example, hospital system Health Quest is using IBM Watson’s Clinical Trial Matching tool for similar work, and New York City-based startup called Antidote makes it easier for patients to search through clinical trials. Meanwhile, other companies are taking on different parts of the drug discovery process, like Insitro, which uses AI to suggest possible therapies.
