Watch Our Webinar on How AI is Changing Patient Identification in Clinical Trials

Biooptronics Webinar Blog Banner
Deep 6 AI Blog

Deep 6 AI Blog

Get insights about clinical trials, patient identification, health IT, data science, machine learning, and more.

The processes for clinical trial feasibility and recruitment are manual, time-consuming, and error-prone. Billions of dollars and months of time are lost each year as sponsors and sites try to determine which sites have the right patient populations to be successful with each trial.

 

We recently teamed up with Bio-optronics for a webinar on how artificial intelligence (AI) is changing how both sites and sponsors are conducting clinical trial feasibility and recruitment. Deep 6 AI’s Vice President of Marketing, Emily Hossellman, walked through the challenges of patient identification and where AI is able to make a difference.

 

Here are a few key takeaways from her presentation:

 

AI changes the game for sites and sponsors

 

Sites have never had a reliable way to conduct feasibility assessments for clinical trials. Most of the relevant data to determine patient eligibility is in unstructured data within the medical record (doctor’s notes, pathology reports, etc.) which needs to be reviewed manually. Even if the team has the time to do a manual review, it will be months before the trial is up and running and the eligible patient pool could be completely different, requiring all the work to be conducted again.

 

 

AI and natural language processing (NLP) provide an opportunity to cut down on all that time-consuming manual review and find eligible patients in real time. Site staff still review the records, but the AI is able to create a targeted pool of patients to review and take researchers directly to the relevant data in the records, saving hundreds of hours for the team and giving sponsors confidence that sites have the right patient population.

 

But it isn’t perfect

 

Artificial intelligence is still in its infancy, so there is a lot of room for improvement. This is particularly true because of how challenging medical records data is to work with. Natural language processing (NLP) derives meaning from unstructured text, but medical records are particularly unstructured. They lack punctuation and sentence structure, and are rife with shorthand, acronyms, and typos. In addition, AI can struggle with boilerplate language on medical forms or information that’s copied and pasted from previous records. (We go into examples of this in much more depth in the webinar.)

 

That’s why in its current state, AI is primarily a tool that makes site staff faster and more effective, rather than a replacement for them. AI can identify a pool of potential patients for consideration, but site staff still needs to review the patients and make the appropriate call about their care.

 

Sites and sponsors that invest in technology will get ahead

 

During the webinar we walked through the impact AI can have on the ability of site staff to recruit patients more quickly and at a much larger scale than they could using their previous methods. The data show that teams that invest in this kind of technology can find and recruit more patients, more quickly – allowing them to run more effective, more profitable clinical trials for the site. It’s a win for sponsors, sites, and most of all, patients.

 

 

Additional considerations in a COVID-19 world

 

While only discussed briefly during the webinar, it’s hard to think about clinical trials operations today without considering what COVID-19 is and will do to affect clinical trials. One of the biggest factors here is that COVID-19 is pausing new recruitment or delaying the start of some clinical trials, with no clear idea of when (or even if) those trials will ramp back up again.

 

Many sites will be caught flat-footed when the time comes to restart their trials. Which trials should you restart? Which of my patients are still eligible (if any)? These are the types of scenarios where the real-time nature of AI can make a huge difference. The day research teams are ready to begin resuming recruitment for their trials, they can conduct a real-time assessment of their eligible patient population to determine which trials they can successfully resume and prioritize their efforts.

 

This is a topic we’ll be discussing in much more depth in the coming months, so be sure to subscribe to our newsletter to stay up to date. And if you’d like to listen to the full recording of the webinar, you can do so here.

 

Watch the recorded webinar!

Share this post

Share on linkedin
Share on facebook
Share on twitter
Share on print
Share on email