BioTel Research Blog

April 01, 2020

AI Techniques Gaining Traction in the Life Science Industry

Artificial Intelligence (AI) and Machine Learning in Clinical Trials

Our own Dr. Jonathan Riek participated in a recent panel discussion on this topic at OCT West in early March. There are several themes that surfaced during the panel discussion – let’s investigate two of them below.

Patient Selection and Recruitment

A majority of clinical trial failures are attributed to insufficient recruiting and selecting techniques, as well as an inability to effectively monitor patients. More than 8 in 10 trials fail to meet enrollment deadlines. Reasons for this include: eligible patients may not be at specific stage of a disease; suitable patients may find the process too complex and cumbersome to navigate.

AI tools can help enhance patient selection by choosing patients who are more likely to have a measurable clinical endpoint, and identifying a population more capable of responding to treatment. Such tools can automatically analyze EHR and clinical trial eligibility databases, find matches between specific patients and recruiting trials, and recommend these matches to doctors and patients.

Identifying Predictive Biomarkers

Drug developers utilize biomarker data during the development process. Machine learning allows researchers to explore all potential biomarkers and the wide variety of interactions among variables to select the most relevant ones for a trial. It also helps researchers understand the cause-and-effect relationships within trial data. By discovering which patients benefit from the drug in relation to the trial population as a whole, researchers can optimize criteria for later trials, speeding up the drug approval process.

AI simulations can also help answer why a given clinical trial fails. Specifically by addressing ‘what if’ questions. What if drug A instead of drug B had been given to patient X? One of AI’s most valuable contributions is not only identifying key patterns in data, but its ability to run simulations that ‘reason’ what could happen based on those patterns.

Our industry will continue to take AI and machine learning further in clinical trials. It is important to emphasize that this is made possible by the ongoing collaboration and data sharing of doctors, scientists, engineers, researchers, and so many others that make this process complete.  The future potential of this technology will continue to grow as we capture lessons learned to accomplish successful clinical trials together. 

Written by George Athas

BioTel Research, One Preserve Parkway, Suite #600, Rockville, MD 20852