Research on AI model to be presented at ASCO 2019 Annual Meeting;
Useful for assessing patient risk and treatment options, evaluating cost and quality of care or determining trial eligibility
BOSTON, MA, May 28, 2019 – ConcertAI announced today that it has developed an AI model predicting survival of lung cancer patients 3-12 months from their last clinical visit. The model allows clinical researchers to gain deeper insights into key variables impacting patient survival. This model will be presented as a research abstract during a poster session at the 2019 American Society of Clinical Oncology’s (ASCO) Annual Meeting in Chicago, on June 1, from 1:15-4:15 p.m.
The Gradient Boosting model was built using de-identified structured data from 55,000 lung cancer patients in ASCO’s CancerLinQ Discovery® database. Nearly 4,000 unique variables, including diagnostic and therapeutic codes, biomarkers, surgeries and lab tests, were accessible to the model. Model validation was done on a randomly selected and reserved set of 8,468 patients. Results were significantly better than a baseline Cox-PH model and compared very favorably to other survival models in the literature created using AI and comparable machine learning techniques.
“A model that can accurately predict risk of mortality in patients at various time points could help guide clinical development teams to more effectively select attributes of patients for different pre- and post-approval studies to bring needed new therapies forward and enhance the effectiveness of existing therapies respectively,” said Smita Agrawal, Ph.D. and Senior Director of Product Management at ConcertAI and the research abstract’s primary author. “This could have a huge impact on the accessibility of new treatments and overall costs of care.”
“There are enormous costs associated with every patient recruited into a clinical trial,” said Jeff Elton, Ph.D. and CEO of ConcertAI. “The stakes are high but researchers just haven’t had the tools to give them the insights to change how they work and design trials with speed and efficiency. Researchers can now determine with a high degree of confidence which patients are expected to survive the next three months and recruit accordingly to ensure a successful clinical trial at a faster pace. This class of ConcertAI models have immense positive implications for both the patient entering the trial and for clinical trial researchers designing those studies.”