Here are five things to know from the study:
- The tool, called RAG-Enabled Clinical Trial Infrastructure for Inclusion Exclusion Review, or RECTIFIER, analyzed electronic health record information to determine if patients met the necessary criteria to enroll in a heart failure study, according to a Feb. 17 news release from the health system.
- Of the 4,476 patients screened for the study, 2,242 were screened with the AI tool and 2,234 were screened manually by staff.
The AI tool found 458 eligible patients compared to manual screening, which found 284 eligible patients.
Of the eligible patients, 35 from the AI group and 19 from the manual group enrolled in the study.
- “The rate of enrollment in the AI-enabled arm was almost double the rate of enrollment in the manual arm,” lead author Ozan Unlu, MD, a clinical informatics fellow at Mass General Brigham and a cardiovascular medicine fellow at Brigham and Women’s Hospital, said in the release. “This means that AI could almost halve the time it takes to complete enrollment in a trial.”
- Because previous research has shown AI can introduce bias, the researchers analyzed the race, gender, and ethnicity of enrolled patients and found no significant differences.
- “By adjusting the eligibility questions that the RECTIFIER tool asks of the medical record notes, AI screening can be applied to trials assessing cancer treatments, diabetes interventions, and many others,” co-senior author Alexander Blood, MD, Brigham and Women’s Hospital cardiologist and associate director of Mass General Brigham’s Accelerator for Clinical Transformation, said in the release.
Read the full study here.
Leave a Reply