BD teamed up with the Mayo Clinic and its patient data-sharing program to help track real-world performance of its current medical devices and help develop new ones.
Dubbed Mayo Clinic Platform_Discover, the database includes de-identified findings from at least 10 million patients—spanning medical records, doctor’s notes and images, and including as many as 1.2 billion lab test results and 3 million echocardiograms.
BD plans to use artficial intelligence and machine learning to analyze this trove of information and supplement its work in controlled clinical studies.
Randomized, controlled clinical trials are the gold standard for evaluating whether medical devices are safe and work as designed. But clinical trials come with tight parameters for patient eligibility and operations, so experts "are seeing added value in leveraging insights from real-world data to truly understand whether we are meeting patients' needs," Lisa Boyle, BD’s vice president of global clinical affairs and medical affairs strategy, said in a statement.
"We need to be leveraging real-world evidence, using datasets like those from Mayo Clinic Platform, to understand the many parameters that we wouldn't normally capture in a clinical trial and understand patients' care pathways and address the needs of diverse patients in order to establish better solutions for better outcomes and experiences," Boyle added.
The device maker said it will also use the Mayo Clinic’s historical data to build predictive models, with the goal of aiding in study design and supporting new claims for its products in regulatory submissions.
Mayo Clinic Platform, the medical center’s health data offshoot, launched a separate program earlier this year focused on incubating artificial intelligence startups. The immersive, 20-week Platform_Accelerate program aims to help validate AI models and offer support in regulatory, clinical and business development—including access to de-identified Mayo patient data in a secure environment.
The first four startups chosen for the program are Cliexa, focusing on people with multiple chronic diseases such as diabetes and cardiovascular conditions; Quadrant Health, which analyzes electronic health records to triage and predict patient harms; ScienceIO, aimed at reducing administrative burdens for physicians; and Seer Medical, with its home-based epilepsy diagnostics and digital biomarkers for predicting seizures.