Mayo Clinic completes deidentification of expansive medical dataset to protect patient privacy, and advance research discoveries and clinical outcomes.
ROCHESTER, Minn. â€” Mayo Clinic announced a significant achievement of the newly formed Clinical Data Analytics Platform: completion of the deidentification of structured portions of 10 million Mayo Clinic patient records, along with 2.5 million unstructured portions. Structured portions of data include laboratory values, diagnosis codes, vital signs and medications, while unstructured portions of data contain clinical notes.
Without the data leaving Mayo Clinic’s possession or providing patient-identifying information, the Clinical Data Analytics Platform enables researchers and clinicians to extract the knowledge and insights within the deidentified dataset to advance medical research and clinical outcomes.
The Clinical Data Analytics Platform is a venture under the Mayo Clinic Platform, a strategic initiative to improve health care through insights and knowledge derived from data. The Clinical Data Analytics Platform applies advanced data analytics on deidentified data from Mayo Clinic and other organizations, as well as vast information in scientific literature, to advance medicine and improve patient health, like identifying targets and biomarkers for new drugs and matching patients with therapies.
“Mayo Clinic is committed to transforming health care and improving patient lives through platform business models and advanced digital technology. This accomplishment reflects that commitment and the ability to realize that vision in a patient-centric manner that prioritizes privacy,” says John Halamka, M.D., president, Mayo Clinic Platform.
Preserving patient privacy is the priority of the Clinical Data Analytics Platform. All clinical data are deidentified in accordance with the Health Insurance Portability and Accountability Act and Mayo’s privacy policies. All deidentified data will be audited and certified by an independent third-party expert in deidentification. The deidentified data remain securely stored in the Mayo Clinic cloud and never leave Mayo’s secure environment. Approved users of the platform will only see the insights extracted from the data. These measures provide additional layers of protection for patients and minimize the risk of reidentification.
Insights derived from the deidentified data will be available to Mayo Clinic research staff and as a service to the biopharmaceutical industry to identify targets for new drugs, find potential new uses for existing drugs and optimize clinical trials. Mayo’s collaborator, nference Inc., a company that is making biomedical knowledge computable to address urgent health needs, is the exclusive conduit for analytic services to the biopharmaceutical industry. nference’s analytic toolset, nferX, relies on federated learning architecture, which is a privacy-preserving approach that allows multiple entities to build a common artificial intelligence model without sharing raw data.
“This milestone is significant in that it protects patient privacy and provides researchers with the types of deidentified health information that enable insights into capturing underlying disease biology â€• providing an invaluable tool to further discoveries and to test hypotheses,” says Clark Otley, M.D., chief medical officer, Mayo Clinic Platform.
Mayo Clinic and nference researchers use Clinical Data Analytics Platform technology, including nferX, to delve into various COVID-19 research studies, ranging from coagulopathy signatures to long-term virus shedding in COVID-19 patients who are not hospitalized.
“The combination of curated, deidentified clinical data and advanced analytics technology will accelerate scientific discovery and development of new therapeutics, and nferX is already assisting COVID-19 research. We look forward to continuing to advance patient-centered discovery through the Clinical Data Analytics Platform,” says Venky Soundararajan, co-founder and chief scientific officer, Nference Inc.
By the end of 2020, the Clinical Data Analytics Platform team will complete the deidentification of the remainder of the unstructured text data portion. The addition of this data, along with datasets involving imaging, pathology and molecular data in 2021, will create a source of information that scientists can use to advance patient care significantly worldwide.
About Mayo Clinic
Mayo Clinic is a nonprofit organization committed to innovation in clinical practice, education and research, and providing compassion, expertise and answers to everyone who needs healing. Visit the Mayo Clinic News Network for additional Mayo Clinic news and Mayo Clinic Facts for more information about Mayo.
- Duska Anastasijevic, Mayo Clinic Public Affairs, email@example.com