One Data Network to Rule Them All
Mayo Clinic Platform_Connect is transforming how patient data is used to generate innovative diagnostic tools and treatment options.
By John Halamka, M.D., President, Mayo Clinic Platform, and Paul Cerrato, senior research analyst and communications specialist, Mayo Clinic Platform
No doubt many readers will ask: What exactly is Mayo Clinic Platform_Connect, and will it really transform patient care? Mayo Clinic Platform_Connect is a distributed data network that partners with health systems, payers, medical device companies, and academic medical centers. Its goal is to enable better ways to diagnose, treat, and even prevent disease, and it seeks to accomplish this by connecting clinical data through a federated, secure architecture. Each partner in Connect brings its own collection of clean, curated, de-identified data to the program, enabling users to find new insights about complex and rare health conditions and discover a wide range of treatments and therapies that can be applied to diverse communities in urban and rural settings.
More specifically, the massive data set enables users to analyze patient records and develop unique AI-driven algorithms while at the same time knowing that any data they contribute to Connect will remain safe and will not be sold, removed, and taken outside their own firewalls. That’s accomplished through a data encryption and de-identification process we refer to as Data Behind Glass*. It’s a secure, federated architecture in which data and intellectual property remain under the control of each respective Connect partner or model developer and are only viewed or used as authorized.
We follow a secure-by-design development approach. A Connect partner’s current infrastructure is configured to operate as a data node within the network. A node represents a database where each partner retains local control and administration rights, including security. Connect partners continuously monitor their respective nodes, while Mayo Clinic Platform provides oversight and governance of the entire network. Using unique cryptographic methods, Connect partners and model developers communicate about subsets of data and models via cloud-enabled, federated tools within their respective nodes. Complete copies of models or data are never exchanged; neither model developers nor Connect partners can view the models or data at any time. In effect, communication between data nodes is blind and secure.
What sets Mayo Clinic Platform_Connect apart from other distributed data networks is its reach. Mayo Clinic recently partnered with provider networks across the globe to transform patient care. Each of these partners has contributed data to the distributed data network. The new partners include Hospital Israelita Albert Einstein in Brazil, Sheba Medical Center in Israel, and University Health Network (UHN) in Canada. These recent additions join Mercy, St. Louis, the sixth largest Catholic Health Care System in the U.S.
Until recently, complex types of medical information, such as integrated electronic health records data, radiology, pathology images, and videos, were too unstructured and complex to analyze. The diversity of languages presented a significant barrier to consolidating clinical data globally. With the combination of privacy-protected, cloud-based storage and the growth of AI and machine learning, Mayo Clinic Platform and the members of Connect will use aggregated, de-identified clinical data to generate patterns to pinpoint disease earlier and identify the best treatment options.