What’s New: The University of California San Francisco’s (UCSF) Center for Digital Health Innovation (CDHI) is using Intel® Software Guard Extensions (Intel® SGX) featured in Intel® Xeon® E processors and Fortanix® Confidential Computing Enclave Manager to help streamline certification of breakthrough medical devices with embedded artificial intelligence (AI) capabilities. The UCSF BeeKeeperAI™ project utilizes the unique hardware enhanced security of Intel SGX to accelerate the validation of the devices’ data and algorithms in order to improve both patient care and privacy. Validating these algorithms against multiple distinct datasets owned by various organizations can be challenging and time-consuming, but essential for user security.
“UCSF’s Center for Digital Health Innovation is pleased to be collaborating with Fortanix, Intel and Microsoft Azure to establish a confidential computing platform with privacy-preserving analytics to accelerate the development and validation of clinical algorithms. The platform will provide a ‘zero trust’ environment to protect both the intellectual property of an algorithm and the privacy of healthcare data. Using Fortanix Enclave Manager for orchestration of Intel’s SGX enclaves on Azure confidential computing infrastructure with Azure Kubernetes Service (AKS), and CDHI’s proprietary BeeKeeperAI data access, transformation, and orchestration workflows, the platform will reduce the time and cost of developing clinical algorithms.”
–Michael Blum, MD, associate vice chancellor for informatics, executive director of CDHI and professor of medicine at UCSF
Why It Matters: UCSF is leveraging the Fortanix Confidential Computing Enclave Manager platform that uses Intel SGX to help protect the privacy of patient data. Intel SGX enables the platform to create a trusted computing environment that offers hardware-based memory encryption that helps isolate specific application code and data in memory. This means the BeeKeeperAI project can use these private regions of memory, called enclaves (or Trusted Execution Environments or TEEs), to increase the security of application code and data (to run signed applications in enclaves). This enables other organizations to confidently work together to validate the algorithms while helping to keep each organization’s data confidential and protect intellectual property.
Gaining regulatory approval for clinical AI algorithms requires highly diverse and detailed clinical data to develop, optimize and validate unbiased algorithm models. Algorithms used in the context of delivering healthcare should be capable of consistently performing across diverse patient populations, socioeconomic groups and geographic locations, and be equipment-agnostic. Few research groups, or even large healthcare organizations, have access to enough high-quality data to accomplish these goals.
How It Works: The platform will provide a “zero-trust” environment designed to protect both the intellectual property of an algorithm and the privacy of healthcare data, while CDHI’s proprietary BeeKeeperAI will provide the workflows to enable more efficient data access, transformation and orchestration. The confidential computing technology helps protect the privacy of patient data by enabling a specific algorithm to interact with a specifically curated dataset, which remains at all times in the control of the healthcare institution. The data is placed into an enclave protected by Intel SGX and leveraging Fortanix cryptographic functions, including validating the signature of the algorithm’s image.
“BeeKeeperAI will accelerate the development of clinical AI algorithms while protecting the privacy of patient data with confidential computing technology,” said Ambuj Kumar, CEO and co-founder, Fortanix. “This groundbreaking new healthcare AI platform will enable life-saving clinical AI algorithms to be validated in days instead of years. The UCSF Center for Digital Health Innovation is using Fortanix technology to secure both the private patient data and the intellectual property of the artificial intelligence algorithms, enabling researchers to combine diverse sets of data while ensuring complete data privacy.”
More Context: Intel Xeon Scalable Platform Built for Most Sensitive Workloads (News Release)