Smarter healthcare: Overcoming data security and privacy concerns through federa...

Sunday, 25 June 2023 10:02

Smarter healthcare: Overcoming data security and privacy concerns through federated learning

Andrew McLean, Managing Director Australia and New Zealand, Intel

GUEST OPINION: Collaboration on a global scale leads to the type of innovation that delivers improved outcomes and better decision-making. Through diverse and expansive datasets, our collective knowledge expands. This enables significant improvements in healthcare, from disease detection to identification of operable regions.

Intel’s federated learning hardware and software recently proved this in the largest medical federated learning study to date, using an unprecedented global dataset from more than 6,000 brain tumour patients across 71 institutions and six continents.

In a joint study led by Intel and the Perelman School of Medicine at the University of Pennsylvania (Penn Medicine), the team demonstrated the ability to improve the detection of the most common and fatal adult brain tumour, glioblastoma (GBM) by 33 percent using a privacy-preserving AI technique called federated learning and improve the identification of operable regions of tumours using an AI software platform called Federated Tumour Segmentation (FeTS).

To treat diseases like GBM, researchers must have access to large amounts of diverse medical data – in most cases, datasets that exceed what one facility can produce. Federated learning’s privacy-preserving approach enables secure data sharing and allows organisations from around the world to collaborate on projects.

Centralised learning is no longer sustainable in healthcare

Centralised learning has long been the traditional norm in AI modelling. This method involves collecting datasets from various locations and devices, and then sending them to a centralised location where the ML model training occurs.

This leads to several risks. Firstly, data stored at a single location can be stolen and exposed, causing huge liabilities to the institution responsible for storing it. Secondly, data owners might not even want to share their raw data in the first place. Though the data owners may be willing to share it for training, the raw data itself may be too sensitive to share.

Security and privacy concerns make joint research and collaborations difficult to scale globally, especially regarding data ownership, intellectual property (IP), and compliance with local data regulations such as Australia’s Privacy Act 1988.

These concerns lead to fewer institutions contributing data, hindering the ML model from learning from diverse and augmented data sets from different institutions and geographical locations. The result is inaccurate and biased data insights. This is why we need federated learning.

What is federated learning and how does it work?

The main idea behind federated learning is to train a ML model on user data without needing to transfer that data to a single location. This involves moving the training computations to the infrastructure at the data-owning institution, instead of moving the data to another location for training. A central server is then responsible for aggregating the insights that result from the training computations of multiple data owners.

Federated learning has training iterations performed on local devices, which brings the main benefit of not compromising or exposing the original data when data is in flight. Data remains with the owner while still being utilised to create global insights.

Federated learning starts with a trusted foundation

With so much riding on data, it is imperative that organisations have a robust data security strategy in place. Key to this is to keep sensitive data in the cloud inside an access-restricted enclave, commonly known as a Trusted Execution Environment (TEE). Privacy protections like these are critical to providing continuous protection of workloads with regulatory requirements or other sensitive data in distributed networks.

As computing moves to span multiple environments – from on-prem to public cloud to edge, organisations need protection controls that help safeguard sensitive IP and workload data wherever the data resides, as well as to ensure that remote workloads are executing with the intended code. This is where confidential computing comes in. Unlike traditional encryption for data at rest or in transit, confidential computing relies on a TEE for enhanced protection and privacy of the code to be executed and the data in use.

Confidential computing means datasets can be processed much more securely, and the risk of attacks can be reduced by isolating code and data from outside incursions. As the most researched and deployed confidential computing technology in the data center today, Intel® Software Guard Extensions (Intel® SGX) offers a hardware-based security solution that helps protect data in use via a unique application-isolation technology.

With a hardware-based security foundation, previously vulnerable attack surfaces can be strengthened to not only protect against software attacks, but also help eliminate threats against data in-use. Organizations can therefore have a peace of mind that their machine learning model can safely use different datasets, and train algorithms with them while remaining compliant with regulations and security.

Future of federated learning

By enabling ML models to gain knowledge from ample and diverse data that would otherwise be unavailable, federated learning has the potential to bring significant breakthroughs in healthcare, improve diagnosis, and better address health disparities.

While we are still at the beginning of exploring federated learning, it holds great promise by bringing organisations closer together to collaborate and solve challenging problems, while mitigating data privacy and security issues. In fact, federated learning can stretch its application beyond healthcare, with great possibilities in areas such as Internet of Things, fintech, and much more.

The future of federated learning will bring AI applications to the next level, and we are just scratching the surface of its true potential.

Read 493 times

GARTNER MARKET GUIDE FOR NDR 2022

DOWNLOAD NOW!

PROMOTE YOUR WEBINAR ON ITWIRE

MORE INFO HERE!