AWS & AEP partner to build world's largest epilepsy dataset

Amazon Web Services (AWS) is collaborating with the Australian Epilepsy Project (AEP) to develop the world's largest connected dataset for epilepsy, leveraging artificial intelligence (AI) and cloud technology.

This partnership aims to aid neurologists, researchers, and clinicians in better comprehending the disorder and improving patient management and treatment methods.

Epilepsy affects approximately 250,000 Australians and poses significant challenges in terms of diagnosis and treatment. The condition costs Australia up to AUD $12.3 billion annually. The journey from diagnosis to effective management is frequently hampered by the complexities associated with analysing brain scans, genetic information, cognitive tests, and extensive medical histories. AWS and AEP's partnership seeks to address these issues.

One of the key benefits of this initiative is the acceleration of the diagnostic and treatment process. Kylie Staats, an individual living with epilepsy, shared that although epilepsy has not deterred her from adventurous pursuits such as skydiving and bungee jumping, she dreams of eventually being able to drive a car. "Cloud technology is providing hope," Staats noted, reflecting a sentiment likely shared by many in the epilepsy community.

The AEP was awarded AUD $30 million in funding through the Medical Research Future Fund (MRFF) in 2021. This funding, the largest government investment in epilepsy research in Australia, supports the design and delivery of a brain-health platform intended to revolutionise epilepsy care by tackling issues of access, uncertainty, and cost. By incorporating AWS cloud technology, the AEP can efficiently handle vast amounts of patient data, thereby reducing the diagnostic and treatment timeline potentially by up to 15 years.

The project has analysed data from more than 600 participants and 300 control volunteers as of July 2024, with an ultimate goal of reaching 4,000 individuals by 2026. This data is securely stored within Australia, and only de-identified information is accessible to researchers and collaborators. The security measures include encryption of data during storage and transmission, as well as multifactor authentication for systems accessing participant data.

Machine learning (ML) models play a crucial role in prioritising patients at risk. By examining MRI scans, genetic tests, medical histories, and participant surveys, AWS technology assists researchers in drawing rapid correlations and identifying anomalies. These insights help refine patient risk predictions and recommended treatments, ultimately offering personalised care pathways.

"The Australian Epilepsy Project demonstrates how healthcare organisations can leverage the secure, scalable computing resources of AWS to gain insights from complex datasets," remarked Simon Elisha, AWS's chief technologist for the public sector in Australia and New Zealand. "With advanced cloud technologies, the AEP can efficiently process patient data and apply AI models to surface relevant clinical information and automate tasks. This enables neurologists to provide more timely and personalised care for those living with epilepsy."

Anton de Weger, AEP's Digital and Technology Lead, commented on the project's significance: "With AWS, we are building a world-leading epilepsy dataset, but the real impact comes from the insights we gain. These insights enhance our understanding of epilepsy and brain function so we can help clinicians improve treatments."

The AEP has employed various AWS services to facilitate this project. For instance, Amazon Textract extracts text from diverse medical document formats, making it easier to search and analyse records. A comprehensive MRI viewer hosted via Amazon EC2 allows neurologists to access high-resolution MRIs from across Australia seamlessly. Amazon QuickSight provides graphical dashboards to display project data, while Amazon Elastic Kubernetes Service (Amazon EKS) enables efficient running of AI and ML workloads.

The project's generative AI capabilities further simplify and accelerate treatment processes. Future plans include developing a generative AI chatbot for patients seeking general advice on epilepsy and treatment options or the latest insights from scientific literature.

This collaborative effort between AWS and AEP exemplifies how advanced technology can be harnessed to address complex health challenges, ultimately aiming to enhance the quality of life for individuals living with epilepsy in Australia.