HPE Blog, UK & Ireland

How healthcare could harness swarm learning

In the era of digital transformation, technologies such as artificial intelligence and machine learning are uncovering insight and value from data like never before. These technologies are solving some of the world’s most pressing challenges by driving fundamental changes in sectors such as healthcare, agriculture, transportation and many others.

As the global economy becomes increasingly connected; as does the need to maximise data by developing algorithms, models and systems in which deep insight can be extracted and analysed for the greater good.

Machine Learning (ML) enables this. ML is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. ML data is, traditionally, aggregated into a central location, typically a public or private cloud, where statistical or ML models are trained using the data.

However, all intelligent data-driven models face the same challenges – data sovereignty, security and privacy – which can cause barriers to sharing the data required to train ML, especially in healthcare. Not to mention the costs a central infrastructure could incur to host and process the aggregated data.

Hospital data includes our most private data, yet it’s arguably the data we can learn the most from. Which, in turn, could pave the way for life-changing advances in patient and medical care.

This is just one of the reasons why HPE set out to develop swarm learning.

What is swarm learning?

Perhaps you’ve driven by a field and seen a flock of tens of thousands of starlings pulsing and whipping through the air, a relatively common sight, especially in Europe. Perhaps you’ve been swimming in the ocean and have seen a “fish ball” doing basically the same thing in the water. Different animals use a type of distributed behaviour, often defensive, that cannot be traced to the movements of a “leader.”

This behaviour, called scale-free correlation, murmuration, or swarming, may provide an answer to what to do with the geometric increase in data collected at the edge.

Today, we create artificial intelligence models by sending data collected at the edge to a central point. The data is used to “train” the model, and then the model is pushed back out to all devices at the edge.

Yet, it’s often impractical to send all of the data collected at the edge to a central server for computing. Latency, heat and energy demands, compliance issues, and mounting transportation and opportunity costs are to blame.

Eventually, the impractical will become the impossible. But researchers, inspired by nature, have an answer.

Dr. Eng Lim Goh, Senior Vice President and Chief Technology Officer for AI at HPE acknowledges the benefits of hospitals using swarm learning.

“HPE swarm learning allows each hospital to continue to learn locally. But at each cycle we collect the learned weights of the neural networks, average them, and send it back down to all of the hospitals. After a few cycles of doing this, the hospitals would have learned from each other, removing biases without having to share any private patient data. That’s the key. So, there’s the ability to learn from everybody without having to share your private patients data. That’s swarm learning.”

This unified sharing of data enables more efficient research and analysis of life-threatening diseases through a collaborative approach that removes bias and data capacity limitations. 

The power of sharing

It is impossible for healthcare professionals to prepare for and specialise in every disease, diagnosis and treatment. Every day doctors are faced with new challenges; continually learning to provide the right care and treatment for their patients.

But what if every doctor could access the experience and learnings of every other doctor, across the globe?

Patients will travel hundreds of miles to visit a consultant that specialises in their condition. This specialism has been fostered through seeing, identifying and treating the specific condition. To date, this consultant’s specialist ‘wisdom’ has been confined to his brain, and perhaps those close enough to observe.

The swarm learning concept enables healthcare professionals to impart their individual knowledge, experiences and, most importantly, the intuition behind the data. The unification of this ‘wisdom’ is priceless and something that simply cannot be depicted through medical journals and data.

The AI algorithms involved in swarm learning can be trained in different disciplines meaning life-changing learnings can be shared with the network. As more hospitals join the swarm, the power and knowledge the swarm can impart simply evolves.

Swarm learning in a COVID world

 The healthcare sector’s response to the coronavirus pandemic has been greatly influenced by the rise in technologies such as AI. Joachim Schultze, Professor for Genomics & Immunoregulation at the German Center for Neurodegenerative Diseases (DZNE), and his team, have applied swarm learning to learn more about the disease and how it impacts those with underlying illnesses.

The technology has proved to improve the diagnosis of severe diseases in patients, which in-turn has enabled a more accurate treatment response for those within that subset, that have tested positive for COVID-19.

Swarm learning offers the opportunity to gain deeper, impartial insight that local machine learning simply cannot provide. To learn more watch HPE’s recorded event, ‘What is swarm learning? AI, Blockchain and IoT working together to uncover real-time intelligence’.

Jennifer Reilly
Hewlett Packard Enterprise



0 Kudos
About the Author


Jennifer shares insights and technology trends that are impacting the healthcare and life sciences industries.

HPE Webinars
Find out about the latest live broadcasts and on-demand webinars
Read more
Online Expert Days
Visit this forum and get the schedules for online Expert Days where you can talk to HPE product experts, R&D and support team members and get answers...
Read more
View all