Data

Summary

At the MARCS Institute we are disrupting big data. We help organisations switch to a ‘point of event’ approach - where only relevant pieces of data are recorded and used – enabling them to eliminate the waste that comes from capturing and analysing ‘big data ’.

Using traditional and neuromorphic sensor technology, artificial intelligence (AI) and machine learning (ML), we are shaping how sectors such as manufacturing, health, and defence capture, analyse and visualise data to improve decision making.

And our approach is having a profound impact.

We’re working with partners to; improve wearable and implantable biometric technologies for defence personnel, patients, and elite athletes; enhance medical imaging to improve diagnostics; enable autonomous systems to operate and maintain communication in complex and changing environments such as in mines and combat zones, and; deploy sensors that can do everything from tracking space debris to detecting koalas by their call, to monitoring “at risk” patients to record when they fall while still protecting their privacy.

Our capability in data capture and analysis is enabling us to:

• Provide evidence-based improvements to ensure data captured in high-stakes environments for augmented reality (AR) and virtual reality (VR) systems is being presented in a way that enhances human performance and optimises decision making.

• Develop new sensors for forecasting and modelling across multiple settings spanning medicine, environment, and physical structures.

• Build models to interpret sensor data for better diagnostic outcomes.

• Enhance imaging, and data capture and analysis in hospital settings to make the patient triage system more efficient.

• Improve decision making by incorporating greater explainability into data processing systems across medical, defence or industrial applications.

• Build systems to utilise emerging technology spanning AR, VR, and AI to understand and improve mental health and education across all ages.

Our unique approach

We develop neuromorphic systems – inspired by how the human brain functions - to shift how we capture, analyse and display data so that it is optimised for human use.

Instead of just developing powerful algorithms that review all the available data, we go to the heart of the problem – reducing the amount of data captured in the first place. Whether it’s audio, visual, or biometric, we can develop new sensors, or modify existing ones to capture only the data that matters to our partners.

In addition to traditional AI and ML approaches we have created ‘spiking neural networks’ to analyse and interpret data. These networks activate and process data when needed and slow down when it is not, improving speed and efficiency of data translation.

We are applying our deep understanding of the human brain and how it is impacted by factors such as stress, fatigue, and information overload to create platforms that integrate and display data so it’s optimised for the human brain to process.

Our fields of interest

• Neuro imaging and brain stimulation: To assess cognitive function including perception, action and, decision making.

• Human interaction with stress: To understand how humans respond to challenging, stressful environments.

• Neuromorphic audio and visual sensors: Creating a new generation of sensors and adapting existing sensors to selective capture information at the time the event happens.

• Neuromorphic audio and visual processing: Creating more efficient ‘neural spike’ algorithms for processing data.

Impact built on collaboration

Our partnerships range from small scale to large-long term projects with hospitals, manufacturing, government and international industry. Some examples include:

• Developing neuromorphic sensors in partnership with defence teams to monitor satellites and space junk.

• Partnering with Thales to create ‘The Astrosite’ - a custom-designed mobile telescope observatory built specifically to track and detect objects in space.

• Working with the United States Air Force Academy to deploy neuromorphic cameras on the International Space Station to monitor high altitude sprites and keep an eye on space debris.

• Collaborating with Liverpool Hospital to analyse electrocardiogram (ECG) and Photoplethysmography (PPG) data for improved diagnosis .

Our research in action

Across MARCS, we are engaged with hundreds of research projects at any given time. MARCS research projects aim to have a profound impact in their particular field and often involve collaboration with local and international researchers. You can explore our current projects in Data here.