Focus areas
Four connected areas of work, from the materials that touch the skin to the models that read what they measure.
Next-generation sensors
Next-generation sensors are soft, thin devices that stretch, bend and twist with the body while measuring it continuously. The centre makes them from wrinkled metal films and stretchable inks, quickly and at low cost, without a clean room.
- Strain sensors that track breathing rate and volume
- Soft pressure sensors for beat-to-beat blood pressure
- Electronic-skin networks and integrated photonic sensors
Breathing and blood pressure are usually checked in a clinic, as a single snapshot. Soft strain sensors worn on the ribs and abdomen have tracked breathing rate and volume continuously and sent the readings to a phone, which could give people with asthma or cystic fibrosis an early warning of a flare instead of a lung test every few months. Wrinkled gold pressure sensors at the wrist have measured beat-to-beat blood pressure with results comparable to an FDA-approved monitor, pointing to continuous blood pressure without an inflatable cuff. Because they are made from shrink film rather than in a clean room, the sensors can be cheap enough to throw away.
- Respiration rate and volume measurements using wearable strain sensorsnpj Digital Medicine, 2019
- Soft wearable pressure sensors for beat-to-beat blood pressure monitoringAdvanced Healthcare Materials, 2019
What makes a sensor next-generation?
It moves with the body instead of fighting it. Matching the softness of skin gives cleaner signals and lets people wear the sensor for longer.
Why wrinkled metal?
A metal film shrunk into tiny wrinkles can stretch like an accordion without cracking, so its resistance changes smoothly as it is pulled.
Wireless and battery-free
Battery-free sensors take their power from a reader’s field and send their data back the same way, using near-field communication. With no battery to replace, they can be worn for longer, sealed against water or placed under the skin.
- Implantable pressure sensors read through the skin
- Body-area networks that keep working underwater
- Waterproof NFC and stretchable circuit boards
A battery is often the biggest, heaviest and shortest-lived part of a wearable. Take it away and a sensor can be thinner, sealed against water and left in place for much longer. In laboratory tests, a thin copper-coil sensor designed to sit under the skin measured fluid pressure and was read through 12.7 mm of skin, with no battery at all. Rising fluid pressure is an early sign that heart failure is getting worse, so a sensor like this could give clinicians time to act before a patient is readmitted to hospital. On the skin, stretchy patches that carry power and data along the body keep battery-free sensors working from dry land to underwater, where Bluetooth and wi-fi fail.
- Wireless and battery-free sensor for interstitial fluid pressure monitoringSensors, 2024
- Amphibious epidermal area networks for uninterrupted wireless data and power transferNature Communications, 2023
How can a sensor work without a battery?
A nearby reader, such as a phone, creates a magnetic field. The sensor’s coil turns that field into just enough power to take a reading and send it back.
Does it work in water?
Yes. The centre’s epidermal networks use magnetic coupling, which passes through water and skin where Bluetooth drops out.
Integrated therapeutics
Integrated therapeutics close the loop between sensing and treatment: a device detects a change in the body and responds to it. Target applications include apnoea of prematurity, wound healing, incontinence and drug delivery.
- Aptamer sensors that detect specific molecules
- Electrodes integrated with microfluidics
- Closed-loop devices that sense and respond
Most care is two separate steps: a test finds a problem, then someone decides on a treatment. Closing that loop means a device that detects a change in the body and responds to it, for problems such as apnoea in premature babies, slow-healing wounds, incontinence and drug delivery. It starts with sensors that can recognise one specific molecule. The team has built an aptamer sensor on a low-cost wrinkled gold electrode that tells cerebrospinal fluid apart from blood and mucus in about 20 minutes, without added reagents. That is a step towards a quick bedside test for spinal fluid leaks, which current methods detect slowly and inconsistently.
- Electrochemical aptamer-based sensor for cerebrospinal fluid detectionAnalytical Methods, 2025
What does closed-loop mean?
The device measures, decides and acts on its own, such as releasing a drug when a biomarker crosses a threshold, rather than waiting for a person to read the data.
What is an aptamer?
A short strand of DNA that folds to grip one specific molecule. When the target binds, the electrical signal from the electrode changes.
Predictive analytics
Predictive analytics turns continuous body signals into early warnings, using machine learning to see what is likely to happen next. Current work predicts sleep stages from pulse and breathing and tracks beat-to-beat blood pressure.
- Sleep stage and sleep apnoea classification
- Beat-to-beat blood pressure variability
- AI-enabled wearable monitoring with AIML
A wearable that records around the clock produces far more data than any clinician can read. Machine learning turns those signals into something useful: a trend, an early warning, a forecast. The team has used deep learning to predict sleep stages from pulse and breathing signals, the kind a wearable can record at home, which could make sleep checks possible outside a sleep laboratory. In people who had a stroke more than a year earlier, beat-to-beat blood pressure varied more than in healthy people, suggesting continuous blood pressure monitoring could help guide long-term care and medication to lower the risk of another stroke.
- Improved sleep stage predictions by deep learning of photoplethysmogram and respiration patternsComputers in Biology and Medicine, 2024
- Changes in beat-to-beat blood pressure and pulse rate variability following strokeScientific Reports, 2023
What can be predicted from a wearable?
Signals like pulse, breathing and blood pressure carry patterns that come before events. Models trained on them can flag changes earlier than a single reading can.
Who does the modelling?
The centre works with the Australian Institute for Machine Learning (AIML) at Adelaide University.
Build the next generation with us
Five fully funded PhD positions for electrical and biomedical engineers. Or bring us an industry or clinical challenge.