Improving epilepsy diagnosis with a wearable device
Using machine-learning algorithms, Seer Medical’s diagnostic system analyses electroencephalography (EEG) data to detect epileptic seizures – highlighted here in pink.
A portable system for diagnosing epilepsy is now available across Australia and will soon be introduced in the UK and Germany. It was developed by Seer Medical, which has raised $A20 million in investment.
- A portable system for improved epilepsy diagnosis is now available across Australia and will soon be introduced in the UK and Germany.
- Methods for diagnosing epilepsy are currently inadequate, even though the disorder affects around 50 million people worldwide.
- The system was developed by Seer Medical, a company co-founded by University of Melbourne researchers.
More than 2700 people across Australia have used an at-home system for improved epilepsy diagnosis. The system, which combines a wearable device and cloud-computing technology, will soon be introduced to the UK and Germany.
The system was developed by Seer Medical, a company founded in 2017 by researchers at the University of Melbourne.
Seer Medical has raised around $A20 million from investors, including medical device company Cochlear Ltd and the Epilepsy Foundation.
Seer Medical has more than 100 employees.
Methods for diagnosing epilepsy are currently inadequate, even though the disorder affects around 50 million people worldwide.
Epileptic seizures can be infrequent and unexpected. Sometimes, people with epilepsy don’t remember that they’ve had a seizure. And heart problems, narcolepsy or migraines can cause similar symptoms and be confused for epilepsy.
For an accurate diagnosis, a person’s brain activity, heart activity and behaviour need to be monitored continuously over several days. However, monitoring patients for long durations in hospital is expensive, and the equipment is not available in most hospitals. In-hospital monitoring is also inconvenient for patients. And it is not available in regional and rural areas.
The technology behind Seer Medical’s portable diagnostic system is based on clinical observations and research led by University of Melbourne researchers. Professor Mark Cook is Director of the University’s Graeme Clark Institute for Biomedical Engineering and Chair of Medicine at St Vincent’s Hospital Melbourne. Dr Dean Freestone is an electronic engineer in the University’s Department of Medicine at St Vincent’s Hospital Melbourne.
Because of the nature of epileptic seizures, the development and testing of machine-learning algorithms for seizure detection require data from the same patients collected over years.
The research team obtained insights from data acquired during a 2010 clinical trial that they conducted in partnership with Neurovista, a medical device company based in Seattle, USA. In the trial, 15 people were implanted with a seizure-detection device and monitored for up to three years. The team’s analysis of the clinical trial data revealed seizure cycles – an increased likelihood that a person would experience a seizure at a particular time of day, week or month. It also showed that seizure duration and frequency are highly predictable within individuals with epilepsy but vary between them.
The team confirmed these findings using data obtained from Seizure Tracker, a website and app that has collected information on epileptic seizures from more than 10 000 people for up to 10 years.
Based on this work, the team developed machine-learning algorithms using data acquired from scalp monitoring studies performed in hospital patients over many years of clinical practice.
In 2015, Professor Cook won Epilepsy Foundation of America’s Shark Tank Competition, receiving $US125 000 to pursue methods for epilepsy diagnosis and management.
Technology development history
In 2017, Professor Cook and Dr Freestone co-founded Seer Medical with colleagues from the University of Melbourne and St Vincent’s Hospital Melbourne. Professor Cook is Chief Medical Officer and Dr Freestone is Chief Executive Office of the company. The company has attracted investment of around $A20 million from Cochlear Ltd, the Epilepsy Foundation and private investors.
Seer Medical developed a portable diagnostic system that can be used at home to monitor brain activity and behaviour for 3–10 days. The system uses electroencephalography (EEG) to measure brain activity via electrodes on the scalp. It also uses electrocardiography (ECG) to measure heart activity via electrodes on the chest. A wearable harness holds the data recorder and a battery pack.
A portable video camera is also supplied, so users can film their own behaviour wherever they go. Users are connected to and disconnected from the system in one of Seer Medical’s 22 clinics around Australia. This service is paid for by Australia’s public healthcare system, Medicare.
Seer Medical's wearable device enables brain activity to be monitored continuously at home rather than at a hospital.
During monitoring, data is uploaded wirelessly to the Seer Cloud, the company’s secure cloud storage. The data is analysed using Seer Medical’s machine-learning algorithms, which it continues to develop. The user can then share this information with their doctor.
The company also developed the Seer App, which allows anyone with epilepsy to record their own seizures, schedule medication, and share this information with their healthcare provider.
In 2018, Seer Medical received $US3 million from Epilepsy Foundation of America to develop methods for managing data from wearable technology. The company is collaborating with the Mayo Clinic in the USA and King’s College Hospital in the UK on this project.
Seer Medical is in discussions to share its cloud-computing technology with Epi-Minder, a Melbourne company that is developing an implantable device for seizure detection. Epi-Minder was also co-founded by Professor Cook.
Seer Medical has more than 100 employees in Australia, the UK and Germany.
Karoly PJ et al (2020) Forecasting cycles of seizure likelihood. Epilepsia 61(4): 776–786. doi: 10.1111/epi.16485
Clarke S et al (2019) Computer-assisted EEG diagnostic review for idiopathic generalized epilepsy. Epilepsy & Behaviour, doi: 10.1016/j.yebeh.2019.106556
Cook MJ et al (2013) Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. Lancet Neurology 12: 563–571. doi: 10.1016/S1474-4422(13)70075-9
AU2020901338, filed 28 April 2020
AU2020202572, filed 16 April 2020
US10506988B2, filed 11 February 2019
WO2019100102A1, filed 26 November 2018
AU2018305723, filed 8 June 2018