Monitoring epileptic seizures with an implantable device

The outcome

An implantable device to detect and eventually predict epileptic seizures is undergoing a Phase I clinical trial in Australia. The trial will test the safety of the device in a small number of people. It will also show whether the device can successfully record brain activity and transmit that data to a remote database.

“The Minder system captures EEG data 24/7, revealing previously unseen brain activity,” says biomedical engineer Professor Mark Cook.

There are five components to the Minder® system:

1. A smartphone app collects data and uploads it to the cloud for processing
2. A wearable processor powers the device and connects to a smartphone by Bluetooth
3. A magnetic coil sits on the scalp, connecting the processor wirelessly to the implant
4. A minimally invasive implant captures EEG, placed between the skin and the skull
5. A sub-scalp electrode senses and records neural events.

The Minder device received Breakthrough Device designation from the US Food and Drug Administration (FDA) in April 2023. It’s granted to medical devices that could provide a more effective treatment or diagnosis of life-threatening or irreversibly debilitating diseases or conditions.

The device is designed and manufactured by Epiminder, a company co-founded in 2018 by University of Melbourne researchers. Investors in Epiminder include medical device company Cochlear Ltd, the Bionics Institute, St Vincent’s Hospital Melbourne and the University of Melbourne.

The need

An estimated 50 million people worldwide live with epilepsy.

Epileptic seizures can cause temporary loss of awareness. They can affect movement, sensation and mood.

“The unpredictability of seizures means that people with epilepsy may be unable to work or drive,” says Professor Cook.

A means to reliably predict seizures would make it easier for people with epilepsy to carry out everyday activities. It would also benefit their safety, mental health and employability.

More accurate seizure monitoring can also help identify which medications work and which don’t, helping people with epilepsy become seizure free while dramatically lowering the side effects of medication sooner.

The research

The technology is based on research in neuroscience and electrical engineering from a team led by the University of Melbourne’s Professor Mark Cook. He is also a clinician at St Vincent’s Hospital Melbourne, where he is the Sir John Eccles Chair of Medicine and Director of Clinical Neurosciences.

”We found evidence of seizure cycles – an increased likelihood of experiencing a seizure at a particular time of day, week or month,” says Professor Cook.

In the study, 86 per cent of people with epilepsy had seizures that followed at least one significant cycle. For example, 83 per cent of people had a daily (circadian) cycle, while 23 per cent had a weekly cycle.

The research team also found that seizure duration and frequency varied considerably between people with epilepsy but was highly predictable within individuals.

Technology development history

Professor Cook’s research team had been studying seizure prediction since 2005. In 2010, they partnered with Neurovista, a medical device company based in Seattle, US. They designed a seizure advisory system that was trialled in 15 people. A device implanted under the skull recorded brain activity using electroencephalography (EEG). A second device, which was implanted in the chest, transmitted the EEG data to external equipment. An algorithm predicted seizures for each person based on their individual brain activity.

“The trial showed that it was possible to record continuous EEG and transmit and analyse this data. But a major limitation was the inability to analyse large amounts of data fast enough to provide reliable seizure forecasts,” says Professor Cook.

To overcome this, Professor Cook’s research team adopted machine-learning algorithms they had developed for epilepsy diagnosis. These algorithms incorporate the team’s findings on seizure cycles and duration.

The team also investigated the best method for long-term implantation of EEG electrodes under the scalp – a much less invasive way of collecting the EEG data than the Neurovista device.

To further develop and commercialise the seizure detection device, Professor Cook co-founded Epiminder in June 2018 with Associate Professor Chris Williams, who led the preclinical research and Minder device development at the Bionics Institute.

Epiminder raised around $A10 million in initial investment. Investors include Cochlear Ltd, the Bionics Institute, St Vincent’s Hospital Melbourne, the University of Melbourne and private investors. Cochlear invested $A2 million and at the initial stage contributed to the design and development of the device to the value of $A1.65 million.

Epiminder has an exclusive license to intellectual property owned by the Bionics Institute, St Vincent’s Hospital Melbourne and the University of Melbourne.

In 2019, Epiminder began a clinical trial with its implantable device. Data will be collected from each person involved in the trial, and transferred wirelessly from the device to their mobile phone. Data will then be transmitted to cloud storage managed by Epiminder.

The trial will test the safety of implanting the device and its ability to record brain activity over extended periods and transmit the data.

In late 2020, Epiminder raised $A18 million in Series A funding from private investors and existing shareholders and an additional $16 million in late 2021.

Partners

Cochlear Ltd

The Bionics Institute

St Vincent’s Hospital Melbourne

Funding support

ARC Linkage Project (LP0560684)
NHMRC Project Grant (APP1048360)
NHMRC Project Grant (APP1065638)
NHMRC Project Grant (APP1075347)

Publications

Karoly PJ et al (2017) The circadian profile of epilepsy improves seizure forecasting. Brain 140(8): 2169–2182. doi: 10.1093/brain/awx173

Benovitski YB et al (2017) Ring and peg electrodes for minimally invasive and long-term sub-scalp EEG recordings. Epilepsy Research 135: 29–37. doi: 10.1016/j.eplepsyres.2017.06.003

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

People

Professor Mark Cook

Associate Professor Chris Williams