As a student, the neurologist Greg Scott had very little interest in following a career in medicine. Computing was the focus of his studies.
Then one day he suffered a grand mal seizure. Also called a tonic-clonic seizure, it causes a loss of consciousness and violent muscle contractions.
"It came out of the blue," he told the Observer last week. "I was taken to hospital where a scan revealed that I had a brain tumour. It was then removed during an operation during part of which I was completely conscious," said the researcher at Imperial College London.
"I was later able to watch the operation during a case presentation. So I can say that I have seen my own brain. I was 19 and the experience transformed the trajectory of my career."
Scott was inspired to study medicine and has put his computational background to good use in his research. "I became fascinated with brain activity and the idea of using AI and other technologies to understand how it generates consciousness and cognition."
In particular, Scott is concerned with delirium, a common but often misdiagnosed condition that is frequently confused with dementia.
His research, which is funded by the National Institute for Health and Care Research, aims to develop an easy-to-wear device that could rapidly detect if a person is suffering from this debilitating condition.
"Delirium is a massive problem," said Scott, who is based at Imperial's UK Dementia Research Institute. "Around 20% of adults in hospitals - around 20,000 NHS patients - are affected by delirium which can be caused by all sorts of different things: the impact of an operation, a urinary tract infection, a chest infection, the side-effects of medication or drug or alcohol misuse."
The result is confused thinking and a lack of awareness of one's surroundings, symptoms that can often have a swift onset - sometimes within a few hours. However, to diagnose the condition, clinicians must make subjective decisions.
"We desperately need to find a simple, straightforward objective test that can tell doctors unambiguously within minutes if a person is suffering from delirium - for that could direct them to look for the underlying cause of their condition and allow it to be treated more quickly and effectively," added Scott.
The key to this work lies with the use of electroencephalograms (EEGs) to record the brain's electrical activity. Brainwaves were first measured a century ago by the German psychiatrist Hans Berger whose work was received with incredulity and derision when first published. The use of EEGs has since become standard in diagnosing epilepsy and other conditions.
However, it can take a long time to fit the 20 or so electrodes to a patient's head and more time for another clinician to interpret the recording. "The whole process is very time consuming, very resource intensive, and poorly available in the NHS. We need to find ways to change that."
This is the goal of Scott's research which is focused on narrowing down the mass of signals that are produced by EEGs to pinpoint only those that are linked to delirium. "That is going to give us a handle on the quick diagnosis of the condition," he said.
By specifying precisely which brainwaves are being triggered by delirium, it will be possible to develop a simple device that can be easily slipped over a patient's head.
"It won't take hours to put on and within minutes it will give a readout that will tell doctors if their patient has delirium and will also reveal its severity. Ideally it will also differentiate those with dementia and those with delirium and allow the right treatment to be used from the start. That is the goal we are now homing in on."