'It is the saddest thing': ECG tests being misinterpreted with tragic consequences

Sly Saint

Senior Member (Voting Rights)
ITV Regional News has been told a “concerning” number of patients have died from heart conditions after their test results were misinterpreted by medical professionals.

Clarissa Nicholls, a student at the University of Cambridge, collapsed and died while hiking with a friend in France last year.

She was a keen athlete with a passion for running and hockey.

Her mother, Hilary Nicholls, told ITV News that her daughter was "incredibly sporty", "hugely dynamic" and "had so much energy".

"She was the model of someone who was fit and healthy but she had no idea that there was something wrong.”

The only clue had been the year before her death when she collapsed while playing sport, Nicholls said of her daughter.

She had an ECG to check her heart but the test results were not looked at by a specialist.

This meant a rare heart condition was not picked up and Clarissa did not realise that exercising was actually putting her life at risk.

She died from a cardiac arrest just days before her 21st birthday.
ITV News approached NHS England who said training for clinical professionals on how to interpret ECGs doesn’t fall within their remit.

In a statement, the Department for Health and Social Care said: “Patient safety is of utmost importance and we expect guidance on assessing electrocardiograms to be followed at all times and for all clinicians reviewing ECGs to have received the appropriate training.

"NHS trusts are responsible for ensuring their staff follow the latest clinical guidance."

It is hoped the interpretation of ECGs could be improved with the development of artificial intelligence. This is currently being looked into at the School of Medical Sciences at City St George’s, University of London.

'It is the saddest thing': ECG tests being misinterpreted with tragic consequences | ITV News
 
It is hoped the interpretation of ECGs could be improved with the development of artificial intelligence
All hail our future AI overlords. Because it truly can't be worse than what we have right now. We will not survive ourselves otherwise.
 
All hail our future AI overlords. Because it truly can't be worse than what we have right now. We will not survive ourselves otherwise.
The thing is that now is the time the literature needs to be filled with more papers about methodology then so the ‘learning’ the AI does at least has those heuristics as part of them underlining how poor most of certain research is

in I’m imagining an ‘if x,y, z missing then not science/valid’ type way

it’s too late on the content itself - they’ve already filibustered real scientific psychology out of search engines with sheer numbers of crap papers you can churn out when you don’t have the time and resource issues of recruiting right or using blinded controls and anything other than surveys

so there needs to be a way those method issues get attached into the trawls any AI learners on a topic will be doing ?
 
NHS to use AI death calculator to tell patients how long they have
Hundreds of Brits going into hospital could soon have the rough date they will die estimated by an AI death calculator.

Using the results of a single electrocardiogram (ECG) test — which takes minutes and records the electrical activity of the heart — it is able to detect hidden health issues that doctors might not be able to spot.

The programme, called AI-ECG risk estimation or AIRE, has proven in studies to correctly identify risk of death in the 10 years after the ECG, with up to 78 per cent accuracy.

The tech will be trialled at two London NHS trusts from the middle of next year, but experts hope it will be used across the health service within five years.

Dr Arunashis Sau, a cardiology registrar at Imperial College Healthcare NHS Trust — one of the two trusts involved — said the aim of AIRE was not to develop something to replace doctors, but to create something 'superhuman'.

The tech 'reads' the ECG results to uncover patterns in the electrical signals and analyse genetic information from the heart's structure to detect issues such as heart rhythm problems and heart failure before they fully develop.

It then gives a prediction figure, measured in years, of a patient's risk level.
NHS to use AI death calculator to tell patients how long they have
 
Back
Top Bottom