News

Robots better than human medics at diagnosing women’s heart attacks



Robots may be better than medics at diagnosing heart attacks in women, research suggests.

A study of 13,000 patients found artificial intelligence could eradicate a “gender bias” that means female cases are often missed.

Previous studies have found that women suffering heart attacks are 50 per cent more likely than men to receive a wrong initial diagnosis – leading to thousands of needless deaths.

Heart attacks in women missed

Experts say heart attacks in women are often missed, partly because medics still assume victims are likely to be male, and partly because symptoms can be less obvious.

The study by Edinburgh University found that the algorithm was developed from data on patients who went to hospital with a suspected heart attack.

It was able to accurately rule out a heart attack 99.5 per cent of the time – meaning patients could be safely sent home.

It was also able to confirm whether a patient required further hospital tests in 83.7 per cent of cases.

This compares to accuracy rates of just 49.4 per cent with standard checks.

App for instant diagnosis

Scientists intend to develop an app to help doctors make an instant accurate diagnosis, simply by inputting blood test results and patient data.

Edinburgh University researchers developed the app using data from 10,038 NHS patients – 48 per cent of whom were women – who went to hospital with a suspected heart attack.

It was then validated on a further 3,035 people outside of the UK.

Prof James Leiper, associate medical director at the British Heart Foundation, said: “This is a huge step forward to help ensure everyone is on a level playing field when it comes to heart attack diagnosis and treatment.

“We know that women are more likely to receive a misdiagnosis, but by harnessing the power of AI this team could help make that an issue of the past.”

The findings were presented at the annual conference of the European Society of Cardiology in Barcelona. 

Current approach ‘imperfect’

Prof Nick Mills of the BHF Centre for Cardiovascular Science at the University of Edinburgh, said: “Each year in the UK there are half-a-million hospital visits from people experiencing chest pain, with more than 200,000 visits due to heart attacks.

“Our current approach is imperfect with one in five people re-attending at 30 days and one in 20 having a heart attack or dying from cardiovascular disease at one year.

“Our mission is to fast track improvements in heart-attack diagnosis, and we hope our CoDE-ACS app will be rolled out in emergency departments across the UK to deliver more personalised care and better outcomes.”

Heart-attack patients can receive clot-dissolving drugs, balloon angioplasty treatment or surgery to restore blood flow.

Measuring troponin in the blood is the current gold standard for diagnosing a heart attack.

But results can be very different for men and women, making diagnosis more difficult.

The algorithm was able to evaluate patients’ risk more precisely so that 29.5 per cent more people had a clear treatment plan.

‘Gender disparity’ in heart care

Dr Allyah Abbas-Hanif, of Imperial College London and co-chair, Paediatrics and Women’s Health Expert Group in the Faculty of Pharmaceutical Medicine, said: “The gender disparity in the assessment of a heart attack has led to worrying misdiagnosis and delays for women, tragically leading to thousands of women unnecessarily dying.

“The reasons are complex but include women themselves as well as doctors having a low suspicion for heart attack. The reduced blood flow to the heart that causes a heart attack needs rapid reversal.

“We have lifesaving treatments for heart disease, but if you don’t get the diagnosis right, or it’s late, you miss the crucial window of opportunity.”

Dimitrios Doudesis, lead researcher at the BHF Centre at Edinburgh University, said: “While the troponin test takes 30 minutes to process, we take an array of other health information and add it into the app alongside the troponin measurement.

“This provides doctors with a precise and instantaneous score to confirm if they can reassure their patient that they haven’t had a heart attack and send them home, or if they require further tests.”

Some of the heart-attack gender gap is thought to be down to women historically being underrepresented in clinical trials.

Not receiving same standard of care

A landmark 2018 study estimated 8,200 women in England and Wales died of heart problems between 2002 and 2013 because they did not receive the same standard of care as men.

It showed women having a heart attack have a 50 per cent greater chance than men of being misdiagnosed by the NHS.

It also found women are also less likely than men to receive potentially life-saving treatment for a heart attack in a timely way.

They are less likely to be admitted to a specialist heart ward and less likely to be prescribed medication to help prevent a second heart attack.

Last year, 17 global experts published a major report in The Lancet to tackle the inequity, saying doctors and the public needed to be better educated about the risks to women.

‘Potential to save lives’

Prof Chris P Gale, a consultant cardiologist at Leeds University who headed the 2018 study, said: “This has the potential to save lives.

“If women get an uncertain diagnosis we know follow-through care is slower and there are missed opportunities and excess deaths.

“I would now like this to be confirmed in a robust scientific setting.”

Dr Abbas-Hanif said: “Historically, clinical research, all the way from animal models to clinical trials has been conducted with a male focus and the biological differences of female physiology overlooked.

“This means subtleties of symptoms, the true impact of risk factors and thresholds of tests, such as troponin are skewed against women – who actually make up the majority of the population at 51 per cent.”

“CoDE-ACS adds another piece of the puzzle to change inequities in women’s health. As we learnt from the Lancet Commission, women face unconscious bias and disadvantage at every stage of the patient pathway.

“To see real change and prevent cardiovascular disease continuing to be the silent killer of women, pre-clinical and clinical trial research needs to become more representative of society, considering both men and women, a range of ages, ethnicities, social classes and beyond.”



Source link

Leave a Reply

Your email address will not be published.