Cough analyser app ‘is better than doctors at diagnosing children with conditions such as asthma, pneumonia and bronchitis’ and is 97% accurate
- Software analyses sounds of child’s cough and provides on the spot diagnosis
- It outperformed panel of doctors in accurately diagnosing five different diseases
- Hoped app will allow treatments to begin sooner by removing physical exam
Thousands of parents could be spared countless trips to the doctors thanks to a smartphone app that can diagnose children with asthma.
The app works by listening to youngsters coughing into the device’s microphone and analysing the sounds associated with five conditions.
A trial found it could distinguish between asthma, bronchiolitis, pneumonia, croup and lower respiratory tract disease with up to 97 per cent accuracy.
Researchers found it outperformed senior doctors at diagnosing all five of the conditions.
A new smartphone app can distinguish between respiratory diseases in children by analysing soundwaves in their coughs (file image)
It can be difficult to differentiate between respiratory disorders in children, even for experienced doctors.
Scientists hope, by removing the need for a physical examination, the app will allow treatments to begin sooner.
Researchers at Curtin University and The University of Queensland, Australia, programmed the app to recognise the soundwaves in coughs, similar to speech recognition technology.
CHILDREN’S NOSES MAY HOLD CLUES TO LUNG INFECTION RISK
Examining the bacterial makeup of a child’s nose could help doctors improve the diagnosis and treatment of serious lung infections, scientists say.
Researchers from the University of Edinburgh found that bacteria and viruses at the back of the nose and throat of children with respiratory infections is different to that of healthy youngsters.
These differences indicate the severity of the condition and could help doctors predict how long the affected child needs to spend in hospital.
The findings could aid doctors when assessing the risk of lung infection (Andrew Matthews/PA)
In less serious cases they could be helped to recover naturally, reducing the need for antibiotics.
Lower respiratory tract infections (LRTIs), including pneumonia and bronchiolitis, are a leading cause of death in under-fives around the world.
Symptoms include shortness of breath, weakness and fever.
Professor Debby Bogaert of Edinburgh University’s Medical Research Council Centre for Inflammation Research, who led the study, said: ‘Lung infections can be extremely serious in children and babies and are very distressing for parents.
‘Our findings show for the first time that the total microbial community in the respiratory tract – rather than a single virus or a bacteria – is a vital indicator of respiratory health.
‘This could really impact on how doctors diagnose LRTIs and use precious antibiotics to fight infections.’
Researchers worked with teams in The Netherlands to take samples from more than 150 children under the age of six hospitalised with an LRTI and compared them with samples from healthy children.
They found the ‘microbiome’ – the population of bacteria and viruses – at the back of the nose and throat was related to that seen in the lungs, making it easier to understand and diagnose infections.
The microbiome profiles allowed them to identify 92 per cent of children as being healthy or ill when combined with factors such as the child’s age.
It also provided a marker of the infection’s severity.
They then used the app to categorise the coughs of 585 children between ages 29 days to 12 years, who were being cared for at two hospitals in Western Australia.
Recordings were done in realistic hospital environments with background noises including talking, crying and medical devices.
The accuracy of the app was determined by comparing its results with the diagnosis of a panel of doctors.
The doctors were allowed to examine the patients and review results of imaging, laboratory findings and hospital charts.
It had 97 per cent accuracy in spotting asthma, compared to 91 per cent by doctors and 87 per cent for pneumonia, compared to 85 per cent by the panel.
It scored 83 per cent accuracy when picking up on lower respiratory tract disease, compared with 82 per cent by doctors.
And it was 85 per cent accurate in spotting croup, compared to 82 per cent by the panel.
The tech also snuffed out bronchiolitis 84 per cent of the time, compared to 81 per cent by the team of experts.
Dr Paul Porter, lead author of the study, said: ‘It can be difficult to differentiate between respiratory disorders in children, even for experienced doctors.
‘This study demonstrates how new technology, mathematical concepts, machine learning and clinical medicine can be successfully combined to produce completely new diagnostic tests utilising the expertise of several disciplines.
‘As the tool does not rely on clinical investigations, it can be used by health care providers of all levels of training and expertise.
‘However, we would advise that where possible the tool should be used in conjunction with a clinician to maximise the clinical accuracy.’
Dr Samantha Walker, Director of Research and Policy at Asthma UK, said described the findings as ‘exciting’.
She said: ‘Asthma is very difficult to diagnose in children, particularly those under five, as the tools available for doctors to use are crude so this smartphone app is potentially an exciting development.
‘Around one in five children have to wait three or more years for an asthma diagnosisis, which places a huge emotional and practical toll on parents as many see their children in and out of hospital, fighting for life.
‘While this new tool shows how technology in healthcare has the potential to transform the lives of children with asthma, more research is needed to make sure that these findings are as reliable as possible and can be used in routine care practice.’
The findings were published in the journal Respiratory Research.