The DeepBreath AI deep learning algorithms developed at EPFL were originally under development for other lung conditions but have used new training data to identify patterns of Covid-19 in breath sounds
“It’s not a relaxing time to study infectious diseases,” said Dr Mary-Anne Hartley, a medical doctor and researcher in EPFL’s intelligent Global Health group (iGH). iGH is based in the Machine Learning and Optimization Laboratory of Professor Martin Jaggi, a world leading hub of AI specialists, and part of EPFL’s School of Computer and Communication Sciences. “We’ve named the new deep learning algorithms DeepChest – using lung ultrasound images – and DeepBreath – using breath sounds from a digital stethoscope. This AI is helping us to better understand complex patterns in these fundamental clinical exams. So far, results are highly promising,” said Prof Martin Jaggi.
Hartley is working with nearby Swiss university hospitals on two major projects.
At HUG, the Geneva University Hospitals, Professor Alain Gervaix has been collecting breath sounds since 2017 to build an intelligent digital stethoscope, the “Pneumoscope”. This was intended to diagnose pneumonia, and a version to be released by the end of the year should enable the diagnosis of COVID-19 from breath sounds. The first results suggest that DeepBreath is even able to detect asymptomatic COVID by identifying changes in lung tissue before the patient becomes aware of them.
“Pneumoscope with the DeepBreath algorithm can be compared to applications which can identify music based on a short sample played. The idea came from my daughter when I explained to her that auscultation allows me to hear sounds which help me identify asthma, bronchitis or pneumonia,” said Prof Gervaix.
CHUV, Lausanne’s University Hospital, is leading the clinical part of the DeepChest project, collecting thousands of lung ultrasound images from patients with Covid-19 compatible symptoms admitted to the Emergency Department.
This started in 2019 to identify markers to distinguish between viral and bacterial pneumonia but pivoted to Covid-19. “Many of the patients who agreed to take part in our study were scared and very ill but they wanted to contribute to broader medical research, just like we do,” said Dr Noémie Boillat-Blanco, principal investigator, I think there is a collective motivation to learn something from this crisis and to rapidly integrate new scientific knowledge into everyday medical practice.”
The algorithms have been pre-published on the EPFL website. Work is also underway to develop an application that allows these complex deep learning algorithms to work on mobile phones, even in the most remote regions. The challenge will be getting the detection engines on low cost microcontrollers, something that is focus of French developer Cartesiam.
“We want to collect data from under-represented communities so that our tools can be accurate even in poor settings,” said Hartley. “Our algorithm is for instance specifically designed to tolerate errors in image or sound collection and inconsistent quality, which are more likely in those types of settings.”
“Covid-19 has sensitized people to the vulnerability of public health, and its enormous complexity. The need to build large scale AI research efforts to understand and react to rapidly emerging data has never been more obvious. Let’s hope the momentum continues beyond the pandemic, and can be used to enable equitable access to health care,” she said.
The algorithms are avaialble at www.epfl.ch/labs/mlo/deepbreath/
Related Covid-19 articles
- XILINX TEAMS FOR CLOUD-BASED OPEN SOURCE X-RAY CLASSIFICATION
- IoT TEMPERATURE INDICATORS FOR VACCINE DISTRIBUTION
- PEOPLE COUNTER USES AI ON ST MICROCONTROLLER
- AI PROJECT TO MONITOR PATIENTS WITH ULTRASOUND
Other articles on eeNews Europe
- Panasonic finally looks at European battery gigafactory
- Battle of the BBC boards
- Slovakia aims to build world’s fastest AI supercomputer
- Dyson to spend $3bn on AI technologies
- ARM spins out its Pelion IoT business