Artificial intelligence recognizes Covid-19 on the breath

What does the breath sound like in Covid 19 patients? And what does their lungs look like? Researchers at the ETH Lausanne (EPFL) want to answer this question with artificial intelligence, as the university announced.

Mary-Anne Hartley is a doctor and researcher at the ETH Lausanne (EPFL). Together with her team and university hospitals, she is working on using algorithms to diagnose Covid-19 and predict how seriously the patients will become ill. The researchers named the new deep learning algorithms DeepChest and DeepBreath. With the DeepChest they use data from lung ultrasound images, with the DeepBreath they use breath sounds that they hear with a digital stethoscope. “The results so far are very promising,” said Martin Jaggi from EPFL.

The basis for DeepChest are lung ultrasound images of thousands of corona patients who were admitted to the emergency room of the University Hospital Lausanne (CHUV). Alain Gervaix, pediatrician and infectiologist, has been listening to patients’ breath sounds at Geneva University Hospital since 2017 in order to develop an intelligent digital stethoscope, the “pneumoscope”. The idea came to him when he was talking to his daughter: he explained to her that while listening to the lungs he could hear noises that would help identify asthma, bronchitis or pneumonia, according to the Geneva doctor.

Chosen for you

Algorithm detects changes in lung tissue

The EPFL researchers used this data to develop the DeepBreath algorithm. Initial results would suggest that he can detect changes in lung tissue before the disease breaks out. In this way, asymptomatic Covid 19 patients can be identified at an early stage. The researchers hope that the application will be available by the end of the year.

The team is further refining the algorithms – among other things with the help of a one-year hackathon that Hartley announced in March. “We want to create robust and trustworthy tools that will still be useful after the pandemic,” she said.

Chosen for you

The researchers hope that the algorithms will one day be able to differentiate between viral and bacterial pneumonia – in order to make a contribution in the fight against antibiotic resistance. They are also working to ensure that the algorithms also work on cell phones. In the spring, EPFL researchers developed an app based on artificial intelligence that is supposed to detect the typical corona cough.

Source link

These were the details of the news Artificial intelligence recognizes Covid-19 on the breath for this day. We hope that we have succeeded by giving you the full details and information. To follow all our news, you can subscribe to the alerts system or to one of our different systems to provide you with all that is new.

It is also worth noting that the original news has been published and is available at and the editorial team at AlKhaleej Today has confirmed it and it has been modified, and it may have been completely transferred or quoted from it and you can read and follow this news from its main source.

NEXT India's foreign ministry launches app to connect to citizens abroad