Roeland Decorte grew up in a nursing home in Belgium, where he learned to spot the first subtle signs of mental decline in small changes in the way residents walked or talked. When Decorte was 11, his father, the owner and director of the home, began waking him up in the middle of the night with chest pains and an overwhelming sense of impending doom.
She went to two doctors, who briefly listened to her heartbeat through stethoscopes and diagnosed her with anxiety. But the symptoms persisted, and it was not until she underwent a full series of scans at a private hospital that a third doctor discovered the source of the problem: a tiny hole between the left and right chambers of her heart. If it had gone undetected, it would have cost her her life: she was 39.
Disaster averted, young Decorte was able to focus on his studies, and by the age of 17 he was a student at Cambridge University, the youngest Belgian to attend the prestigious institution. (This caused some logistical problems: His tutor had to become his legal representative, and a new payment system had to be set up at the faculty bar to prevent him from buying alcohol like his classmates.)
He spent the next seven years specializing in the deciphering ancient codesand a comfortable career in academia (or a more exciting one as an Indiana Jones-style relic hunter) appealed to him. But Decorte never stopped thinking about what had happened to his father, and how he could have been diagnosed much sooner if a doctor—any doctor—had spent more than 30 seconds listening to his heart. So in 2019, with no medical training but armed with the confidence that only an Oxbridge education brings, Decorte, then 27, founded a company and turned his attention to cracking a different ancient code: the secret rhythm of the heart.
Algorithms for listening to conditions in the body
Artificial intelligence (AI) is on the rise in healthcare, and the only thing holding it back is a lack of data. Meanwhile, doctors, pressed for time, only gather information sporadically. Wearable devices, such as smartwatchesthey measure the pulse, but they are poor for more specific diagnoses (partly because the wrist is practically far from the truly vital organs).
Decorte wanted to develop technology that would continuously and accurately monitor the body, so that people like his father could get the treatment they needed more quickly. At first, he tried embedding sensors in clothing so people could track their vital signs without going to a doctor. Then he designed an elaborate exoskeleton packed with sensors to measure all sorts of ailments. This attracted some military interest, but it wouldn’t really have helped someone like Decorte’s father. “I was very naive,” he said when we met recently in the wood-paneled basement of a seedy coffee shop in London’s Mayfair. “I spent about two years working full-time in the spare room of my house doing nothing else.” But the problem he continually encountered was noise: Unless he built a contraption that pressed each sensor against the skin, there was too much random interference from people moving around the world to get a good sense of what was actually going on in the body.
But perhaps, Decorte thought, noise would also be the solution. During the pandemic, he met Erika Bondareva, a PhD student, who had published a paper on diagnosing Covid by analysing audio data collected from people coughing into an app. Her software Bondareva and Decorte looked for patterns common to people with the disease, and then researched those same patterns to try to detect it early in other individuals. Together, Bondareva and Decorte worked to expand the approach to other ailments, starting with heart disease. Over time, Decorte says, she ended up replacing all the sensors on the exoskeleton she had designed with an audio sensor. Eventually, she realized that the only way to detect the disease was to use the sensor. hardware What I needed was a microphone.
