You sleep, but your heart keeps telling stories. The rhythm of beats, the pauses between them, the respiratory waves — all shift from one sleep stage to the next. In a sleep lab, this is read from a host of sensors on the head, face, legs, and chest. It's cumbersome, expensive, and not something you can do at home.
The new study proposes to get by with just a single ECG lead — the same kind already used in Holter monitors and wearable patches. The algorithm, dubbed Holter-to-Sleep, takes an overnight heart recording and produces a sleep profile: when you fell asleep, how long the deep and REM phases lasted, how often you woke up. And at the same time, it delivers a full set of cardiac metrics, just like a standard Holter exam.
The method was tested on 10,439 studies from four public databases collected at different clinics. Then it was run on independent recordings from wearable patches — to make sure the algorithm hadn't adapted to any particular lab. The agreement of sleep stages with reference polysomnography was high enough to suggest practical utility.
This approach removes the main obstacle to large-scale studies of the link between sleep and heart disease. Previously, you had to bring people into the lab for a night, and even then, the conditions were far from home. Now you can wear a lightweight patch for a week, live your normal life, and get a complete picture: how you sleep, how your heart beats, and how the two influence each other.
During sleep, the heart doesn't just slow down: it beats differently in REM and non-REM phases. In REM, the rhythm becomes irregular, almost as if under mild stress, even though you're dreaming.