In nature, 1/f fluctuation appears everywhere: in the wavelengths of light, in music (Mozart and singers alike), in electrical signals, butterflies, fish, wind, sunlight filtering through leaves, the sound of a stream, and flames.

Our heartbeat follows the same 1/f fluctuation day to day. Taking regular Apple Watch readings in the morning, at midday and in the evening therefore makes the analysis far more meaningful.

By reviewing the slope of 1/f fluctuation and the Lorenz plot from daily measurements, and by following real-time and resting heart rate through figures and graphs, you come to feel the work of your heart up close.

And by watching those figures and graphs change, you build a habit of paying attention to your own health — this is what is known as biofeedback, and it is recognised as an effective approach to self-care.

To improve analytical accuracy, aim for a clean heart-rate waveform.

Rest your arm on a table or your knee and keep your finger still for 30 seconds. After scrolling down to the result and tapping "Done", take another 30-second measurement.

Four measurements in total are needed to gather two minutes of data — which makes HearTomo roughly four times as accurate as apps that rely on a single 30-second reading.

Try to take the four measurements back to back, without long gaps in between.

Built on a heart-rate data analysis server system (patent 6430729) developed through joint research between Motohisa Osaka — a graduate of the Faculty of Mathematics at the University of Tokyo and of the Nippon Medical School, now Professor Emeritus at the Nippon Institute of Life Sciences — and ZMP.

Osaka carried out heart-rate fluctuation analysis using a proprietary formula together with the spectral analysis algorithm developed by Professor Ronald Berger at Johns Hopkins Hospital.