Though millions of lives around the globe end in sudden cardiac death (SCD) each year, signs of a troubled heart can be notoriously difficult to spot.

A new method for identifying cardiac rhythms associated with imminent heart failure could one day buy precious time for those at risk.

Created by researchers from Tampere University in Finland, the new algorithm makes use of a particular metric called detrended fluctuation analysis (DFA2 a1), which can detect changes in heart rate variability over time.

Where heart attacks occur when blood flow to the heart is restricted, SCD involves the heart being overwhelmed by short electrical impulses. Though it primarily occurring in older individuals, the interrupted rhythms often appear without any prior symptoms.

Based on an analysis of 2,794 adults over an average follow-up period of 8.3 years, the team found that DFA2 a1 is a "powerful and independent predictor" of SCD. The association is strongest when the body is at rest, rather than engaging in physical activity.

"The most interesting finding of the study is the identification of differences specifically during measurements at rest," says Tampere University physicist Teemu Pukkila.

"The characteristics of heart rate intervals of high-risk patients at rest resemble those of a healthy heart during physical exertion."

The team used statistical analysis methods to connect DFA a1 patterns to SCD incidents. The approach included factoring in the impact of other important variables, including age and existing heart health conditions.

Encouragingly, the reading of the metric only takes a minute, and could be done via sensors simple enough to fit into a smartwatch. There wouldn't have to be any trips to a clinic or complicated scans to assess someone's SCD risk.

"Accelerometers in wearable consumer devices can easily distinguish between the states of physical activity and rest and perform the measurement when applicable," write the researchers in their published paper.

The new predictive algorithm is significantly more accurate than current methods, which typically involve measuring cardiorespiratory fitness: that means someone's capacity to send oxygen to the muscles, and the extent to which those muscles can use the oxygen during physical exercise.

The next steps are to test the approach with larger and more diverse groups of people, and to see how the findings might relate to other types of heart disease too. Ultimately, the predictive algorithm could end up saving a substantial number of lives, by warning those at risk from this sudden and quick killer.

"It is possible that in many previously asymptomatic individuals, who have suffered sudden cardiac death or who have been resuscitated after sudden cardiac arrest, the event would have been predictable and preventable if the emergence of risk factors had been detected in time," says cardiologist Jussi Hernesniemi from Tampere University.

The research has been published in JACC: Clinical Electrophysiology.