I’d like to welcome you to the first article of our new blog The Journal. I am Niclas Granqvist, the Chief Technology Officer of Nanoleq AG and responsible for the technology inside our products. This Journal will focus on the science of wearables, but I also hope we can touch on other topics such as the development of wearables, the human mind and monitoring of vital signs. Sometimes, we will also invite expert guest bloggers. The Journal’s goal is to be scientific but only to the extent that everybody understands. We appreciate dialogues with readers, so feel free to mail comments to science@nanoleq.com.

This premiere post deals with the hot topic of accuracy. We all encountered inaccurate wearables and devices in the past. I’m often frustrated because my watch is not giving the right readings of my heart rate, or the activity monitor is always overestimating my energy expenditure. But why is the accuracy of wearables so difficult, and why do we think Nanoleq technologies improve accuracy?

Accuracy depends greatly on the measurement method. Heart rate today is typically measured from the pulse on the wrist using scattered light (photoplethysmogram, PPG), but the measurement has inherently poor performance as the wrist is badly perfused and the pulse is weak. The majority collect information using optical sensors 24/7, but the truth is that the information is often poor and dubious. The following graphs from the Nanoleq Biometrics Lab show the difference between a smartwatch and a Polar H10 heart rate sensor during rest and exercise.

The graphs are illustrative of the many poor performances of smartwatches, as we notice that sometimes the signal is delayed (Figure 1), the signal is just off (Figure 2), or there is no consistent signal at all (Figure 3). Sometimes the smartwatch follows the reference „fairly“ well (Figure 4).

Nanoleq’s solution to these accuracy problems is unique. Integration with textiles allows us to place measurements around the body and connect sensors using Body Area Networks (BAN) of flexible, soft and stretchable cables. This way, we place high-performance electrodes around the body that are stretchable, comfortable and invisible. The sticky electrode surface mimics the compression of a belt without compromising comfort. We use the electrical signal (electrocardiogram, ECG) of the heart to measure heart rate, which has a significantly better signal quality and is far stronger than wrist measurement.

In exercise physiology, heart behaviour is a key metric in understanding our body and the effect of movement. Optical methods are particularly challenging in sports such as cycling and weight lifting. The reason is the motion artefacts from under the skin. It may come as a surprise that optical methods work better in running, and the explanation is the repetitiveness of movement. ECG, on the other hand, works well in all sports. In fact, the quality of the Nanoleq electrode is so high that it also works well in swimming.

In [1], the authors compare different wearables and methods (ECG and optical) for heart rate variability (HRV) and stress. HRV is notable because it is difficult to measure as one must catch every heartbeat down to a few milliseconds’ precision. The research shows that the best signal quality is obtained with ECG/Polar H10 transmitter. ECG is considered the gold standard and largely accepted throughout the medical community. Nanoleq technology is based on ECG and has the benefit of the very best accuracy of data available.

Niclas Granqvist

CTO Nanoleq

References:
[1] Umair, Chalabianloo, Sas and Ersoy, “HRV and Stress: A mixed-methods approach for comparison of wearable heart rate sensors for feedback.”.  IEEE Access 2021, 10.1109/ACCESS.2021.3052131