Revision as of 16:05, 29 May 2016

Optical Heart Rate Monitoring detects the blood filling the capillaries under your skin as your heart beats. Each time your heart beats the capillaries expand with blood, and this ebb and flow can be used to determine your heart rate. Most optical heart rate monitors for use while exercising shine a green light into the skin and use a receptor to detect the changes in the reflected light.

This is a view of the optical sensors of the Basis Peak, Garmin 225, TomTom Cardio Runner, and the Garmin 235 (left to right).

This approach has been used for decades, and I had an early version back in the 1980s. The latest optical heart rate monitors are vastly superior, but still have many accuracy issues. The most accurate form of heart rate monitoring is to use a chest strap that picks up the electrical signal from the heart. While a chest strap is not perfect, it works remarkably well as long as it has good contact with your skin, the battery is not flat, and it's not malfunctioning.

1 Anecdotal Accuracy

Many reviews of optical heart rate monitors will include anecdotal comparisons such as the ones shown below.

During this run you see the 235 having a couple of major dropouts. For the rest of the run, the 235 roughly tracks the true heart rate.

Here we see the 235 giving an accurate reading, but one that is rather misleading. While I frequently see the 235 displaying a heart rate that is wildly too high or too low, I know I can ignore that information. Where the 235 is more problematic than other optical systems is that it will display a plausible but inaccurate value.

For this run the 235 initially gives an inaccurate reading that is somewhat close to the real heart rate, but then spikes to wildly too high. I tried several times to adjust the tension and position of the 235, but nothing helped.

These comparisons can be quite useful to give a sense of the type of problems that are common with optical heart rate monitors. However, they provide no quantification of the accuracy, making it impossible to evaluate if one device is better than another. Therefore, I decided it would be worthwhile to undertake a slightly more rigorous approach.

2 Methodology

To evaluate the accuracy in a more quantifiable manner I've analyzed the heart rate of readings of several optical heart rate monitors compared with a chest strap based monitor. My first step was to verify that the chest strap based monitor is reasonably accurate. To do this, I ran with two different chest strap systems, a Garmin Ant+ (HRM4) and a Polar Bluetooth (H7). I compared over 10,000 readings and found that the two systems matched extremely well. I have uploaded the results below, using two different visualizations. The first graph shows the heart rate measured by the Garmin HRM4 on the horizontal against the Polar H7 on the vertical. If the two systems match exactly then the point will be on the green line of equality. You can see a few places where the two systems don't line up perfectly, but the vast majority of the readings align within a couple of heartbeats. I've used transparent points to give a better impression of the density of data, with black areas having at least 10 data points lining up. There is a blue regression line, which will be aligned with the green line if the system is accurate. I've also included two red lines that are 25 bpm away from the true value. The second graph shows the distribution of differences between the two systems, and again we can see that the vast majority of data points are within a couple of beats. For those with a statistical background, the standard deviation is 1.55 BPM with a bias of -0.06 BPM.

The distribution of readings between the Polar Bluetooth system vertically and the Garman Ant+ system horizontally. This is how the distribution should look if the system being tested is accurate.

Here is another view of the same data, showing the most readings were exactly the same, or differing by just a couple of beats.

3 Garmin 235 Accuracy

The graphs below highlight the accuracy problems off at the Garman 235's optical heart rate monitor. Of the 20,000 readings, slightly more than 10% were out by more than 25 BPM (demarked by the red lines on the first graph.) You can see a distinct cloud of points clustered high above the true readings. It's possible that this cloud represents the optical system becoming confused by the impact of my feet landing, as they are vaguely in the vicinity of my typical Cadence. Other runners have reported the phenomenon of optical heart rate monitors displaying wildly high values. You can also see this set of high values in the distribution graph, way it is displayed as a small bump to the right hand side of the main spike. The average error (standard deviation) is 19.1 BPM, with an average reading that was 5.7 BPM too high. If you're an experienced runner that has a good idea what your heart rate should be, then you may be able to ignore values that are out by more than 25 bpm. In that case, the standard deviation drops to 4.2 BPM with an average that is 0.03 too low. Of course, if you know your heart rate to within 25 BPM, then the 235 will only get you slightly closer.

The distribution of readings between the Garmin 235 OHRM vertically and the Garman Ant+ system horizontally.

Here is another view of the same data, showing the greater error.