Follow-Up: Garmin Vivoactive HR & Polar H10: Which measures heart rate more accurately?

Heart Rate Measurement Using Garmin & Polar Wearables

A study was made of the Garmin Vivoactive HR and Polar H10 chest strap in terms of comparative heart rate assessments. The units are shown in Figure 1 below. The two units involved included a wrist-based sensor (Garmin Vivoactive HR) and a chest strap (Polar H10).

Polar h10 chest strap and garmin vivoactive hr smart watch were used in the comparison

This follow-up focuses on 20 minutes of water rowing using both units in an effort to assess the heart rate measurement consistency and reliability. Both watch and chest strap were properly attached with no movement between these devices and the skin. Data were collected and then downloaded and processed through a Microsoft Excel spreadsheet. The data were time-synchronized so that corresponding data points from each device were associated in time. A summary of the analysis is provided here.

Time-Based Plots of Heart Rate

Overlay scatter plots of heart rate measurements versus time were made and are as shown in Figure 2.

Figure 2: Heart rate measurement while water-rowing approximately 20 minutes. Shown are overlays of Garmin Vivoactive HR and polar h10 heart rate versus time.

A general observation from the data is that the heart rate measurements from the two devices seem to overlap reasonably well as viewed by the naked eye. But there are key drops in measurements, particularly with the wrist-based heart rate sensor, that show as deviations in the overlap of the two signals. This can be seen more readily via the correlation curve shown in Figure 3. The correlation coefficient of 0.91 was determined between the two sets of measurements. It should be noted that the wrist-based sensor was snug with no movement on the wrist. Ambient temperature was approximately 80F.

As I showed in a previous post, there was a serious issue with the wrist-based sensor in which there were data dropouts with some significant time lags between measurements. In the case of the wrist-based sensor for the associated measurements here, this was also experienced. For comparison, I show histogram plots of the time intervals between measurements for both the wrist-based sensor (Figure 4) and the chest strap (Figure 5). The wrist-based sensor experiences a significant number of events in which the time between actual measurements are greater than one second. Indeed, from the figure, only 83 measurements during this interval were obtained within one-second of one another! There were a significant number of measurements in which the interval was > 1 second, with one as high as 40 seconds. The overall quantity of measurements was thus reduced to approximately 430 during the workout. On the other hand, the chest strap consistently measured at one-second intervals for a total of approximately 1320 measurements.

Figure 3: Scatter plot of heart rate as measured between the wrist-based Garmin device and the chest-strap Polar H10. A correlation coefficient of 0.91 was determined between the measurements. Perfect correlation is shown by the diagonal line.
Figure 4: Historgram of time between measurements for Garmin wrist-based sensor. Note the significant quantities of measurements in which the interval is greater than 1 second (the advertised measurement interval). For example, there were 20 instances in which the measurement interval was 6 seconds, and one instance in which the measurement interval was 40 seconds! Note that only 83 measurements were in the one-second interval width!
Figure 5: Historgram of time between measurements for the chest strap polar H10 sensor. All measurements (of which there were more than 1300) were reliably at one-second intervals.

Conclusions

Chest straps are much more reliable for heart rate measurement versus wrist-based sensors. Users of wrist-based sensors for heart rate measurement should be advised that measurements can be in question, as results illustrate here. This is not to say that chest straps are the gold-standard. Clearly, ECG measurement similar to those obtained through stress-testing are of diagnostic quality. Yet, for rate measurement chest straps are quite adequate and seemingly reliable.

Garmin Vivoactive HR & Polar H10: Which measures heart rate more accurately?

Figure 1: Polar h10 chest strap and garmin vivoactive hr smart watch were used in the comparison.

Heart Rate Measurement Using Garmin & Polar Wearables

A study was made of the Garmin Vivoactive HR and Polar H10 chest strap in terms of comparative heart rate assessments. Three different types of tests were conducted while the author wore these devices. The units are shown in Figure 1. The Garmin unit is able to be used with a number of sports, including rowing, and provides measurements of heart rate, stroke rate, distance per stroke, split times, and also provides for location tracking during the workout. Data can be uploaded to http://connect.garmin.com/and are also available for download in TCX (an XML format) as well as splits downloads in CSV format. The Polar H10 is strapped around the chest just below the level of the breast bone. This unit, too, can upload data to the http://flow.polar.com/web site, where data can be downloaded in TCX format, as well.  In order to provide some variety, I considered three different activities:

  • General workout, involving weight lifting, sit-ups, squats;
  • Walking for 1 mile; and,
  • Indoor rowing for 15 minutes.

In all cases, both the Vivoactive HR and the H10 were attached, with the Vivoactive HR snuggly affixed to the left wrist. Both watch and chest strap were properly attached with no movement between these devices and the skin. Data were collected and then downloaded and processed through a Microsoft Excel spreadsheet. The data were time-synchronized so that corresponding data points from each device were associated in time. Plots of the measurements were made.

Time-Based Plots of Heart Rate

Overlay scatter plots of heart rate measurements versus time were made of all three activities, shown in Figure 2 through Figure 4. Data were downloaded from the Garmin & Polar cloud sites and were uploaded into MS Excel. The data were then time synchronized using visual basic to align the measurements.

Figure 2: Heart rate measurement while walking 1 mile. Scatter overlay of Garmin Vivoactive HR and polar h10 heart rate versus time.
Figure 3: Heart rate measurement during general exercise activity. Scatter overlay of Garmin Vivoactive HR and polar h10 heart rate versus time.
Figure 4: Heart rate measurement while  rowing indoors on Concept 2 ergometer. Scatter overlay of Garmin Vivoactive HR and polar h10 heart rate versus time.

Heart Rate Comparison: Walking

Measurements of heart rate were taken during a one mile walk. The heart rates were plotted against one another and the correlation coefficient was computed between the two sets of measurements. In the case of the comparison shown in Figure 5, the correlation among measurements was rather poor: the correlation coefficient was determined to be -0.54. Perfect correlation is given by the diagonal line in the figure. Interesting to note is that the data points taken from the Vivoactive HR time variance. In the case of the Vivoactive HR, in some instances, the time between measurements was as high as 47 seconds with 62 measurements in the 12-14 second interval range, whereas in the case of the Polar H10, all measurements were 1 second interval. Thus, the number (quantity) of measurements taken by the Polar H10 were far denser than those of the Vivoactive HR.

Figure 5: Scatter plot of heart rate measured while walking one mile using the Polar H10 versus Garmin Vivoactive HR. The correlation coefficient of -0.54 was determined between the two sets of measurements. Perfect correlation is shown by the diagonal line.

Heart Rate Comparison: General Activity

In the case of general activity, which included some weight lifting, sit-ups, leg raises and standing exercises, the heart rate comparison is as shown in Figure 6. The correlation coefficient among these measurements is a bit higher at 0.60. The variation in measurement collection time associated with the Garmin HR was even higher here, with one measurement interval as high as 88 seconds!

Figure 6: Scatter plot of heart rate measured while performing weight lifting, sit-ups and general standing exercises using the Polar H10 versus Garmin Vivoactive HR. The correlation coefficient of 0.60 was determined between the two sets of measurements. Perfect correlation is shown by the diagonal line.

I have hypothesized that the wide variation in data collection time may be due to arm motion that is not experienced to the degree in walking. I also have hypothesized that the improved correlation may be due to the higher heart rate, which is more easily detected by the Vivoactive HR. We will see some supporting evidence of this in the final section on indoor rowing.

Heart Rate Comparison: Indoor Rowing

Rowing on the Concept 2 PM5 unit while wearing both the Vivoactive HR and the Polar H10 produced the results as illustrated in Figure 7. The correlation between the Vivoactive HR and the Polar H10 is much higher here, with a correlation coefficient of 0.95. Several items of note: the variation in measurements with the Vivoactive HR is much lower, with only two measurements 19 seconds apart and most measurements having 1-2 second intervals. This complies much more closely with the 1-second measurement intervals of the Polar H10. Furthermore, heart rate measurements are much higher here: some measurements as high as 165 beats/minute (during sprints). In general, corroboration between the two units is better as heart rate measurement is higher. This could be due to more accurate peripheral measurement.

Figure 7: Scatter plot of heart rate measured while performing indoor rowing using the Polar H10 versus Garmin Vivoactive HR. The correlation coefficient of 0.95 was determined between the two sets of measurements. Perfect correlation is shown by the diagonal line.

Conclusions

Based on the limited sampling and workouts thus far, the general conclusion regarding heart rate measurement “trust” is that the Polar H10 is more reliable based on several observations: (1) data collection time variation remains consistent at 1 second; and, (2) data density remains high with no dropouts in any of the workouts. This is not a surprise in general as the conventional wisdom is that chest straps are much more reliable. Yet, I wanted to quantify this reliability using some objective measures. It should be noted that while heart rate remains somewhat questionable with the Vivoactive HR, I have found that stroke rate measurement in comparison with the Concept 2 PM5 measurement is dead on accurate (at least based on the data I have observed).

Garmin Vivoactive HR for Rowing & Sculling

Vivoactive HR

Sculling and Rowing

I am a rower and sculler. I first cut my teeth in the sport over 30 years ago while at college rowing on the Charles River. I had been looking for the longest time for a device that I could use to track my heart, stroke rate, and also support GPS mapping of my workout while on the water. There are professional devices that track stroke rate and the like, such as Speed Coach GPSStroke Coach and Coxmate GPS. These are all excellent pieces of equipment, by the way. But, I am not in varsity rowing any more and I was looking for a piece of equipment that could support my rowing “habit” both for indoor and outdoor rowing (aside: I also possess a Concept 2 ergometer, which I love) while also serving the utilitarian purpose of being a good watch that can track heart rate full time.

When I row, however, I am really interested in being able to map the analytics to the motion. The Vivoactive HR enables me to do this as well as to post-process the data. I am into data. As a Chief Analytics Officer in the healthcare field for a medical device and real-time patient surveillance company, it is important to me to be able to access and understand the information collected during an activity. The connectivity and access to data provided by the Vivoactive HR are phenomenal.

Data view from Garmin Connect web site.

 

 

 

 

The figure above details an example analytics screen, which shows the map of the workout, heart rate, stroke rate, distance traveled at each measurement point, and allows tracking the entire workout with a cross-hair that is dynamic and interactive on the web screen. The unit supports many other types of workouts, including running, biking, pool, golf, walking, indoor rowing on ergometer, SUP rowing, XC skiing, indoor walking, indoor biking, and indoor running, and tracks sleep. The unit can be submerged in water and the battery life is amazing. I normally live with the unit on my wrist, and after 3 days of continuous use, battery is down to, perhaps 80%. I will take it off for an hour or so to charge, and it is good-to-go. I highly recommend this unit for the avid professional or veteran rower (like myself).

Update June 29th, 2017: Comparison among NK, Coxmate, Minimax

Robin Caroe of RowPerfect kindly left me a comment to this post last evening and provided an updated article on comparison among the NK, Coxmate GPS and Catapult Minimax which contains quite valuable data on performance related to these products. I have provided the hyperlink to the article above. Technological differences in sampling rate (e.g.: 5 Hz for NK versus 10 Hz for Coxmate) are important for accuracy. I must say that I was very close to purchasing the Coxmate GPS prior to investigating the Garmin. Upon reading the brochure for the Minimax S4, I am intrigued. The Minimax offers an update rate on the GPS that provides for precision in terms of location. In the Rowperfect article, of the key measures of performance identified, (1) heart rate & heart rate variability; (2) force and length of stroke; and, (3) GPS update rate are important measures for the elite athlete. In the case of the Minimax, GPS update on the order of 100 times per second (10 milliseconds) can reveal boat pitch, roll & yaw. Highly impressive. I would agree, though, that this level of accuracy and precision would be important for the competitive athlete. Yet, in my case (non-competitive, casual athlete), I still love my Garmin. I am able to see and track my position very accurately, monitor my stroke and heart rate, and in terms of heart rate variability, I can write an algorithm in R or Matlab to monitor that measure fairly directly.

As an added resource, Reviews.com has posted a comparison between best rowing machines for training and rowing experience. You can read that review at this link: The Best Rowing Machine: get a total-body workout on dry land.

A suggested method to control heart rate pacing and stroke volume in left-ventricular assist devices and for patients undergoing heart transplantation

Controller Design Concept

Update: this weblog article has been updated recently and a PDF of the document along with the new article is available here: Autonomic Heart Rate Controller Device Concept

I present a concept for autonomic cardiac pacing as a method to augment existing physiological pacing for both ventricular assist devices (VAD) and heart transplantations. The following development represents a vision and reflects an area that has yet to be fully exploited in the field. Therefore, the analysis is meant to be a starting point for further study in this area. Furthermore, an automatic control system methodology for both heart rate and contractile force (stroke volume) of patients having either an artificial left ventricular assist device (LVAD) or who have experienced degenerative performance of the Sinoatrial node is suggested. The methodology is described both in terms of a device and associated operational framework, and is based on the use of the naturally-occurring hormones epinephrine, norepinephrine, and dopamine contained in the return blood flow through the superior vena cava. The quantities of these hormones measured in the blood stream are used to derive a proportional response in terms of contractile force and pacing of the Sinoatrial node. The method of control suggests features normally described using cyclic voltammetry, expert systems, and feedback to pacing an artificial assist device.

Predicting Sepsis with Early Warning Shock Index

Sepsis hospitalizations increased between 2000 & 2008

Between 2000 and 2008, annual hospitalizations for septicimia increased from 326,000 to 727,000, in which 46% of hospitalized septic patients (patients diagnosed with sepsis or septicimia–blood infections) were admitted through the Emergency Department (ED). [Source: Hall MJ, Williams SN, DeFrances CJ, et al. Inpatient Care for Septicemia or Sepsis: A Challenge for Patients and Hospitals. U.S. Department of Health and Human Services – National Center for Health Statistics. 2011; 62:1-7.]

Predicting the future has never been shown to be possible with 100% certainty. It would seem to be a tautology that uncertainty invades every aspect of life. To a large degree, the inability to predict the future is a direct demonstration of the inability to control or, perhaps more correctly, the inability to fully account for all external influencing events that cause deviation from expected behavior. Thus, establishing certainty in causal relationships (that is, cause and effect) can be difficult.

Such is the case in medicine as in other aspects of life. Over the span of the last 150 years a great deal of research into epidemiology has led to great advances in the ability to predict behaviors based on prior causal evidence. The efficacy of drugs is testament to this statement—just look at penicillin to see a simple example of this. Perhaps the most illustrative example of the use of analysis and modeling to predict or assess the correlation between causal events and outcomes is the famous work of Dr. John Snow relative to identifying the root cause of the cholera outbreak in mid-19th century London. This work presented a cause-and-effect relationship surrounding the presence of cholera in a Broad Street well in 1854. Dr. Snow is identified as one of the fathers of epidemiology. In his analysis he identified the occurrence of cholera in Broad Street and the surrounding environs and traced the cause to a contaminated well. Once the pump handle was removed from the well, the epidemic began to subside. One could argue, perhaps, that early warning as to the likelihood of improvement was the removal of the pump handle on the offending well.

In other areas, though, the ability to predict the causal relationships is much less certain. The battle with cancer in its various forms is testament to this in that the complexity of the cause and effect relationship is not so well understood as to claim that any one approach or cause will guarantee an individual will or will not develop cancer in his or her lifetimes. To be clear, there are correlations and associations based upon past history that establish likely relationships assuming large populations (e.g.: smoking is likely to cause lung cancer). Yet, plenty of individual exceptions exist that violate the general population. My father is an example of an individual who smoked first cigarettes and then a pipe for the vast majority of his life and passed away at the age of 90…of old age. On the other hand, my mother, who did not smoke and yet watched her diet her entire life passed away from breast cancer at the age of 54. However, these individual examples, while outliers, do not provide me, as their progeny, with the rationale or an excuse to engage in destructive behavior. In other words, I am not willing to bet on the gene pool superseding the statistics from the larger population.

Other types of early warning measures (Sepsis and Other Syndromes)

In order to reduce uncertainty in predicting causal relationships it is necessary to improve the fidelity of the models with which the relationships are represented. But, the relationships based upon generalities can provide guidance for expected behavior, and this may be enough to do good, particularly when the expected behavior can target specific outcomes that can mean the difference between life and death. Two examples are given below to illustrate this point.

1) Viability for weaning from mechanical ventilation. Yang & Tobin [Yang KL, Tobin MJ: A prospective study of indexes predicting the outcome of trials of weaning from mechanical ventilation. N Engl J Med 1991; 324:1445-1450] put forth a measure based on the ratio of the respiratory rate to the tidal volume in patients undergoing spontaneous breathing trials. The so-called “rapid shallow breathing index”, or RSBI, is defined as:

RSBI = RR/TV.

In their study, Yang & Tobin determined that this ratio was a good discriminator for weaning success and failure. The threshold RSBI = 105 was identified as the point of demarkation relative to successful weaning (RSBI < 105) and unsuccessful weaning (RSBI > 105). This ratio is used operationally today yet with varying levels of success. However, one of its key benefits is its simplicity. In terms of predicting the causal relationships between successful weaning and failed weaning attempts, the RSBI is, perhaps, moderate to good. Yet, its simplicity lends itself to a general rule-of-thumb that clinicians can opt to use (or ignore) operationally that could motivate the seeking of further, more accurate or telling information that will have a higher, more telling influence on clinical predictability (e.g.: end-tidal CO2, for one).

2) Modified Shock Index and mortality in emergency department patients. Ye-cheung Liu et al. showed that the modified shock index (MSI) (Modified shock index and mortality rate of emergency patients; World J Emerg Med, Vol 3, No 2, 2012) is a superior measure to Shock Index (SI) alone for determining hemodynamic stability. Shock Index is a common measure to assess hypovolemic shock. Shock Index is given by:

SI = HR / BP-SYS,

this is the ratio of the heart rate (pulse) to the systolic component of blood pressure.

Modified Shock Index is given by:

MSI = HR / MBP,

this is the ratio of heart rate (pulse) to mean blood pressure (MBP), which is given by:

MBP = [(BP-DIAS x 2) + BP-SYS] / 3.

In the referenced study, MSI > 1.3 or MSI < 0.7 is associated with an increased probability of ICU admission and death.

Many more examples exist in which predicting causal relationships can be shown. The absolute predictability of the event is not possible to assess with 100% certainty. But, the ability to identify the likelihood of an event on the basis of the causal relationship is possible to show. In posts to follow we will take a look at the possible uses for these causal relationships in operational care management of patients.

Studies of Shock Index as an Early Warning Marker for Sepsis Onset is Promising

A cross-institution 2011 study presented in The Western Journal of Emergency Medicine (Berger et al., “The Shock Index and Early Recognition of Sepsis in the Emergency Department – A Pilot Study”) conducted by researchers at UC Davis, NY Hospital and Harvard University proffered using the Shock Index (SI) as an early warning marker for sepsis in the emergency department (ED). The objective of the study was to:

“…compare the ability of SI, individual vital signs, and the systemic inflammatory response syndrome (SIRS) criteria to predict the primary outcome of hyperlactatemia (serum lactate ≥ 4.0 mmol/L) as a surrogate for disease severity, and the secondary outcome of 28-day mortality.”

Per the studies presented above, a cohort of adult patients suspected of infection were screened for sepsis using data that included vital signs, laboratory data, and initial serum lactate levels.
Test characteristics were computed for two specific outcomes:
1) hyperlactatemia (identified as a marker for morbidity); and,
2) 28 day mortality (see here and here).
Covariates considered in the analysis for sepsis diagnosis included:
a. heart rate (HR) > 90 beats/min
b. mean arterial pressure (MAP) < 65 mmHg
c. respiratory rate (RR) > 20 breaths/min
Shock Index = HR / SYS-BP
Results of the analysis showed the following causal relationships:
of 2524 patients with complete medical record data considered in the analysis, 290 patients presented with hyperlactatemia and 361 died within 28 days.

Threshold measures for Shock Index as an Early Warning Score

 Patients with SI > 0.7 (representing 15.8%) were 3 times more likely to present with hyperlactatemia than those with normal SI (4.9%). Negative predictive value of SI > 0.7 was 95%. This was identical to that of SIRS. Furthermore, an SI > 1.0 was “the most specific predictor of both outcomes.”

Autonomic Heart Controller Device Concept

What is an Autonomic Heart Rate Controller?

The idea for extending the performance of left-ventricular assist devices (LVADs) occurred to me more than 15 years ago. The idea led me to write a white paper at that time which has been maintained and archived on this web site.

Discussions regarding heart rate variability (HRV) caused me to research a conceptual controller for autonomic, chemoreceptor-based sinoatrial heart control. As HRV is affected by sympathetic and parasympathetic control, this fact reminded me of a paper (unpublished) which I had written 14 years ago. While I included a web-adaptation of this paper very early in the history of the web log, I never included the actual paper itself. Much has transpired over the years in the regard to medical device integration, control and research into physiologic monitoring. Yet, I have not seen any writing in particular associated or closely related to the topic at hand.

Heart Rate Autonomic Control White Paper

Originally written in August 2002, the attached white paper presents a concept for a mechanism to automatically controller for heart rate pacing and contractile force (stroke volume) of either an artificial left ventricular assist device (LVAD) or a patient’s own heart who has experienced degenerative performance of the Sinoatrial node.

Autonomic control is based on the hormonal action of concentrations of catecholamines within the blood stream. These, in turn, influence the sinoatrial node through uptake. The concept laid out in the attached paper “operates” by analyzing the chemical nature of the epinephrine, norepinephrine, and dopamine content of the return blood flow through the superior vena cava and then using this information via cyclic voltammetry, neural network control, and feedback to the pacing device to control the heart rate of the assist device.

Why Autonomic Control is Potentially Interesting?

The basic premise of left ventricular assist devices (LVAD) and in heart pacing in general is to provide enough contractile force to move blood throughout the system. Most systems these days rely upon vasodilation, which is related to hormones as well as work. Yet, heart contractility is also related to emotion and release of hormones which are unrelated to direct movement or action of the human body. Hence, the idea was to accommodate a more “realistic” concept that took into account not only the contractile or vasodilation / vasoconstriction aspects of the arteries and veins, but also the hormonal changes measured through catecholamine changes in the blood stream. It is recognized that the attached paper is very raw and, clinically, it is somewhat naive. Yet, the objective was to present the idea as a potential starting point for research and, eventually, a product that could extend the performance of existing LVADs to support more human-like, natural behavior of artificial hearts.

Update on measurement of blood enzymes to support autonomic control:

Recently, a team at Ecole Polytechnique Federale de Lausanne (EPFL) in Switzerland developed an implant to monitor chemicals in the blood. This implant, according to this recent article is reportedly the world’s smallest at 14 mm and measures a maximum of 5 indicators to include troponin, lactase, glucose, ATP, to show whether a heart attack has occurred or to track (in the case of diabetic patients) blood enzymes and protein levels. The information can subsequently be transmitted via Bluetooth to a smartphone for online tracking.

This work is encouraging as it lays the foundation for real-time sensing, data collection and analysis at a level that is necessary for more accurate modeling and monitoring of cardiovascular systems. Such work can lead to earlier detection and more accurate prostheses — particularly, artificial hearts utilizing autonomic control to augment heart rate pacing– as well as more accurate clinical decision support methods that can determine whether patients have experienced critical events. I can see applications to earlier stroke detection where time-is-brain. Such real-time autonomic methods, when integrated with mobile technology and health information technology, can lead to great advances in patient care management through early warning resulting in early intervention.