Cover Story

Current Advances in Wearable Health Technology: A Review

Tawseef Dar, MD1; Bharath Yarlagadda, MD1; Rakesh Gopinathannair, MD, FACC, FHRS2; Dhanunjaya Lakkireddy, MD, FACC, FHRS3

1Cardiac Arrhythmia Research Fellow, Division of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital & Medical Center, Kansas City, KS; 2Associate Professor of Medicine, Director of Cardiac Electrophysiology, University of Louisville, Louisville, KY; 3Professor of Medicine, Division of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital & Medical Center, Kansas City, KS

Tawseef Dar, MD1; Bharath Yarlagadda, MD1; Rakesh Gopinathannair, MD, FACC, FHRS2; Dhanunjaya Lakkireddy, MD, FACC, FHRS3

1Cardiac Arrhythmia Research Fellow, Division of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital & Medical Center, Kansas City, KS; 2Associate Professor of Medicine, Director of Cardiac Electrophysiology, University of Louisville, Louisville, KY; 3Professor of Medicine, Division of Cardiovascular Diseases, Cardiovascular Research Institute, University of Kansas Hospital & Medical Center, Kansas City, KS

Introduction

The concept of “wearable health technology” has gained much popularity in the health sector over the past several years. Although portable heart rate monitors and fitness devices are nothing new, the recent introduction of smartwatches and devices have taken wearable health technology to a new horizon. Besides monitoring heart rate, these wearable devices have the potential to identify the underlying rhythm, and the data can be transferred via Bluetooth link to a suitable mobile phone. Around 33 million people worldwide currently use smart wearable technology, with an expected growth to 90 million users in 2019.1 Some of the smart devices use photoplethysmographic (PPG) technology used by conventional pulse oximeters to measure heartbeat. The current focus of these companies is to develop more robust PPG-based algorithms for detection of atrial fibrillation (AF), that will be as accurate as ECG-based algorithms. 

For the purpose of easy understanding, we have divided these wearable technologies for monitoring heart rate and rhythm into the following sections below. (Table 1) 

Besides monitoring heart rate, these wearable devices have the potential to identify the underlying rhythm, and the data can be transferred via Bluetooth link to a suitable mobile phone. Around 33 million people worldwide currently use smart wearable technology, with an expected growth to 90 million users in 2019.1 Some of the smart devices use photoplethysmographic (PPG) technology used by conventional pulse oximeters to measure heartbeat. The current focus of these companies is to develop more robust PPG-based algorithms for detection of atrial fibrillation (AF), that will be as accurate as ECG-based algorithms. 

For the purpose of easy understanding, we have divided these wearable technologies for monitoring heart rate and rhythm into sections (Table 1). 

Mobile Cardiac Outpatient Telemetry (MCOT or MCT)

MCT is a device system that continuously records the electrical beat-to-beat activity of the heart for up to 30 days. The standard technology consists of a small sensor and monitor, attached to the patient’s chest by three leads, that the patient wears while continuing their normal routine. Recently, patch MCT devices that communicate by Bluetooth to a cell phone-sized monitor have been introduced, which provide one or two ECG leads with application of the patch sensor (which has no lead wires) to the patient’s chest. Direct contact between the patient and monitoring company is constantly maintained, so any event occurrence is instantaneously transmitted to a monitoring center for analysis and response. The patient can also trigger a recording at any time by pushing a record event button on the monitor. In contrast, a Holter monitor continuously records a patient’s EKG for only 24-48 hours, while an event monitor records a patient’s EKG only at the time of an event, with no real-time transmission. A multicenter, randomized control trial (RCT) comparing MCT to loop monitors showed that the arrhythmia detection or exclusion rate was 88% with MCT vs 75% with loop monitors.2 However, the question of whether real-time transmission and interpretation of data has any impact on overall morbidity and mortality is still debated. Examples of devices for MCT include the MCOT Mobile Cardiac Outpatient Telemetry (BioTel Heart) and Mobile Cardiac Telemetry 3 Lead (MCT 3L) (BioTel Heart). 

In more recent years, continuous technological evolution in medical science has led to the development of miniaturized medical devices for continuous cardiac outpatient monitoring. These adhesive ambulatory electrocardiographic monitors or patches have integrated microelectronics for short- to medium-term (days to weeks) continuous cardiac monitoring, and are challenging conventional, widely used multi-lead wearable devices. The patch monitoring system consists of a sensor, a microelectronic circuit with recorder and memory storage, and an internal battery embedded in a relatively flexible synthetic matrix, resin, or other material. They are easy to use, leadless, water resistant, and usually intended for single use. Currently available patch systems are the Zio XT system (iRhythm Technologies, Inc.), MCOT Patch (BioTel Heart), ePatch® Extended Holter (BioTel Heart), and SEEQ MCT System (Medtronic). The Zio XT system does not have real-time transmission capability, and therefore, requires the user to return the device in a postage-paid envelope upon study completion. The MCOT Patch and SEEQ MCT System have a separate cellular data transmitter besides the sensor (two-piece design), which ensures real-time data transmission to the company data network. However, both of these devices are dependent on a company technician's accurate collection and reporting of raw data as well as generation of a summary report. The Zio Patch has shown a higher arrhythmia detection rate when compared to 24-hour Holter (96 vs 61 arrhythmic events; P<0.001).3 Also, the results of the Early Prolonged Ambulatory Cardiac Monitoring in Stroke (EPACS) study, comparing the Zio Patch with 24-hour Holter monitoring in cryptogenic stroke patients, were recently presented at the 3rd European Stroke Organisation Conference (ESOC) 2017. Teo et al showed that AF was detected in 16.3% of patients in the patch group compared with only 2.1% in the standard Holter group (odds ratio of 8.9 95% confidence interval, 1.1-6.0; P=0.047).4 The Zio XT, MCOT Patch, ePatch, and SEEQ MCT are approved by the U.S. Food and Drug Administration (FDA), and Zio XT, ePatch and SEEQ MCT have received the CE Mark for use in the European Union.

More recently, Qardio launched their new device, QardioCore, for continuous EKG outpatient monitoring. Equipped with powerful sensors that record 20 million data points, the device is worn as a chest strap, with no wires and no need for patches or gel. It is a three-lead wireless continuous EKG monitor with Bluetooth transmission of data to a smartphone/watch app (iOS only), which can be viewed and analyzed in real time by the user or their doctor. Besides heart rate, it can record several other biomarkers, such as respiratory rate, skin temperature, and posture changes. It was FDA approved in January 2017.

Due to lack of substantial evidence showing the overall clinical benefit of these new devices, Holter monitors, event monitors, and implantable loop recorders continue to remain the gold standard for diagnosis of arrhythmias. However, these devices are being widely used to assess arrhythmia burden to help clinicians in further decision making.5

Smartphone Event Monitors (Generating Lead I Rhythm Strip)

More recently, technological advancements have allowed users to generate rhythm strips from their smartphones or smartwatches. This system uses an external case (that acts as electrodes when touched by the right or left hand of the user) and a downloadable application on a smartphone, which together form an “event monitor.” The cardiac electrical signal is converted to an ultrasound FM sound signal, and the application in the smartphone then demodulates the signal to a digital EKG tracing. The EKG is generated in real time, and can be stored and instantaneously transmitted by the phone to a secure server for further analysis.

The first product of its kind is the KardiaMobile case or card (AliveCor), which consists of two metal electrodes on the back of a case that clips on or attaches to a smartphone. Following its FDA 510(k) approval in 2012, the device became available for prescription in 2013, and received over-the-counter approval in 2014 as a single-channel ECG recorder.6 The technology has demonstrated a sensitivity of 94-100% and a specificity of 91-95% for detecting atrial fibrillation, and has proven to be a simple and cost-effective method of screening for this arrhythmia.7,8 The utility of this device system has been studied in varied clinical settings.9,10 The fourth-generation KardiaMobile ECG system is compatible with most iOS and Android devices.

The ECG Check mobile heart monitor from Cardiac Designs received FDA approval in January 2013. The ECG Check monitor, which attaches to the back of a smartphone, can record, store, transfer, and analyze single-channel ECGs wirelessly through the ECG Check app and the ECG Check Web Center. The user receives their results within seconds.11 

However, this device system has its own limitations like AliveCor. Tracings tend to have more baseline artifacts than standard 12-lead EKGs, making accurate recording and documentation of ventricular rhythms such as monomorphic VT or polymorphic VT very difficult at times.12 Another limitation of such device systems is that the patient must be able to activate the smartphone for an event to be recorded. Therefore, its utility in patients with very transient symptoms is questionable. Implantable loop recorders or event monitors remain the gold standard in such cases.

AliveCor also recently developed a wearable technology called the KardiaBand, which is compatible with the Apple Watch. The users can record a single-lead EKG by touching KardiaBand’s integrated sensor, which communicates with the Kardia app. Using the “atrial fibrillation detector” and “normal detector” algorithms, the app then indicates whether the heart rate and rhythm are normal. Recordings are stored and viewed in the Kardia app, and can also be sent to the user’s doctor. The KardiaBand was recently approved by the FDA.

AF screening using frequent intermittent short EKG recordings at home has shown increased sensitivity for AF detection (STROKESTOP study).13 Therefore, these smartphone event monitors can prove to be very convenient, effective, and user-friendly tools for AF screening in the future.

Photoplethysmographic (PPG) Based Smart Device AF Detection

Several downloadable applications have been developed (both for iOS and Android users) for smartphones that use PPG-based algorithms to detect atrial fibrillation or normal sinus rhythm. These applications do not require external hardware, and use only the phone’s camera and light. The user puts his right index or second finger on the sensor, and the signal recorded is then processed through an app with built-in algorithms for AF detection. McManus et al demonstrated excellent specificity (0.975), sensitivity (0.962), and accuracy (0.968) using an iPhone 4S camera and algorithm.14 Chan et al also compared the Cardiio Rhythm smartphone PPG application with the AliveCor automated algorithm, and found a sensitivity of 92.9% for PPG technology (higher than that of the AliveCor automated algorithm) as well as a specificity of 97.7% for PPG technology (comparable with AliveCor). However, the positive predictive value of PPG technology was lower (53.1% vs 76.9%) than the AliveCor automated algorithm.15 The PPG technology also has its own limitations; for example, it cannot differentiate between frequent atrial premature beats or atrial arrhythmias, which may lead to false diagnoses of atrial fibrillation.

Furthermore, in 2016, McManus et al demonstrated that an enhanced smartphone app could accurately discriminate pulse recordings during AF from sinus rhythm, premature atrial contractions (accuracy of 0.955), and premature ventricular contractions (accuracy of 0.960).16

PPG-based technology has been extended to wearables such as smartwatches. A recent poster presentation at HRS 2017 by Sanchez et al showed that the Cardiogram app for the Apple Watch accurately distinguished pulse recordings during AF from those obtained during normal sinus rhythm (using deep neural network [DNN] based algorithms).17 

Oster and Clifford reported that AF detection accuracies of PPG-based technology using DNN algorithms rapidly declines to 70-80% with the addition of typical amounts of noise observed in the real ambulatory settings.18 However, Shashikumar et al proposed a new model using the Simband (Samsung),19 in which they applied a continuous wavelet transform to the PPG data, and a convolutional neural network (CNN) algorithm was derived to detect AF. Combining the output of the CNN with features calculated based on beat-to beat variability and signal quality significantly improved AF detection accuracy, and also provided additional discriminatory power to DNN algorithms to separate out noise from AF data.20

Future Directions

Earlobe sensor: Conroy et al recently tested the feasibility of an earlobe PPG sensor in detecting AF and showed promising results, with a sensitivity and specificity of 90.9%.21 This technology can be integrated into wearable devices to enable continuous cardiac monitoring.  

Stretchable heart rate monitor using gold nanoparticles: This device is an ultra-thin, stretchable electronic device using gold nanoparticles that can be applied to human skin like an adhesive sticker. It can sustain complicated mechanical deformations or stretching associated with any movement. The device consists of EKG sensors and amplifiers that monitor the heart rate and can also store the data. It is still under development.

Stretchy skin patch powered by a cell phone: It can measure heart rate and oxygen saturation using four LEDs to shine different colors of light into the skin; any changes in the reflected light are picked up by the photodetectors. The heart rate is displayed as a flashing light. It has no built-in battery, and radio signals from a nearby phone power this device.

Polar Team Pro shirt: Polar announced its smart workout shirt designed for professional athletes for continuously monitoring their heart rate and other metrics. It has two sensors woven into the fabric, which come into contact with the skin to monitor the heart rate, as well as a pouch underneath the collar that holds a pod to track other metrics (such as motion, speed, acceleration, etc.). This has not yet been studied in any trials.

Conclusion

With the availability of a variety of smart devices and apps, cardiac monitoring has a very bright future. These device systems and apps attempt to integrate monitoring tools into an everyday lifestyle, thereby ensuring compliance to its best. Smartphone monitors and patches have the potential to replace traditional event and Holter monitors for diagnostic purposes. Similarly, PPG-based AF detection can serve as excellent screening tools. However, these device systems and apps need to be further validated in larger randomized studies in order for wide acceptance among physicians. For now, their use is mostly limited to screening for arrhythmias (such as AF) in apparently healthy individuals.

Disclosures: The authors have no conflicts of interest to report regarding the content herein.  

References

  1. Projected smart/basic wearables unit shipments worldwide 2017 to 2021. Statista. Available online at http://bit.ly/2EWO2jp. 
  2. Rothman SA, Laughlin JC, Seltzer J, et al. The diagnosis of cardiac arrhythmias: a prospective multi-center randomized study comparing mobile cardiac outpatient telemetry versus standard loop event monitoring. J Cardiovasc Electrophysiol. 2007;18(3):241-247.
  3. Barrett PM, Komatireddy R, Haaser S, et al. Comparison of 24-hour Holter monitoring with 14-day novel adhesive patch electrocardiographic monitoring. Am J Med. 2014;127(1):95.e11-17.
  4. Kaura AK, Lenane JL, Sztriha LS, et al. Early prolonged ambulatory cardiac monitoring in stroke (EPACS) ­— an open-label randomised controlled trial. EP Europace. 2017;19(suppl_1):i6-i.
  5. Tarakji KG, Natale A, Wazni OM, et al. Incidence of asymptomatic atrial fibrillation recurrence post pulmonary vein isolation using a novel continuous event monitoring system. Heart Rhythm. 2005;2(5):S193-S194.
  6. AliveCor’s Heart Monitor for iPhone Receives FDA Clearance. AliveCor. Published December 3, 2012. Available online at http://bit.ly/2EUpPub. Accessed January 18, 2018.
  7. Lowres N, Neubeck L, Salkeld G, et al. Feasibility and cost-effectiveness of stroke prevention through community screening for atrial fibrillation using iPhone ECG in pharmacies. The SEARCH-AF study. Thromb Haemost. 2014;111(6):1167-1176.
  8. Tarakji KG, Callahan T, et al. Evaluating the efficacy and feasibility of a novel wireless recording system using smartphone in monitoring patients after atrial fibrillation ablation procedure: the iTransmit study. Heart Rhythm Scientific Sessions; San Francisco, CA. Abstract AB07-04. 2014.
  9. Waks JW, Fein AS, Das S. Wide complex tachycardia recorded with a smartphone cardiac rhythm monitor. JAMA Intern Med. 2015;175(3):437-439.
  10. Chung EH, Guise KD. QTC intervals can be assessed with the AliveCor heart monitor in patients on dofetilide for atrial fibrillation. J Electrocardiol. 2015;48(1):8-9.
  11. Cardiac Designs ECG Check Device for Heart Rhythm Analysis Cleared for Over the Counter Use. Cardiac Designs. Published May 6, 2013. Available online at http://bit.ly/2DrUsdj. Accessed January 18, 2018.
  12. Garabelli P, Reynolds D. Accuracy and novelty of an inexpensive iPhone-based event recorder. Heart Rhythm Scientific Sessions; May 11, 2012; Boston, MA. Abstract IA02-1.
  13. Svennberg E, Engdahl J, Al-Khalili F, Friberg L, Frykman V, Rosenqvist M. Mass Screening for Untreated Atrial Fibrillation: The STROKESTOP Study. Circulation. 2015;131(25):2176-2184.
  14. McManus DD, Lee J, Maitas O, et al. A novel application for the detection of an irregular pulse using an iPhone 4S in patients with atrial fibrillation. Heart Rhythm. 2013;10(3):315-319.
  15. Chan PH, Wong CK, Poh YC, et al. Diagnostic Performance of a Smartphone-Based Photoplethysmographic Application for Atrial Fibrillation Screening in a Primary Care Setting. J Am Heart Assoc. 2016;5(7).
  16. McManus DD, Chong JW, Soni A, et al. PULSE-SMART: Pulse-Based Arrhythmia Discrimination Using a Novel Smartphone Application. J Cardiovasc Electrophysiol. 2016;27(1):51-57.
  17. Sanchez JM, Ballinger B, Olgin JE, et al. Detecting Atrial Fibrillation Using a Smart Watch - The mRhythm Study. Heart Rhythm Scientific Sessions; May 11, 2017; Chicago, IL. Abstract C-AB12-01.
  18. Oster J, Clifford GD. Impact of the presence of noise on RR interval-based atrial fibrillation detection. J Electrocardiol. 2015;48(6):947-951.
  19. Welcome to Simband. Available online at https://www.simband.io/. Accessed January 18, 2018.
  20. Shashikumar SP, Shah AJ, Li Q, Clifford GD, Nemati S. A deep learning approach to monitoring and detecting atrial fibrillation using wearable technology. 2017 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI), Orlando, FL. 2017, pp. 141-144. doi: 10.1109/BHI.2017.7897225.
  21. Conroy T, Guzman JH, Hall B, Tsouri G, Couderc JP. Detection of atrial fibrillation using an earlobe photoplethysmographic sensor. Physiol Meas. 2017;38(10):1906-1918.
/sites/eplabdigest.com/files/dar.pdf