Sources of Inaccuracy in Photoplethysmography for Continuous Cardiovascular Monitoring

In the paper

“All six identified devices with FDA status appear to be able to provide patients and providers with data for oxygen saturation (SpO2), respiration rate, and pulse rate.”

In the paper

“In some cases, FDA approval is only for when the wearable is used within a software suite or healthcare framework. This is the case with the Samsung Gear S2, which is FDA cleared to monitor heart rate toward detecting atrial fibrillation when done with the LIVMOR Halo™ Detection System.”

In the paper

“To the authors’ knowledge, there is no work that comprehensively summarizes the literature surrounding these sources of noise and how they affect the waveform of the PPG and its derivatives.”

In the paper

“To the authors’ best knowledge, literature does not exist that demonstrates this relationship for the finger, a common location for pulse oximeters.”

In the paper

“It has been shown that the temperature of the body site where PPG data are being collected can influence the resultant waveform amplitude, and even the time between waveforms. This is summarized in Table 5.”

In the paper

“While many sources of noise had documented potential solutions, few had a comprehensive solution and even our understanding of how conditions such as obesity affect the skin and cardiovascular monitoring are still developing.”

In the paper

“PPG can fill this gap if a sufficiently accurate and precise device is developed.”

In the paper

“PPG systems developed for remote and wearable use are typically for general wellness and fitness. This precludes it from being prescribed for medical use at home.”

In the paper

“The indications for use in these devices are very significant advancements in remote monitoring, but still lag in the potential prognostic capabilities of the PPG.”

In the paper

“Remote and continuous/intermittent monitoring (RCIM) has proven to be a promising route to deliver preventative care by reducing both the death rate and burdens placed on the healthcare system [1,2,3].”

What the sources say

• Not checked yetcites [1] Wilkoff, 2008, HRS/EHRA expert consensus on the monitoring of cardiovascular implantable electronic devices (CIEDs): Description of techniques, indications, personnel, frequency and ethical considerations: Developed in partnership with the Heart Rhythm Society (HRS) and the European Heart Rhythm Association (EHRA); and in collaboration with the American College of Cardiology (ACC), the American Heart Association (AHA), the European Society of Cardiology (ESC), the Heart Failure Association of ESC (HFA), and the Heart Failure Society of America (HFSA). Endorsed by the Heart Rhythm Society, the source
• Not checked yetcites [2] Roger, 2011, Heart disease and stroke statistics—2011 update: A report from the American Heart Association, Circulation source
• Not checked yetcites [3] Jencks, 2009, Rehospitalizations among patients in the Medicare fee-for-service program, N. Engl. J. Med. source

In the paper

“The blood pressure cuff, ECG patch, and Holter monitor are also not often used for long-term remote monitoring [24].”

What the sources say

• Not checked yetcites [24] Bansal, 2018, Portable out-of-hospital electrocardiography: A review of current technologies, J. Arrhythmia source

In the paper

“In addition, the hardware and software within the device itself can contribute to the noise.”

In the paper

“Additionally, these tests are rarely administered preventatively, despite research which concludes that preventative testing could reduce deaths by up to 25% [1].”

What the sources say

• Not checked yetcites [1] Wilkoff, 2008, HRS/EHRA expert consensus on the monitoring of cardiovascular implantable electronic devices (CIEDs): Description of techniques, indications, personnel, frequency and ethical considerations: Developed in partnership with the Heart Rhythm Society (HRS) and the European Heart Rhythm Association (EHRA); and in collaboration with the American College of Cardiology (ACC), the American Heart Association (AHA), the European Society of Cardiology (ESC), the Heart Failure Association of ESC (HFA), and the Heart Failure Society of America (HFSA). Endorsed by the Heart Rhythm Society, the source

In the paper

“These sources of error pertain to variation within and across individuals (e.g., skin tone, obesity, age, and gender), physiology (e.g., respiration, venous pulsation, body site of measurement, and body temperature), and external perturbations of the device itself (e.g., motion artifact, ambient light, and applied pressure to the skin).”

In the paper

“Additionally, some parameters in the literature such as pulse transit time (PTT), which is used to determine pulse wave velocity (PWV) and estimate blood pressure without a cuff, requires extraction of the time delay from two PPG waveforms or from an ECG and PPG waveform [31,32].”

What the sources say

• Not checked yetcites [31] Nabeel, 2018, Arterial blood pressure estimation from local pulse wave velocity using dual-element photoplethysmograph probe, IEEE Trans. Instrum. Meas. source
• Not checked yetcites [32] Nabeel, P., Karthik, S., Joseph, J., and Sivaprakasam, M. (2017, January 8–10). Experimental validation of dual PPG local pulse wave velocity probe. Proceedings of the 2017 IEEE International Symposium on Medical Measurements and Applications (MeMeA), Rochester, MN, USA. source

In the paper

“Numerous non-FDA cleared/approved fitness devices exist and can estimate heart rate by quantifying the number of systolic peaks in a period of time, but a single parameter limits the amount of extractable information.”

In the paper

“However, complex analyses of the PPG waveform yield valuable clinical information, such as: blood pressure, respiratory information, sympathetic nervous system activity, and heart rate variability.”

In the paper

“Overall, the literature has demonstrated the potential diagnostic and prognostic strength for the PPG; however, the PPG features can only be utilized if the waveform is of a high quality with high signal-to-noise ratio (SNR).”

In the paper

“Beyond fitness and heart rate monitoring, the primary medical use of the PPG has been focused on obtaining information about the cardiovascular system towards cardiovascular disease (CVD) diagnosis and treatment [17,18].”

What the sources say

• Not checked yetcites [17] Elgendi, 2019, The use of photoplethysmography for assessing hypertension, NPJ Digit. Med. source
• Not checked yetcites [18] Vakily, 2017, A system for continuous estimating and monitoring cardiac output via arterial waveform analysis, J. Biomed. Phys. Eng.

In the paper

“A wearable monitoring tool to be used beyond fitness and heart rate must account for noise sources originating from individual patient variations (e.g., skin tone, obesity, age, and gender), physiology (e.g., respiration, venous pulsation, body site of measurement, and body temperature), and external perturbations of the device itself (e.g., motion artifact, ambient light, and applied pressure to”

In the paper

“These results indicate that overall, the mean modulation goes down with increasing Fitzpatrick scale, and with exercise one would expect greater blood volume changes causing a higher signal for the more absorbing blue/green wavelengths, at least for the source to detector distances used in the study.”

In the paper

“With 20 subjects ranging from Fitzpatrick II to Fitzpatrick IV, the perfusion index (AC/DC), pulsatile strength, and SNR were all found to increase when data collected with the 590 nm LED were analyzed.”

What the sources say

• Not checked yetcites [51] Mohapatra, P., Preejith, S., and Sivaprakasam, M. (2017, January 22–25). A novel sensor for wrist based optical heart rate monitor. Proceedings of the 2017 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Torino, Italy. source

In the paper

“Obesity, determined by having a body mass index greater than 30, affects 40% of the United States and leads to cardiovascular disease, hypertension, and type 2 diabetes [57].”

What the sources say

• Not checked yetcites [57] Poirier, 2006, Obesity and cardiovascular disease: Pathophysiology, evaluation, and effect of weight loss: An update of the 1997 American Heart Association Scientific Statement on Obesity and Heart Disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism, Circulation source

In the paper

“However, the wavelength range needs to be optimized for both skin tone and blood absorption, particularly as more advanced parameters are derived from PPG signals, as the absorption of green light by melanin in individuals with a darker skin tone limits the light penetration to the subcutaneous tissue where the blood is located.”

In the paper

“It was found that relative amplitude of the AC component to the entire signal (AC + DC) increased as LED/photodiode spacing increased from 4 to 11 mm, although no statistical analyses were presented.”

What the sources say

• Not checked yetcites [53] Mendelson, 1988, Noninvasive pulse oximetry utilizing skin reflectance photoplethysmography, IEEE Trans. Biomed. Eng. source

In the paper

“There was an error of 1.04 beats per minute (BPM) for “fair” skinned individuals and an error of 10.90 BPM for “dark” skinned participants, citing the lack of usability of their device for dark skinned individuals.”

What the sources say

• Not checked yetcites [40] Preejith, S., Alex, A., Joseph, J., and Sivaprakasam, M. (2016, January 12–14). Design, development and clinical validation of a wrist-based optical heart rate monitor. Proceedings of the 2016 IEEE International Symposium on Medical Measurements and Applications (MeMeA), Benevento, Italy. source

In the paper

“Blood flow regulation and oxygen saturation are both known to deviate with respect to obesity and BMI [71].”

What the sources say

• Not checked yetcites [71] Limberg, 2016, Peripheral Blood Flow Regulation in Human Obesity and Metabolic Syndrome, Exerc. Sport Sci. Rev. source

In the paper

“Preejith et al. confirmed that skin tone matters when using green light only. The analysis of their PPG biosensor (single source 535 nm LED) showed a significant direct correlation SNR of the heart rate with skin tone, indicating that a darker skin tone yields higher error in the measurement [40].”

What the sources say

• Not checked yetcites [40] Preejith, S., Alex, A., Joseph, J., and Sivaprakasam, M. (2016, January 12–14). Design, development and clinical validation of a wrist-based optical heart rate monitor. Proceedings of the 2016 IEEE International Symposium on Medical Measurements and Applications (MeMeA), Benevento, Italy. source

In the paper

“However, many PPG devices use green light (~550 nm). The decision to use green light in most PPG systems is primarily driven by the relatively high absorption spectrum of hemoglobin in this range (Figure 3b), which is the main absorber in blood and thus can potentially give a strong pulsatile signal with changes in blood volume.”

In the paper

“In particular, obesity can lead to changes in skin thickness, blood flow, and oxygen saturation. This affects the optical properties of skin in addition to the distance light has to travel to reach a target vessel or vessels [48,59,60].”

What the sources say

• Not checked yetcites [48] Jacques, 2013, Optical properties of biological tissues: A review, Phys. Med. Biol. source
• Not checked yetcites [59] Elgendi, 2012, On the analysis of fingertip photoplethysmogram signals, Curr. Cardiol. Rev. source
• Not checked yetcites [60] Iacopi, 2020, Using Skin Bioengineering to Highlight How Weight and Diabetes Mellitus Modify the Skin in the Lower Limbs of Super-Obese Patients, Diabetes Metab. Syndr. Obes. Targets Ther. source

In the paper

“Individuals with obesity experience increased cutaneous blood flow to meet the oxygenation needs of tissue. However, the blood flow of adipose tissue generally decreases with obesity both after a meal and during a fasting state [61,72].”

What the sources say

• Not checked yetcites [61] Chin, 1999, Increased cutaneous blood flow but impaired post-ischemic response of nutritional flow in obese children, Atherosclerosis source
• Not checked yetcites [72] Altintas, 2016, In vivo reflectance-mode confocal microscopy assessments: Impact of overweight on human skin microcirculation and histomorphology, J. Biomed. Opt. source

In the paper

“These often-overlooked disparities with diversity (e.g., skin tone and obesity) are now becoming more documented in the literature [34,35,36].”

What the sources say

• Not checked yetcites [34] Colvonen, P.J., DeYoung, P.N., Bosompra, N.-O.A., and Owens, R.L. (2020). Limiting Racial Disparities and Bias for Wearable Devices in Health Science Research, Oxford University Press US. source
• Not checked yetcites [35] Hidalgo, 2021, Critical care trainees call for pulse oximetry reform, Lancet Respir. Med. source
• Not checked yetcites [36] Rodriguez, 2021, Synthetic Photoplethysmography (PPG) of the radial artery through parallelized Monte Carlo and its correlation to Body Mass Index (BMI), Sci. Rep. source

In the paper

“This is due to the high absorbance of eumelanin with a peak in the ultraviolet wavelength (220 nm) and a steady decay through the visible wavelength region.”

In the paper

“Since the absorption of epidermal melanin is much higher in the visible region of light and much lower in the near infrared region (NIR), the NIR range of light will travel further through pigmented skin.”

In the paper

“In determining heart rate, it was found that bias (defined as the mean difference between the OH1 and ground truth) increased with darker skin (p < 0.001), and heart rate accuracy was positively correlated to skin tone (p < 0.05).”

What the sources say

• Not checked yetcites [41] Hermand, 2019, Validation of a photoplethysmographic heart rate monitor: Polar OH1, Int. J. Sports Med. source

In the paper

“Additionally, the wearable devices used in this study utilized red and near infrared light, which can more easily penetrate the epidermis.”

What the sources say

• Not checked yetcites [46] Bent, 2020, Investigating sources of inaccuracy in wearable optical heart rate sensors, NPJ Digit. Med. source

In the paper

“For Fitzpatrick IV individuals, the perfusion index increased between 1.2 and 7.1 times with the use of the 590 light, while pulsatile strength increased by 1.1 to 3.1 times and SNR increased by 1.3 to 2.6 times, although no statistical significance analyses were performed [51].”

What the sources say

• Not checked yetcites [51] Mohapatra, P., Preejith, S., and Sivaprakasam, M. (2017, January 22–25). A novel sensor for wrist based optical heart rate monitor. Proceedings of the 2017 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), Torino, Italy. source

In the paper

“In the resting condition, green light had larger mean modulation (p < 0.001) than other wavelengths, and after exercise, blue and green had greater SNR ratios than red or infrared (p < 0.001).”

What the sources say

• Not checked yetcites [52] Fallow, 2013, Influence of skin type and wavelength on light wave reflectance, J. Clin. Monit. Comput. source

In the paper

“It is possible that the lack of increase in error observed is due to the already large error present in the results reported—the mean absolute error is approximately 9 BPM across all skin tones.”

In the paper

“Most of the listed changes are due to the cardiovascular system, as arteries change how they respond to a cardiac pulse as the vessel ages.”

In the paper

“Perhaps most detrimental to the PPG waveform is skin thickness, as it is directly correlated to BMI and can dampen PPG signal amplitude [56,70].”

What the sources say

• Not checked yetcites [56] Gibney, 2010, Skin and subcutaneous adipose layer thickness in adults with diabetes at sites used for insulin injections: Implications for needle length recommendations, Curr. Med. Res. Opin. source
• Not checked yetcites [70] Derraik, J.G., Rademaker, M., Cutfield, W.S., Pinto, T.E., Tregurtha, S., Faherty, A., Peart, J.M., Drury, P.L., and Hofman, P.L. (2014). Effects of age, gender, BMI, and anatomical site on skin thickness in children and adults with diabetes. PLoS ONE, 9. source

In the paper

“The ratio b/a has been found to correlate positively to age, while c/a, d/a and e/a have all been correlated negatively to age [29,59].”

What the sources say

• Not checked yetcites [29] Takazawa, 1998, Assessment of vasoactive agents and vascular aging by the second derivative of photoplethysmogram waveform, Hypertension source
• Not checked yetcites [59] Elgendi, 2012, On the analysis of fingertip photoplethysmogram signals, Curr. Cardiol. Rev. source

In the paper

“Keratinocytes become shorter, water content decreases, lipid content decreases, and most notably, collagen synthesis and turnover decrease [93].”

What the sources say

• Not checked yetcites [93] Farage, 2013, Characteristics of the aging skin, Adv. Wound Care source

In the paper

“This change in skin thickness decreases the necessary propagation of light to obtain a PPG signal, which can contribute to an increase in PPG signal strength, along with capillary depletion.”

In the paper

“The increased absorption of deoxygenated hemoglobin compared to that of oxygenated hemoglobin in the far-red region could decrease SNR of PPG measurements that use those wavelengths of light.”

In the paper

“PPG principally measures the response/elasticity of arteries to blood flow, and thus will change if the properties governing artery compliance changes.”

In the paper

“A Holtain skin caliper applied to the dorsal forearm found a decrease in skin thickness, and an infrared PPG was used to determine PPG amplitude. The authors attribute the positive, direct relationship between PPG amplitude and age to capillary depletion, due to similar skin thicknesses observed between older participants and children, despite an increased PPG amplitude in the older participants.”

What the sources say

• Not checked yetcites [94] Leveque, 1984, In vivo studies of the evolution of physical properties of the human skin with age, Int. J. Dermatol. source

In the paper

“Using data from 93 individuals of various ages, Ahn et al. extracted features from fingertip-based PPG and the second derivative PPG, acceleration plethysmography (APG, Figure 4), and correlated them to chronological age in order to determine what they defined as the vascular age index [83].”

What the sources say

• Not checked yetcites [83] Ahn, 2017, New aging index using signal features of both photoplethysmograms and acceleration plethysmograms, Healthc. Inform. Res. source

In the paper

“Another study by Hartmann et al. determined that there is no statistically significant change in PPG amplitude as a function of age; however, this work featured 36 individuals ranging in age from only 33 to 58.”

What the sources say

• Not checked yetcites [95] Hartmann, 2019, Quantitative comparison of photoplethysmographic waveform characteristics: Effect of measurement site, Front. Physiol. source

In the paper

“However, the integumentary system also responds to aging.”

In the paper

“Suzuki and Oguri similarly used artificial intelligence to determine blood pressure from a cuffless monitor and incorporated age into their algorithm [96].”

What the sources say

• Not checked yetcites [96] Suzuki, S., and Oguri, K. (2009, January 2–6). Cuffless blood pressure estimation by error-correcting output coding method based on an aggregation of adaboost with a photoplethysmograph sensor. Proceedings of the 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Hilton Minneapolis, MI, USA. source

In the paper

“Phuong and Maibach noted that gender differentiation in average blood pressure and average heart rate indicate that baseline differences in vascular and cardiovascular parameters must be considered to determine cardiovascular health between males and females, which also extends into analyses of the PPG waveform [98].”

What the sources say

• Not checked yetcites [98] Phuong, C., and Maibach, H.I. (2016). Gender differences in skin. J. Textb. Aging Skin, 1729–1755. source

In the paper

“This leads to dramatic changes in PWV and subsequently PTT.”

In the paper

“Proctor et al. reported that the average heart rate for males was 70–72 BPM and the average for females was 78–85 BPM in 16 endurance-trained men and 14 endurance-trained women [99].”

What the sources say

• Not checked yetcites [99] Proctor, D.N., Beck, K.C., Shen, P.H., Eickhoff, T.J., Halliwill, J.R., and Joyner, M.J. (1998). Influence of age and gender on cardiac output-V o 2 relationships during submaximal cycle ergometry. J. Appl. Physiol. source

In the paper

“Artificial intelligence and machine learning have been used to determine blood pressure from PPG and mitigate the effects age have on the estimate.”

What the sources say

• Not checked yetcites [92] 2011, Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques, Artif. Intell. Med. source
• Not checked yetcites [96] Suzuki, S., and Oguri, K. (2009, January 2–6). Cuffless blood pressure estimation by error-correcting output coding method based on an aggregation of adaboost with a photoplethysmograph sensor. Proceedings of the 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Hilton Minneapolis, MI, USA. source

In the paper

“An algorithm that incorporates age into systolic and diastolic blood pressure determination was created.”

What the sources say

• Not checked yetcites [92] 2011, Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means of machine learning techniques, Artif. Intell. Med. source

In the paper

“Skin thickness is another gender-dependent variable known to affect signal strength (as mentioned previously), and is observed to be higher in men than women across all age ranges [97,103].”

What the sources say

• Not checked yetcites [97] Dao, 2007, Gender differences in skin: A review of the literature, Gend. Med. source
• Not checked yetcites [103] Shuster, 1975, The influence of age and sex on skin thickness, skin collagen and density, Br. J. Dermatol. source

In the paper

“These significant differences in artery diameter lead to significantly different radial artery flow rates as well, with a flow rate of 21 ± 4 and 10 ± 1 mL/min for males and females, respectively [102].”

What the sources say

• Not checked yetcites [102] Joannides, 2002, Influence of vascular dimension on gender difference in flow-dependent dilatation of peripheral conduit arteries, Am. J. Physiol. Heart Circ. Physiol. source

In the paper

“However, post-menopausal females have higher blood pressure than males, which can be attributed to arterial stiffness [101].”

What the sources say

• Not checked yetcites [101] Reckelhoff, 2001, Gender differences in the regulation of blood pressure, Hypertension source

In the paper

“Males have a 15–30% increase in heart mass compared to females resulting in females’ hearts having to beat faster to maintain the same output [100].”

What the sources say

• Not checked yetcites [100] Prabhavathi, 2014, Role of biological sex in normal cardiac function and in its disease outcome—A review, J. Clin. Diagn. Res. JCDR

In the paper

“The physiological differences between men and women extend into cardiovascular health and are thus noticeable in the PPG waveform [97,98].”

What the sources say

• Not checked yetcites [97] Dao, 2007, Gender differences in skin: A review of the literature, Gend. Med. source
• Not checked yetcites [98] Phuong, C., and Maibach, H.I. (2016). Gender differences in skin. J. Textb. Aging Skin, 1729–1755. source

In the paper

“Shuster et al. measured skin thickness in 90 Caucasian men and 107 women and found that forearm skin thickness is greater in men than women, but no statistics were presented to explore statistical significance [103].”

What the sources say

• Not checked yetcites [103] Shuster, 1975, The influence of age and sex on skin thickness, skin collagen and density, Br. J. Dermatol. source

In the paper

“Post-puberty, both males and females have an increase in arterial compliance, determining no gender-dependent differences between the measurement.”

What the sources say

• Not checked yetcites [104] Ahimastos, 2003, Gender differences in large artery stiffness pre-and post puberty, J. Clin. Endocrinol. Metab. source

In the paper

“For respiratory measurements, Nilsson et al. have reported that respiratory synchronous variation in the PPG signal is irrespective of gender, and thus no action is required [111].”

What the sources say

• Not checked yetcites [111] Nilsson, 2006, Age and gender do not influence the ability to detect respiration by photoplethysmography, J. Clin. Monit. Comput. source

In the paper

“Dehghanojamahalleh et al. demonstrated variations in PPG morphology attributed to gender differences [105]. Interestingly, direct influence of gender variation was only significantly different at the upper peripheral measurement site such as the hand and fingers as opposed to lower extremities such as the ankle and feet.”

What the sources say

• Not checked yetcites [105] Dehghanojamahalleh, 2019, Sex-Related Differences in Photoplethysmography Signals Measured From Finger and Toe, IEEE J. Transl. Eng. Health Med. source

In the paper

“However, for females, PWV stayed roughly the same, while there was an increase in PWV for males.”

What the sources say

• Not checked yetcites [104] Ahimastos, 2003, Gender differences in large artery stiffness pre-and post puberty, J. Clin. Endocrinol. Metab. source

In the paper

“It was discovered during prepuberty, females have lower arterial compliance and higher central and peripheral PWV yielding stiffer arteries compared to males.”

What the sources say

• Not checked yetcites [104] Ahimastos, 2003, Gender differences in large artery stiffness pre-and post puberty, J. Clin. Endocrinol. Metab. source

In the paper

“The finger is arterially dominated while the forearm has a larger venous component [120].”

What the sources say

• Not checked yetcites [120] Ysehak Abay, T., Shafqat, K., and Kyriacou, P.A. (2019). Perfusion Changes at the Forehead Measured by Photoplethysmography during a Head-Down Tilt Protocol. Biosensors, 9. source

In the paper

“Pulse wave propagation delay between the genders indicates baseline differences in arterial stiffness with women displaying lower degrees of pulse arrival latency delay indicating higher vascular stiffness [106,107].”

What the sources say

• Not checked yetcites [106] Mukkamala, 2015, Toward ubiquitous blood pressure monitoring via pulse transit time: Theory and practice, IEEE Trans. Biomed. Eng. source
• Not checked yetcites [107] Guajardo, 2018, Sex differences in vascular dysfunction and cardiovascular outcomes: The cardiac, endothelial function, and arterial stiffness in ESRD (CERES) study, Hemodial. Int. source

In the paper

“The venous system similarly contributes periodicity to the PPG signal, but this is often considered noise.”

What the sources say

• Not checked yetcites [116] Shelley, 1993, The detection of peripheral venous pulsation using the pulse oximeter as a plethysmograph, J. Clin. Monit. source

In the paper

“Shelley et al. and Sami et al. both demonstrated the influence of the venous signal on SpO2, showing that forehead and earlobe pulse oximeters (respectively) produced a significantly biased variable component of the PPG determined to originate from central venous pressure (CVP) [125,127].”

What the sources say

• Not checked yetcites [125] Shelley, 2005, The effect of venous pulsation on the forehead pulse oximeter wave form as a possible source of error in Spo2 calculation, Anesth. Analg. source
• Not checked yetcites [127] Sami, 1991, Central venous pulsations associated with a falsely low oxygen saturation measured by pulse oximetry, J. Clin. Monit. source

In the paper

“They have been shown to artificially inflate the systolic peak amplitude and interfere with the rising systolic edge of the typical PPG waveform; an important parameter used in cardiovascular health determination [116].”

What the sources say

• Not checked yetcites [116] Shelley, 1993, The detection of peripheral venous pulsation using the pulse oximeter as a plethysmograph, J. Clin. Monit. source

In the paper

“Nijland et al. also confirmed these results in fetal lambs, where reflectance pulse oximetry SpO2 readings were significantly lower than fiberoptic SaO2 values. However, when the vessel was coagulated, this difference became negligible [128].”

What the sources say

• Not checked yetcites [128] Nijland, 1996, Reflectance pulse oximetry in fetal lambs: Subcutaneous vessels and vasoconstriction affect its reliability, J. Clin. Monit. source

In the paper

“Dicrotic notch duration at the finger was not significantly different from measurements under the wrist and forehead, but was different from the earlobe, arm, and wrist.”

What the sources say

• Not checked yetcites [95] Hartmann, 2019, Quantitative comparison of photoplethysmographic waveform characteristics: Effect of measurement site, Front. Physiol. source

In the paper

“However, it is possible that this will eliminate the arterial pulsation in hypotensive patients as well as increase pressure-induced injury risk to the measurement site [125].”

What the sources say

• Not checked yetcites [125] Shelley, 2005, The effect of venous pulsation on the forehead pulse oximeter wave form as a possible source of error in Spo2 calculation, Anesth. Analg. source

In the paper

“For silicon photodiodes, the absorption coefficient increases with temperature, and thus the detectors will absorb less light at higher temperatures, inducing an apparent shift in PPG amplitude [161].”

What the sources say

• Not checked yetcites [161] Optoelectronics, O. (2020, August 06). Temperature and Bias Effects. Available online: http://www.osioptoelectronics.com/technology-corner/frequently-asked-questions/temperature-bias-effects.aspx.

In the paper

“The back of the ear and earlobe yielded significantly higher PPG signal amplitude than the hand and postauricular region (i.e., neck behind the ear), but LDV values and comparisons to the fingertips and the rest of the face were not significant.”

What the sources say

• Not checked yetcites [144] Tur, 1983, Basal perfusion of the cutaneous microcirculation: Measurements as a function of anatomic position, J. Investig. Dermatol. source

In the paper

“Overall, the literature has shown that the fingertips have the thinnest skin, followed by the forearm, dorsal hand, cheek, and forehead [139].”

What the sources say

• Not checked yetcites [139] Dahlqvist, 2018, Vibrotactile and thermal perception and its relation to finger skin thickness, Clin. Neurophysiol. Pract. source

In the paper

“They reported a 2.6-times increase in PPG amplitude, 93% decrease in LDF flux, and median decrease of 0.089 degrees Celsius [150].”

What the sources say

• Not checked yetcites [150] Allen, 2002, Microvascular blood flow and skin temperature changes in the fingers following a deep inspiratory gasp, Physiol. Meas. source

In the paper

“All these motions range between 0.1 and 20 Hz, which is within the frequency range for heart rate (1–4 Hz) [163].”

What the sources say

• Not checked yetcites [163] Zong, C., and Jafari, R. (2015, January 25–29). Robust heart rate estimation using wrist-based PPG signals in the presence of intense physical activities. Proceedings of the 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, Italy. source

In the paper

“Fallet et al. found via iPPG that for all wavelengths of light, the forehead is superior to the cheeks and the whole face for determining heart rate.”

What the sources say

• Not checked yetcites [146] Fallet, S., Moser, V., Braun, F., and Vesin, J.-M. (2016, January 11–14). Imaging photoplethysmography: What are the best locations on the face to estimate heart rate?. Proceedings of the 2016 Computing in Cardiology Conference (CinC), Vancouver, BC, Canada. source

In the paper

“Adjusting positions or tapping a finger can be considered a micro-motion, while a macro-motion is performing an exercise such as walking or jogging—both types of motion cause more significant fluctuations to the signal than sedentary motions.”

In the paper

“Grabovskis et al. noted that the amount of pressure applied is often underreported and can manipulate the resultant PPG waveform [132].”

What the sources say

• Not checked yetcites [132] Grabovskis, 2013, Effect of probe contact pressure on the photoplethysmographic assessment of conduit artery stiffness, J. Biomed. Opt. source

In the paper

“Additionally, PTT is also similar bilaterally at the ears and finger, while a small difference exists at the toes [91].”

What the sources say

• Not checked yetcites [91] Allen, 2002, Age-related changes in peripheral pulse timing characteristics at the ears, fingers and toes, J. Hum. Hypertens. source

In the paper

“Thus, while physiological temperature changes due to thermoregulation may impact a signal, studies detailing changes caused by external temperatures also illustrate significant PPG impact.”

In the paper

“The authors conclude that PPG quality is improved with warm temperatures, yet do not mention the effect of thermoregulatory responses [155].”

What the sources say

• Not checked yetcites [154] Khan, 2015, Investigating the effects of temperature on photoplethysmography, IFAC Pap. Online source

In the paper

“With the addition of motion artifacts within a PPG signal, results for SpO2, heart rate, and other PPG dependent measurements can be skewed. This can create false alarms or inaccurate readings.”

In the paper

“Additionally, the authors noted a consistent increase in signal amplitude in the wrist PPG compared to the finger PPG.”

What the sources say

• Not checked yetcites [26] Rajala, 2018, Comparison of photoplethysmogram measured from wrist and finger and the effect of measurement location on pulse arrival time, Physiol. Meas. source

In the paper

“They determine that the appropriate amount of pressure is such that the b/a ratio of the SDPPG is equal to 0.70. This means that the external pressure on the arterial wall is equal to the mean arterial pressure [132].”

What the sources say

• Not checked yetcites [132] Grabovskis, 2013, Effect of probe contact pressure on the photoplethysmographic assessment of conduit artery stiffness, J. Biomed. Opt. source

In the paper

“It was determined that green was the best option due to its higher correlation coefficient, but blue is an alternative option due to its significantly similar ΔSNR as green.”

What the sources say

• Not checked yetcites [164] Lee, J., Matsumura, K., Yamakoshi, K.-I., Rolfe, P., Tanaka, S., and Yamakoshi, T. (2013, January 3–7). Comparison between red, green and blue light reflection photoplethysmography for heart rate monitoring during motion. Proceedings of the 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, Japan.

In the paper

“With their longer wavelengths, red and IR have deeper penetration depths compared to blue and green, making them prone to more motion artifacts such as the factors that directly affect the DC baseline [162,164].”

What the sources say

• Not checked yetcites [162] Maeda, 2011, Relationship between measurement site and motion artifacts in wearable reflected photoplethysmography, J. Med. Syst. source
• Not checked yetcites [164] Lee, J., Matsumura, K., Yamakoshi, K.-I., Rolfe, P., Tanaka, S., and Yamakoshi, T. (2013, January 3–7). Comparison between red, green and blue light reflection photoplethysmography for heart rate monitoring during motion. Proceedings of the 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Osaka, Japan.

In the paper

“In another study, performed by Dooley et al., the comparison was made between the Apple Watch (AP), FB, and Garmin Forerunner 225 (GF), where the AP was considered the best in terms of MAPE (1.14–6.70%), while the FB and the GF had a MAPE of 2.38–16.99% and 7.87–24.38%, respectively [167].”

What the sources say

• Not checked yetcites [167] Dooley, 2017, Estimating accuracy at exercise intensities: A comparative study of self-monitoring heart rate and physical activity wearable devices, JMIR mHealth uHealth source

In the paper

“The AP (p = 0.84) and GF (p = 0.35) had no significant difference in HR readings from the ECG during vigorous intensity.”

What the sources say

• Not checked yetcites [167] Dooley, 2017, Estimating accuracy at exercise intensities: A comparative study of self-monitoring heart rate and physical activity wearable devices, JMIR mHealth uHealth source

In the paper

“They found their algorithm to have an accuracy of 83 ± 11%, a sensitivity of 89 ± 10 %, and a specificity of 77 ± 19%.”

What the sources say

• Not checked yetcites [169] Sukor, 2011, Signal quality measures for pulse oximetry through waveform morphology analysis, Physiol. Meas. source

In the paper

“When comparing their algorithm to an ECG, they were able to lower the error in heart rate readings to 0.49 ± 0.66 bpm, whereas without the algorithm, the error was 7.23 ± 5.78 bpm.”

What the sources say

• Not checked yetcites [169] Sukor, 2011, Signal quality measures for pulse oximetry through waveform morphology analysis, Physiol. Meas. source

In the paper

“This shows that, depending on the exercise, each tracker will have a different accuracy, which could be attributed to the different noise-canceling algorithms used on board.”

In the paper

“This study also showed that the Apple Watch 4 had the highest accuracy with an MAE of 2.7 BPM at rest and 4.6 BPM during the walking phase,”

What the sources say

• Not checked yetcites [46] Bent, 2020, Investigating sources of inaccuracy in wearable optical heart rate sensors, NPJ Digit. Med. source

In the paper

“During the low-intensity events, the BP provided accurate heart rate readings (r = 0.84, p < 0.05), while the FB had a moderate correlation (r = 0.73, p < 0.05).”

What the sources say

• Not checked yetcites [165] Jo, 2016, Validation of biofeedback wearables for photoplethysmographic heart rate tracking, J. Sports Sci. Med.

In the paper

“When going from a resting phase to a low-intensity walking phase, they found that the correlation coefficient went down but had slowly rebounded when moved into a more intense walking phase.”

What the sources say

• Not checked yetcites [166] Stahl, 2016, How accurate are the wrist-based heart rate monitors during walking and running activities? Are they accurate enough?, BMJ Open Sport Exerc. Med. source

In the paper

“High-intensity events caused the BP to have a weaker correlation (r = 0.77, p < 0.05), whereas the FB had the weakest correlation (r = 0.58, p < 0.05).”

What the sources say

• Not checked yetcites [165] Jo, 2016, Validation of biofeedback wearables for photoplethysmographic heart rate tracking, J. Sports Sci. Med.

In the paper

“Aside from location and wavelength, the intensity of motion also plays a role in how accurate heart rate monitoring is.”

In the paper

“These results align with an assessment done by Stahl et al., where they measured the mean absolute percentage error values (MAPE) of BP and the FB to be 3.6% and 6.2%, respectively [166].”

What the sources say

• Not checked yetcites [166] Stahl, 2016, How accurate are the wrist-based heart rate monitors during walking and running activities? Are they accurate enough?, BMJ Open Sport Exerc. Med. source

In the paper

“Accelerometers and gyroscopes are the standards, but when trying to measure micro motions such as finger tapping, accelerometers and gyroscopes do not always pick up on the motion because the wrist remains still.”

What the sources say

• Not checked yetcites [179] Casson, 2016, Gyroscope vs. accelerometer measurements of motion from wrist PPG during physical exercise, ICT Express source
• Not checked yetcites [180] Lee, 2018, Motion artifact cancellation in wearable photoplethysmography using gyroscope, IEEE Sens. J. source

In the paper

“They were able to achieve an accuracy of 93.7% during a daily motion experiment, and they saw a significant reduction in error for SpO2 and heart rate measurements.”

What the sources say

• Not checked yetcites [170] Chong, 2014, Photoplethysmograph signal reconstruction based on a novel hybrid motion artifact detection–reduction approach. Part I: Motion and noise artifact detection, Ann. Biomed. Eng. source

In the paper

“The discussed detection methods are all real-time ways to detect and eliminate sections of acquired PPG signals that contain corrupted areas due to motion artifacts.”

In the paper

“Kurtosis and Shannon Entropy were the measurements used by Selvaraj et al. to determine motion artifacts, and they were able to perform an accuracy of 88.8% in a laboratory setting [174].”

What the sources say

• Not checked yetcites [174] Selvaraj, N., Mendelson, Y., Shelley, K.H., Silverman, D.G., and Chon, K.H. (September, January 30). Statistical approach for the detection of motion/noise artifacts in Photoplethysmogram. Proceedings of the 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Boston, MA, USA. source

In the paper

“With the use of adaptive filters and secondary sensors, motion artifacts can be identified and eliminated, or they can be identified and reconstructed to restore the original PPG signal, void of all motion artifacts.”

In the paper

“They discovered skew and kurtosis measurements in the time domain to be higher when there was corruption, and the frequency-domain kurtosis was smaller when there was corruption.”

What the sources say

• Not checked yetcites [173] Krishnan, R., Natarajan, B., and Warren, S. (April, January 31). Analysis and detection of motion artifact in photoplethysmographic data using higher order statistics. Proceedings of the 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, Las Vegas, NV, USA. source

In the paper

“They were able to achieve a specificity of 92.7% and a sensitivity of 82.7%.”

What the sources say

• Not checked yetcites [172] Couceiro, R., Carvalho, P., Paiva, R.P., Henriques, J., and Muehlsteff, J. (September, January 28). Detection of motion artifacts in photoplethysmographic signals based on time and period domain analysis. Proceedings of the 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA. source

In the paper

“They were able to achieve a positive predictive value of 96.68% and a sensitivity of 98.93%.”

What the sources say

• Not checked yetcites [171] Karlen, W., Ansermino, J.M., and Dumont, G. (September, January 28). Adaptive pulse segmentation and artifact detection in photoplethysmography for mobile applications. Proceedings of the 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, San Diego, CA, USA. source

In the paper

“Both the piezoelectric transducer and another wavelength of light can pick up these finer motions [181,182].”

What the sources say

• Not checked yetcites [181] Lee, 2019, Motion artifact identification and removal from wearable reflectance photoplethysmography using piezoelectric transducer, IEEE Sens. J. source
• Not checked yetcites [182] Zhang, Y., Song, S., Vullings, R., Biswas, D., Simões-Capela, N., van Helleputte, N., van Hoof, C., and Groenendaal, W. (2019). Motion artifact reduction for wrist-worn photoplethysmograph sensors based on different wavelengths. Sensors, 19. source

In the paper

“Without force regulation, the coefficient of variation between five posture stages was calculated at 50.9%. The addition of the force regulator dropped the coefficient of variation to 1.8%.”

What the sources say

• Not checked yetcites [212] Sim, 2018, A contact-force regulated photoplethysmography (PPG) platform, AIP Adv. source

In the paper

“Grabovskis et al. reported with an applied pressure range of 0–15 kPa, there was a variation of over 300 percent in the b/a ratio [209]. When an optimal pressure was reached, the variation coefficient dropped to less than 5 percent.”

What the sources say

• Not checked yetcites [209] Grabovskis, A., Marcinkevics, Z., Rubenis, O., Rubins, U., and Lusa, V. (2012, January 16–19). Photoplethysmography system for blood pulsation detection in unloaded artery conditions. Proceedings of the Biophotonics: Photonic Solutions for Better Health Care III, Brussels, Belgium. source

In the paper

“For shorter wavelengths, the occlusion pressure was lower, as more superficial arterioles with low MAP were in the light path at these wavelengths.”

What the sources say

• Not checked yetcites [210] Spigulis, J., Gailite, L., Erts, R., and Lihachev, A. (2007, January 17–19). Contact probe pressure effects in skin multi-spectral photoplethysmography. Proceedings of the European Conference on Biomedical Optics, Munich, Germany. source

In the paper

“The change in these fundamental features can affect other measurements such as the b/a ratio, derived from the a and b peaks of the second derivative of a PPG signal, used to characterize arterial stiffness [132], the relationship between the frequency response of a PPG signal and a blood pressure signal [207], or the pulse transit time (PTT) which is used as an indicator for fluctuating stiffness”

What the sources say

• Not checked yetcites [132] Grabovskis, 2013, Effect of probe contact pressure on the photoplethysmographic assessment of conduit artery stiffness, J. Biomed. Opt. source
• Not checked yetcites [207] Hsiu, 2011, Effects of different contacting pressure on the transfer function between finger photoplethysmographic and radial blood pressure waveforms, Proc. Inst. Mech. Eng. Part H J. Eng. Med. source

In the paper

“As the external pressure exceeds the point where the transmural pressure is no longer zero, the vessel begins to occlude, and the AC amplitude begins to decrease until there is no longer a signal where the vessel is occluded.”

In the paper

“At 60 mmHg, there were the greatest R2 values for four out of five harmonics, and with the different applied pressures, the R2 values ranged from 0.13 to 0.77 for the first harmonic.”

What the sources say

• Not checked yetcites [207] Hsiu, 2011, Effects of different contacting pressure on the transfer function between finger photoplethysmographic and radial blood pressure waveforms, Proc. Inst. Mech. Eng. Part H J. Eng. Med. source

In the paper

“As reported by Teng and Zhang, with an applied force at the finger of 0.2–1.8 N, there was an increase in AC amplitude and AC/DC ratio until the parameters reached a peak and then decreased, while the pulse area decreased, and the DC offset increased [206].”

What the sources say

• Not checked yetcites [206] Teng, 2004, The effect of contacting force on photoplethysmographic signals, Physiol. Meas. source

In the paper

“Lastly, for the PTT calculated at 90% of the pulse amplitude, there was a similar increasing trend for the PTT but without a significant difference (p = 0.107). The PTT leveled out to a constant value for a transmural pressure of around 0.1 N.”

What the sources say

• Not checked yetcites [208] Teng, 2006, The effect of applied sensor contact force on pulse transit time, Physiol. Meas. source

In the paper

“This band would apply a skin pressure of about 75 mmHg and, due to its compliance, it would hold steady the applied pressure while the finger was in different positions.”

What the sources say

• Not checked yetcites [213] Rhee, 2001, Artifact-resistant power-efficient design of finger-ring plethysmographic sensors, IEEE Trans. Biomed. Eng. source

In the paper

“Various attempts are also made in the PPG sensing circuitry level to effectively reject DC photocurrent.”

In the paper

“PPG signals are low-amplitude signals and have a normal pulse frequency within the range of 0.5 to 4 Hz [195,196].”

What the sources say

• Not checked yetcites [195] Wong, 2005, A near-infrared heart rate measurement IC with very low cutoff frequency using current steering technique, IEEE Trans. Circuits Syst. I Regul. Pap. source
• Not checked yetcites [196] Moraes, J.L., Rocha, M.X., Vasconcelos, G.G., Vasconcelos Filho, J.E., de Albuquerque, V.H.C., and Alexandria, A.R. (2018). Advances in photopletysmography signal analysis for biomedical applications. Sensors, 18. source

In the paper

“With the change in transmural pressure, there is also a related compliance change in the affected vessel. The vessel reaches maximum compliance when the transmural pressure is zero [205].”

What the sources say

• Not checked yetcites [205] Forster, 1986, Oscillometric determination of diastolic, mean and systolic blood pressure—A numerical model, J. Biomech. Eng. source

In the paper

“With a low bandwidth of around 5 Hz, PPG signals can be extracted out from high frequency noise using a low-pass filter and from low frequency noise using a high-pass filter (e.g., to reject 60 Hz room lights and to reject DC sunlight).”

In the paper

“This method was found to be effective in shielding from noise due to ambient light, as the DC level of the PPG signal was considerably low (measured to be less than 2 nA) when the LEDs were switched off.”

What the sources say

• Not checked yetcites [200] Wang, 2007, Multichannel reflective PPG earpiece sensor with passive motion cancellation, IEEE Trans. Biomed. Circuits Syst. source

In the paper

“Several characteristics of the PPG waveform are affected by changes in contact force, including; the AC amplitude, the DC offset amplitude, the ratio of AC/DC amplitudes, and the normalized pulse area.”