Www.dovepress.com



Supplementary dataInfluence of apnea-hypopnea index (AHI) on performanceThe apnea-hypopnea index (AHI) is an index used to indicate the severity of obstructive sleep apnea (OSA). Individual apneic events have impact on hemodynamics and it has been demonstrated that repetitive episodes of apneas trigger fluctuations in heart rate with consequent effects on the estimates of cardiovascular variability. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"qJIFHGdP","properties":{"formattedCitation":"\\super 1\\nosupersub{}","plainCitation":"1","noteIndex":0},"citationItems":[{"id":194,"uris":[""],"uri":[""],"itemData":{"id":194,"type":"article-journal","container-title":"The Journal of Clinical Investigation","DOI":"10.1172/JCI118235","ISSN":"0021-9738","issue":"4","journalAbbreviation":"J Clin Invest","language":"en","note":"publisher: American Society for Clinical Investigation\nPMID: 7560081","page":"1897-1904","source":"","title":"Sympathetic neural mechanisms in obstructive sleep apnea.","volume":"96","author":[{"family":"Somers","given":"V. K."},{"family":"Dyken","given":"M. E."},{"family":"Clary","given":"M. P."},{"family":"Abboud","given":"F. M."}],"issued":{"date-parts":[["1995",10,1]]}}}],"schema":""} 1 Our results show a weak Spearman’s rank correlation between AHI and κ (ρ = -0.21, p=0.001), suggesting a lower performance as AHI increases. No significant correlation was found between AHI and accuracy. These results demonstrate only a minor effect of AHI on performance, indicating that our classifier can be used in patients with different severities of OSA without a priori knowledge of their condition.Performance in children/adolescentsFor the children/adolescents, the algorithm achieved an average κ of 0.66 ± 0.09 and accuracy of 77.9% ± 6.5% (Table S1) for 4-class sleep staging, which is higher compared to the performance in the entire cohort.Table S SEQ Supplementary_Table_S \* ARABIC 1 Epoch-per-epoch agreement for children/adolescentsTaskκ (-)Accuracy (%)Sensitivity (%)Specificity (%)PPV (%)Wake/N1+N2/N3/REM0.66 ± 0.0977.9 ± 6.5n/an/an/aWake/NREM/REM0.71 ± 0.1087.3 ± 3.7 n/an/an/aWake (vs. Sleep)0.66 ± 0.1693.7 ± 2.8 74.8 ± 17.3 95.5 ± 3.268.5 ± 19.1N1+N20.56 ± 0.1278.5 ± 6.0 78.0 ± 7.3 79.1 ± 9.577.2 ± 13.7 N30.70 ± 0.1390.4 ± 5.074.4 ± 18.195.3 ± 4.2 81.9 ± 13.1 REM0.73 ± 0.1393.1 ± 3.480.9 ± 13.495.3 ± 3.274.8 ± 13.8 κ: Kappa; PPV: positive predictive valuePerformance in patients with Non-REM parasomnia A detailed look at the classification for Non-REM parasomnias showed an increased performance compared with the overall performance for four- and three-class classification (Table S2). This was associated with an increase for almost all metrics for all classes, and particularly with an increase in κ for N3 (from 0.58 in the entire cohort to 0.70 for patients with a Non-REM parasomnia). It could be that no parasomnia event occurred during the recording night, and that these patients had a relatively normal sleep pattern, comparable to a healthy person of the same age, which makes it possibly easier to perform sleep staging. It is also noteworthy that the patient in the non-REM parasomnia group were younger compared to the overall validation dataset, with an average age of 24.8 ranging from 3 to 54 years. Table S SEQ Supplementary_Table_S \* ARABIC 2 Performance for patients with Non-REM parasomniaTaskκ (-)Accuracy (%)Sensitivity (%)Specificity (%)PPV (%)Wake/N1+N2/N3/REM0.69 ± 0.0780.1 ± 4.3n/an/an/aWake/NREM/REM0.74 ± 0.0688.5 ± 2.4 n/an/an/aWake (vs. Sleep)0.69 ± 0.1194.9 ± 2.6 75.9 ± 14.5 96.4 ± 3.771.7 ± 13.5N1+N20.61 ± 0.0880.6 ± 4.1 79.7 ± 7.881.4 ± 4.881.9 ± 5.4 N30.70 ± 0.1191.5 ± 3.475.8 ± 16.0 95.0 ± 4.279.4 ± 14.5 REM0.77 ± 0.0693.2 ± 2.483.6 ± 8.995.1 ± 2.579.1 ± 8.3 κ: Kappa; PPV: positive predictive valuePerformance in patients with REM-parasomnia The performance of the algorithm was lower in patients with REM parasomnias. There was a decrease for all metrics, but the most negatively impacted class was REM, with a substantial decrease in sensitivity and κ ( REF _Ref51676000 \h \* MERGEFORMAT Table S3). A potential reason for this is the presence of comorbid autonomic dysfunction, as explained by Fonseca et al. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"8oUvuTHb","properties":{"formattedCitation":"\\super 2\\nosupersub{}","plainCitation":"2","noteIndex":0},"citationItems":[{"id":101,"uris":[""],"uri":[""],"itemData":{"id":101,"type":"article-journal","container-title":"Sleep","DOI":"10.1093/sleep/zsaa048","title":"Automatic sleep staging using heart rate variability, body movements and recurrent neural networks in a sleep disordered population","author":[{"family":"Fonseca","given":"Pedro"},{"family":"Gilst","given":"Merel M.","dropping-particle":"van"},{"family":"Radha","given":"Mustafa"},{"family":"Ross","given":"Marco"},{"family":"Moreau","given":"Arnaud"},{"family":"Cerny","given":"Andreas"},{"family":"Anderer","given":"Peter"},{"family":"Dijk","given":"Johannes P.","dropping-particle":"van"},{"family":"Overeem","given":"Sebastiaan"}],"issued":{"date-parts":[["2020"]]}}}],"schema":""} 2 Another reason might be the fact that human inter-rater-agreement is lower for REM sleep and wake in patients with a REM behavior disorder, suggesting less reliable comparison with ground-truth PSG. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"lPc79Mzn","properties":{"formattedCitation":"\\super 3\\nosupersub{}","plainCitation":"3","noteIndex":0},"citationItems":[{"id":188,"uris":[""],"uri":[""],"itemData":{"id":188,"type":"article-journal","abstract":"Abstract:. Recently described functional connections between basal ganglia and brainstem circuits provide a neurobiologic basis for the absence of REM sleep at","container-title":"Sleep","DOI":"10.1093/sleep/23.5.1j","ISSN":"0161-8105","issue":"5","journalAbbreviation":"Sleep","language":"en","note":"publisher: Oxford Academic","page":"1-6","source":"academic.","title":"Inter-rater Reliability for Identification of REM Sleep in Parkinson's Disease","volume":"23","author":[{"family":"Bliwise","given":"Donald L."},{"family":"Williams","given":"Monique L."},{"family":"Irbe","given":"Dainis"},{"family":"Ansari","given":"Farzaneh Pour"},{"family":"Rye","given":"David B."}],"issued":{"date-parts":[["2000",8,1]]}}}],"schema":""} 3Table S SEQ Supplementary_Table_S \* ARABIC 3 Confusion matrix for patients with REM parasomniaPred. Ref. ↓WakeN1+N2N3REMPrev. (%)Sens. (%)κ (-)Wake3030 (18.2%/69.3%)1164 (7.0%/26.6%)17 (0.1%/0.4%)160 (1.0%/3.7%)4371 (26.3)69.30.65N1+N2713 (4.3%/8.6%)6444 (38.8%/77.5%)694 (4.2%/8.4%)460 (2.8%/5.5%)8311 (50.0)77.50.49N336 (0.2%/1.7%)751 (4.5%/36.1%)1288 (7.8%/62.0%)3 (0.02%/0.1%)2078 (12.5)62.00.58REM107 (0.6%/5.8%)470 (2.8%/25.4%)3 (0.02%/0.2%)1271 (7.7%/68.7%)1851 (11.1)68.70.64PPV(%)78.0%73.0%64.3%67.1%Each cell contains the number of epochs for all classes according to the sleep stages identified by the algorithm (“Pred.”) and ground truth PSG (“Ref.”). In addition, between parentheses, the percentage relative to the total number of epochs of all classes is listed, followed by the percentage relative to the total number of epochs with the corresponding reference sleep stage, as estimated with PSG, for that row.Prev.: prevalence; Sens.: sensitivity; PPV: positive predictive valuePerformance in patients with insomnia The ability to accurately detect wake is important in patients with sleep disorders, especially in patients with insomnia. The use of actigraphy in this group is problematic due to its main limitation: its low ability to correctly detect wake (low sensitivity to wake or low specificity to sleep). Patients with insomnia often lie awake in bed motionless, where the classifier is being “fooled” and identifies motionless periods incorrectly as sleep. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"sfWBdBab","properties":{"formattedCitation":"\\super 4\\uc0\\u8211{}6\\nosupersub{}","plainCitation":"4–6","noteIndex":0},"citationItems":[{"id":201,"uris":[""],"uri":[""],"itemData":{"id":201,"type":"article-journal","abstract":"To assess the use of actigraphy in evaluating insomnia, 36 patients with a serious complaint of insomnia slept 3 nights each in the laboratory, where the usual polysomnograms (PSGs) were obtained as well as actigraphic assessments of their sleep. Patients also wore actigraphs for 7 days at home, were extensively interviewed and filled out psychometric tests. Based on all this information, the patients were then diagnosed according to the International Classification of Sleep Disorders. Averaged over the 3 nights for each insomniac, the mean discrepancy between actigram and PSG was 49 minutes per night. In three-fourths of the cases, actigram and PSG agreed to within 1 hour on the total amount of sleep per night. Discrepancies, however, were not random: In patients with psychophysiologic insomnia and in insomnia associated with psychiatric disease, the actigram typically overestimated sleep when compared with the PSG. In patients with sleep-state misperception, the actigram was either quite accurate or it underestimated sleep when compared with the PSG. Comparing laboratory with home sleep, one-third of all insomniacs slept better in the laboratory and two-thirds slept better at home. In addition, night-by-night variability was higher at home than in the laboratory. Based on our study, we now recommend actigraphy as an additional tool in the clinical evaluation of insomnia, but we believe that in complex cases it should be combined with 1 PSG night in the sleep disorders center.","container-title":"Sleep","DOI":"10.1093/sleep/15.4.293","ISSN":"0161-8105","issue":"4","journalAbbreviation":"Sleep","language":"eng","note":"PMID: 1519002","page":"293-301","source":"PubMed","title":"Wrist actigraphy in insomnia","volume":"15","author":[{"family":"Hauri","given":"P. J."},{"family":"Wisbey","given":"J."}],"issued":{"date-parts":[["1992",8]]}}},{"id":223,"uris":[""],"uri":[""],"itemData":{"id":223,"type":"article-journal","abstract":"Summary:. This paper, which has been reviewed and approved by the Board of Directors of the American Sleep Disorders Association, provides the background for t","container-title":"Sleep","DOI":"10.1093/sleep/18.4.288","ISSN":"0161-8105","issue":"4","journalAbbreviation":"Sleep","language":"en","note":"publisher: Oxford Academic","page":"288-302","source":"academic.","title":"The Role of Actigraphy in the Evaluation of Sleep Disorders","volume":"18","author":[{"family":"Sadeh","given":"Avi"},{"family":"Hauri","given":"Peter J."},{"family":"Kripke","given":"Daniel F."},{"family":"Lavie","given":"Peretz"}],"issued":{"date-parts":[["1995",6,1]]}}},{"id":261,"uris":[""],"uri":[""],"itemData":{"id":261,"type":"paper-conference","abstract":"This paper presents an actigraphy-based approach for sleep/wake detection for insomniacs. Due to its relative unobtrusiveness, actigraphy is often used to estimate overnight sleep-wake patterns in clinical practice. However, its performance has been shown to be limited in subjects with sleep complaints such as insomniacs. Quantifying activity counts on 30-s epoch basis, as usually done in regular actigraphy, may lead to an underestimation of wake periods where the subject shows reduced body movements. We therefore propose a new actigraphic feature to characterize the `possibility' of epochs being asleep (or awake) before or after its nearest epoch with a very high activity levels. It is expected to correctly identify some wake epochs when they are very close to the high activity epochs, although they can be motionless. A data set containing 25 insomnia subjects and a linear discriminant classifier were used to test our approach in this study. Leave-one-subject-out cross validation results show that combining the new and the traditional actigraphic features led to a markedly improved performance in sleep/wake detection compared to that using the traditional feature only, with an increase in Cohen's kappa from 0.49 to 0.55.","container-title":"2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","DOI":"10.1109/BSN.2017.7935711","event":"2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN)","note":"ISSN: 2376-8894","page":"1-4","source":"IEEE Xplore","title":"Actigraphy-based sleep/wake detection for insomniacs","author":[{"family":"Long","given":"X."},{"family":"Fonseca","given":"P."},{"family":"Haakma","given":"R."},{"family":"Aarts","given":"R. M."}],"issued":{"date-parts":[["2017",5]]}}}],"schema":""} 4–6 Taking a look at recent published literature (see REF _Ref54857181 \h \* MERGEFORMAT Table S5), Kahawage et al. validated both actigraphy and a consumer wearable in patients with insomnia. They reported an accuracy to detect wake of 91.0% and 81.1%, and specificity to detect sleep (or sensitivity to detect wake) of 39.09% and 44.76% for actigraphy and the consumer wearable, respectively. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"meGWRMcG","properties":{"formattedCitation":"\\super 7\\nosupersub{}","plainCitation":"7","noteIndex":0},"citationItems":[{"id":181,"uris":[""],"uri":[""],"itemData":{"id":181,"type":"article-journal","abstract":"Consumer activity trackers claiming to measure sleep/wake patterns are ubiquitous within clinical and consumer settings. However, validation of these devices in sleep disorder populations are lacking. We examined 1?night of sleep in 42 individuals with insomnia (mean = 49.14 ± 17.54 years) using polysomnography, a wrist actigraph (Actiwatch Spectrum Pro: AWS) and a consumer activity tracker (Fitbit Alta HR: FBA). Epoch‐by‐epoch analysis and Bland?Altman methods evaluated each device against polysomnography for sleep/wake detection, total sleep time, sleep efficiency, wake after sleep onset and sleep latency. FBA sleep stage classification of light sleep (N1 + N2), deep sleep (N3) and rapid eye movement was also compared with polysomnography. Compared with polysomnography, both activity trackers displayed high accuracy (81.12% versus 82.80%, AWS and FBA respectively; ns) and sensitivity (sleep detection; 96.66% versus 96.04%, respectively; ns) but low specificity (wake detection; 39.09% versus 44.76%, respectively; p = .037). Both trackers overestimated total sleep time and sleep efficiency, and underestimated sleep latency and wake after sleep onset. FBA demonstrated sleep stage sensitivity and specificity, respectively, of 79.39% and 58.77% (light), 49.04% and 95.54% (deep), 65.97% and 91.53% (rapid eye movement). Both devices were more accurate in detecting sleep than wake, with equivalent sensitivity, but statistically different specificity. FBA provided equivalent estimates as AWS for all traditional actigraphy sleep parameters. FBA also showed high specificity when identifying N3, and rapid eye movement, though sensitivity was modest. Thus, it underestimates these sleep stages and overestimates light sleep, demonstrating more shallow sleep than actually obtained. Whether FBA could serve as a low‐cost substitute for actigraphy in insomnia requires further investigation.","container-title":"Journal of Sleep Research","DOI":"10.1111/jsr.12931","ISSN":"0962-1105, 1365-2869","issue":"1","journalAbbreviation":"J Sleep Res","language":"en","source":" (Crossref)","title":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I: In‐lab validation against polysomnography","title-short":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I","URL":"","volume":"29","author":[{"family":"Kahawage","given":"Piyumi"},{"family":"Jumabhoy","given":"Ria"},{"family":"Hamill","given":"Kellie"},{"family":"Zambotti","given":"Massimiliano"},{"family":"Drummond","given":"Sean P. A."}],"accessed":{"date-parts":[["2020",10,20]]},"issued":{"date-parts":[["2020",2]]}}}],"schema":""} 7 A study by Marino at al. showed an accuracy of 83.3% and sensitivity to wake of 34.7% obtained with actigraphy in insomnia patients. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"YWN6o1E4","properties":{"formattedCitation":"\\super 8\\nosupersub{}","plainCitation":"8","noteIndex":0},"citationItems":[{"id":247,"uris":[""],"uri":[""],"itemData":{"id":247,"type":"article-journal","abstract":"We validated actigraphy for detecting sleep and wakefulness versus polysomnography (PSG).Actigraphy and polysomnography were simultaneously collected during sleep laboratory admissions. All studies involved 8.5 h time in bed, except for sleep restriction studies. Epochs (30-sec; n = 232,849) were characterized for sensitivity (actigraphy = sleep when PSG = sleep), specificity (actigraphy = wake when PSG = wake), and accuracy (total proportion correct); the amount of wakefulness after sleep onset (WASO) was also assessed. A generalized estimating equation (GEE) model included age, gender, insomnia diagnosis, and daytime/nighttime sleep timing factors.Controlled sleep laboratory conditions.Young and older adults, healthy or chronic primary insomniac (PI) patients, and daytime sleep of 23 night-workers (n = 77, age 35.0 ± 12.5, 30F, mean nights = 3.2).N/A.Overall, sensitivity (0.965) and accuracy (0.863) were high, whereas specificity (0.329) was low; each was only slightly modified by gender, insomnia, day/night sleep timing (magnitude of change &lt; 0.04). Increasing age slightly reduced specificity. Mean WASO/night was 49.1 min by PSG compared to 36.8 min/night by actigraphy (β = 0.81; CI = 0.42, 1.21), unbiased when WASO &lt; 30 min/night, and overestimated when WASO &gt; 30 min/night.This validation quantifies strengths and weaknesses of actigraphy as a tool measuring sleep in clinical and population studies. Overall, the participant-specific accuracy is relatively high, and for most participants, above 80%. We validate this finding across multiple nights and a variety of adults across much of the young to midlife years, in both men and women, in those with and without insomnia, and in 77 participants. We conclude that actigraphy is overall a useful and valid means for estimating total sleep time and wakefulness after sleep onset in field and workplace studies, with some limitations in specificity.","container-title":"Sleep","DOI":"10.5665/sleep.3142","ISSN":"0161-8105","issue":"11","journalAbbreviation":"Sleep","page":"1747-1755","source":"Silverchair","title":"Measuring Sleep: Accuracy, Sensitivity, and Specificity of Wrist Actigraphy Compared to Polysomnography","title-short":"Measuring Sleep","volume":"36","author":[{"family":"Marino","given":"Miguel"},{"family":"Li","given":"Yi"},{"family":"Rueschman","given":"Michael N."},{"family":"Winkelman","given":"J. W."},{"family":"Ellenbogen","given":"J. M."},{"family":"Solet","given":"J. M."},{"family":"Dulin","given":"Hilary"},{"family":"Berkman","given":"Lisa F."},{"family":"Buxton","given":"Orfeu M."}],"issued":{"date-parts":[["2013",11,1]]}}}],"schema":""} 8 The addition of PPG measurements on motion-based measurements enables us to derive HRV features that depend on autonomic activity. Since autonomic nervous system activity cannot be “fooled”, it will help the classifier to detect wake in these motionless periods. This important addition is emphasized in our study by a relatively high agreement on Wake/Sleep classification and the detection of wake with a sensitivity of 75.3% in patients with insomnia, as reported in REF _Ref54856763 \h \* MERGEFORMAT Table S4.Table S SEQ Supplementary_Table_S \* ARABIC 4 Performance for patients with insomniaTaskκ (-)Accuracy (%)Sensitivity (%)Specificity (%)Wake (vs. Sleep)0.68 ± 0.1291.0 ± 5.175.3 ± 16.494.0 ± 6.9N1+N20.55 ± 0.1278.1 ± 6.477.7 ± 9.477.9 ± 11.1N30.60 ± 0.2691.6 ± 5.072.8 ± 24.094.8 ± 5.3REM0.68 ± 0.1893.5 ± 3.176.5 ± 22.6 95.7 ± 3.1Overview of recent published literature on automatic sleep staging REF _Ref54857181 \h \* MERGEFORMAT Table S5 gives an overview of recently published literature on automatic sleep staging using different devices. We included studies using PPG signals and accelerometer data, similar to our work and studies using exclusively PPG signals. In addition, we included studies using clinically accepted actigraphy, which measures solely accelerometer data, and widely available sleep trackers used by a substantial amount of today’s population. Table S SEQ Supplementary_Table_S \* ARABIC 5 Overview of recently published research on sleep staging using photoplethysmography (PPG) data, actigraphy or consumer wearablesType of data/deviceYear of publicationAuthorsSample characteristicsType of sleep stagingResultsRaw PPG and accelerometer dataApple Watch2019Walch et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"44Q2HZj8","properties":{"formattedCitation":"\\super 9\\nosupersub{}","plainCitation":"9","noteIndex":0},"citationItems":[{"id":20,"uris":[""],"uri":[""],"itemData":{"id":20,"type":"article-journal","abstract":"Abstract. Wearable, multisensor, consumer devices that estimate sleep are now commonplace, but the algorithms used by these devices to score sleep are not open","container-title":"Sleep","DOI":"10.1093/sleep/zsz180","ISSN":"0161-8105","issue":"12","journalAbbreviation":"Sleep","language":"en","source":"academic.","title":"Sleep stage prediction with raw acceleration and photoplethysmography heart rate data derived from a consumer wearable device","URL":"","volume":"42","author":[{"family":"Walch","given":"Olivia"},{"family":"Huang","given":"Yitong"},{"family":"Forger","given":"Daniel"},{"family":"Goldstein","given":"Cathy"}],"accessed":{"date-parts":[["2020",1,27]]},"issued":{"date-parts":[["2019",12,24]]}}}],"schema":""} 9 39 healthy adultsAge: 29.42 ± 8.52 yearsRange 19.0 to 55.0Three classκ = 0.3Accuracy = 72%Two classAccuracy = 90%Specificity* = 59.6%Philips CE-marked logging device2017Fonseca et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"iiY7yV3i","properties":{"formattedCitation":"\\super 10\\nosupersub{}","plainCitation":"10","noteIndex":0},"citationItems":[{"id":230,"uris":[""],"uri":[""],"itemData":{"id":230,"type":"article-journal","abstract":"AbstractStudy Objectives:. To compare the accuracy of automatic sleep staging based on heart rate variability measured from photoplethysmography (PPG) combined","container-title":"Sleep","DOI":"10.1093/sleep/zsx097","ISSN":"0161-8105","issue":"7","journalAbbreviation":"Sleep","language":"en","note":"publisher: Oxford Academic","source":"academic.","title":"Validation of Photoplethysmography-Based Sleep Staging Compared With Polysomnography in Healthy Middle-Aged Adults","URL":"","volume":"40","author":[{"family":"Fonseca","given":"Pedro"},{"family":"Weysen","given":"Tim"},{"family":"Goelema","given":"Maaike S."},{"family":"M?st","given":"Els I. S."},{"family":"Radha","given":"Mustafa"},{"family":"Lunsingh Scheurleer","given":"Charlotte"},{"family":"Heuvel","given":"Leonie","non-dropping-particle":"van den"},{"family":"Aarts","given":"Ronald M."}],"accessed":{"date-parts":[["2020",11,11]]},"issued":{"date-parts":[["2017",7,1]]}}}],"schema":""} 10 51 healthy adultsAge: 51.6 ± 7.7 yearsRange 41 to 66Four classκ = 0.42Accuracy = 59.3%Three classκ = 0.46Accuracy = 72.9%Two classκ = 0.55Accuracy = 91.5%Fitbit Surge2017Beattie et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"H5HGbwi0","properties":{"formattedCitation":"\\super 11\\nosupersub{}","plainCitation":"11","noteIndex":0},"citationItems":[{"id":3,"uris":[""],"uri":[""],"itemData":{"id":3,"type":"article-journal","abstract":"Objective: This paper aims to report on the accuracy of estimating sleep stages using a wrist-worn device that measures movement using a 3D accelerometer and an optical pulse photoplethysmograph (PPG). Approach: Overnight recordings were obtained from 60 adult participants wearing these devices on their left and right wrist, simultaneously with a Type III home sleep testing device (Embletta MPR) which included EEG channels for sleep staging. The 60 participants were self-reported normal sleepers (36 M: 24 F, age = 34 ± 10, BMI = 28 ± 6). The Embletta recordings were scored for sleep stages using AASM guidelines and were used to develop and validate an automated sleep stage estimation algorithm, which labeled sleep stages as one of Wake, Light (N1 or N2), Deep (N3) and REM (REM). Features were extracted from the accelerometer and PPG sensors, which reflected movement, breathing and heart rate variability. Main results: Based on leave-one-out validation, the overall per-epoch accuracy of the automated algorithm was 69%, with a Cohen’s kappa of 0.52 ± 0.14. There was no observable bias to under- or over-estimate wake, light, or deep sleep durations. REM sleep duration was slightly over-estimated by the system. The most common misclassifications were light/REM and light/wake mislabeling. Significance: The results indicate that a reasonable degree of sleep staging accuracy can be achieved using a wrist-worn device, which may be of utility in longitudinal studies of sleep habits.","container-title":"Physiological Measurement","DOI":"10.1088/1361-6579/aa9047","ISSN":"0967-3334","issue":"11","journalAbbreviation":"Physiol. Meas.","language":"en","page":"1968–1979","source":"Institute of Physics","title":"Estimation of sleep stages in a healthy adult population from optical plethysmography and accelerometer signals","volume":"38","author":[{"family":"Beattie","given":"Z."},{"family":"Oyang","given":"Y."},{"family":"Statan","given":"A."},{"family":"Ghoreyshi","given":"A."},{"family":"Pantelopoulos","given":"A."},{"family":"Russell","given":"A."},{"family":"Heneghan","given":"C."}],"issued":{"date-parts":[["2017",10]]}}}],"schema":""} 1160 adults (self-reported normal sleepers) Age: 34 ± 10 yearsRange 19 to 60 yearsFour classκ = 0.52Accuracy = 69%Raw PPG dataTransmissive PPG data from a finger pulse oximeter2020Korkalainen et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"Ceg1YWVN","properties":{"formattedCitation":"\\super 12\\nosupersub{}","plainCitation":"12","noteIndex":0},"citationItems":[{"id":191,"uris":[""],"uri":[""],"itemData":{"id":191,"type":"article-journal","abstract":"AbstractStudy Objectives. Accurate identification of sleep stages is essential in the diagnosis of sleep disorders (e.g. obstructive sleep apnea [OSA]) but rel","container-title":"Sleep","DOI":"10.1093/sleep/zsaa098","journalAbbreviation":"Sleep","language":"en","source":"academic.","title":"Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea","URL":"","author":[{"family":"Korkalainen","given":"Henri"},{"family":"Aakko","given":"Juhani"},{"family":"Duce","given":"Brett"},{"family":"Kainulainen","given":"Samu"},{"family":"Leino","given":"Akseli"},{"family":"Nikkonen","given":"Sami"},{"family":"Afara","given":"Isaac O."},{"family":"Myllymaa","given":"Sami"},{"family":"T?yr?s","given":"Juha"},{"family":"Lepp?nen","given":"Timo"}],"accessed":{"date-parts":[["2020",10,22]]}}}],"schema":""} 12894 patients suspected of OSAAge: 56.1 (45.3 – 63.3) (median (IQR)Five classκ = 0.51Accuracy = 64.1%Four classκ = 0.54Accuracy = 68.5%Three classκ = 0.65Accuracy = 80.1%PPG data from optical wrist heart rate monitor (PulseOn Ltd)2019Molkkari et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"24ANh7Y1","properties":{"formattedCitation":"\\super 13\\nosupersub{}","plainCitation":"13","noteIndex":0},"citationItems":[{"id":167,"uris":[""],"uri":[""],"itemData":{"id":167,"type":"paper-conference","abstract":"We assess the feasibility of heart rate variability (HRV) estimated from interbeat interval (IBI) data measured with wrist-worn photoplethysmography device for sleep stage classi?cation. In particular, we examine fractal correlations in the IBIs as the function of both time and scale.","DOI":"10.22489/CinC.2019.287","event":"2019 Computing in Cardiology Conference","language":"en","source":" (Crossref)","title":"Non-Linear Heart Rate Variability Measures in Sleep Stage Analysis with Photoplethysmography","URL":"","author":[{"family":"Molkkari","given":"Matti"},{"family":"Tenhunen","given":"Mirja"},{"family":"Tarniceriu","given":"Adrian"},{"family":"Vehkaoja","given":"Antti"},{"family":"Himanen","given":"Sari-Leena"},{"family":"R?s?nen","given":"Esa"}],"accessed":{"date-parts":[["2020",10,19]]},"issued":{"date-parts":[["2019",12,30]]}}}],"schema":""} 13 18 healthy adults Age: 28 years (average) Range 21 to 42 yearsFive classκ = 0.35Accuracy = 54.2%Four classκ = 0.38Accuracy = 60.1%Three classκ = 0.43Accuracy = 72.5%ActigraphyaActiwatch Spectrum Pro2020Kahawage et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"9kZ9I3nj","properties":{"formattedCitation":"\\super 7\\nosupersub{}","plainCitation":"7","noteIndex":0},"citationItems":[{"id":181,"uris":[""],"uri":[""],"itemData":{"id":181,"type":"article-journal","abstract":"Consumer activity trackers claiming to measure sleep/wake patterns are ubiquitous within clinical and consumer settings. However, validation of these devices in sleep disorder populations are lacking. We examined 1?night of sleep in 42 individuals with insomnia (mean = 49.14 ± 17.54 years) using polysomnography, a wrist actigraph (Actiwatch Spectrum Pro: AWS) and a consumer activity tracker (Fitbit Alta HR: FBA). Epoch‐by‐epoch analysis and Bland?Altman methods evaluated each device against polysomnography for sleep/wake detection, total sleep time, sleep efficiency, wake after sleep onset and sleep latency. FBA sleep stage classification of light sleep (N1 + N2), deep sleep (N3) and rapid eye movement was also compared with polysomnography. Compared with polysomnography, both activity trackers displayed high accuracy (81.12% versus 82.80%, AWS and FBA respectively; ns) and sensitivity (sleep detection; 96.66% versus 96.04%, respectively; ns) but low specificity (wake detection; 39.09% versus 44.76%, respectively; p = .037). Both trackers overestimated total sleep time and sleep efficiency, and underestimated sleep latency and wake after sleep onset. FBA demonstrated sleep stage sensitivity and specificity, respectively, of 79.39% and 58.77% (light), 49.04% and 95.54% (deep), 65.97% and 91.53% (rapid eye movement). Both devices were more accurate in detecting sleep than wake, with equivalent sensitivity, but statistically different specificity. FBA provided equivalent estimates as AWS for all traditional actigraphy sleep parameters. FBA also showed high specificity when identifying N3, and rapid eye movement, though sensitivity was modest. Thus, it underestimates these sleep stages and overestimates light sleep, demonstrating more shallow sleep than actually obtained. Whether FBA could serve as a low‐cost substitute for actigraphy in insomnia requires further investigation.","container-title":"Journal of Sleep Research","DOI":"10.1111/jsr.12931","ISSN":"0962-1105, 1365-2869","issue":"1","journalAbbreviation":"J Sleep Res","language":"en","source":" (Crossref)","title":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I: In‐lab validation against polysomnography","title-short":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I","URL":"","volume":"29","author":[{"family":"Kahawage","given":"Piyumi"},{"family":"Jumabhoy","given":"Ria"},{"family":"Hamill","given":"Kellie"},{"family":"Zambotti","given":"Massimiliano"},{"family":"Drummond","given":"Sean P. A."}],"accessed":{"date-parts":[["2020",10,20]]},"issued":{"date-parts":[["2020",2]]}}}],"schema":""} 7 42 adults with insomnia Age: 49.14 ± 17.54Range 19 to 82Two classκ** = 0.66Accuracy = 81.1%Specificity* = 39.09%Actiwatch 22018Pesonen et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"o8kDt5pb","properties":{"formattedCitation":"\\super 14\\nosupersub{}","plainCitation":"14","noteIndex":0},"citationItems":[{"id":168,"uris":[""],"uri":[""],"itemData":{"id":168,"type":"article-journal","abstract":"STUDY OBJECTIVES: The validity of consumer-targeted wrist-worn sleep measurement systems has been little studied in children and adolescents. We examined the validity of a new fitness tracker (PFT) manufactured by Polar Electro Oy and the previously validated Actiwatch 2 (AW2) from Philips Respironics against polysomnography (PSG) in children and adolescents.\nMETHODS: Seventeen children (age 11.0 ± 0.8 years) and 17 adolescents (age 17.8 ± 1.8 years) wore the PFT and AW2 concurrently with an ambulatory PSG in their own home for 1 night. We compared sleep onset, offset, sleep interval (time from sleep on to offset), actual sleep time (time scored as sleep between sleep on and offset), and wake after sleep onset (WASO) between accelerometers and PSG. Sensitivity, specificity, and accuracy were calculated from the epoch-by-epoch data.\nRESULTS: Both devices performed adequately against PSG, with excellent sensitivity for both age groups (> 0.91). In terms of specificity, the PFT was adequate in both groups (> 0.77), and AW2 adequate in children (0.68) and poor in adolescents (0.58). In the younger group, the PFT underestimated actual sleep time by 29.9 minutes and AW2 underestimated actual sleep time by 43.6 minutes. Both overestimated WASO, PFT by 24.4 minutes and AW2 by 20.9 minutes. In the older group, both devices underestimated actual sleep time (PFT by 20.6 minutes and AW2 by 26.8 minutes) and overestimated WASO (PFT by 12.5 minutes and AW2 by 14.3 minutes). Both devices were accurate in defining sleep onset.\nCONCLUSIONS: This study suggests that this consumer-targeted wrist-worn device performs as well as, or even better than, the previously validated AW2 against PSG in children and adolescents. Both devices underestimated sleep but to a lesser extent than seen in many previous validation studies on research-targeted accelerometers.","container-title":"Journal of clinical sleep medicine: JCSM: official publication of the American Academy of Sleep Medicine","DOI":"10.5664/jcsm.7050","ISSN":"1550-9397","issue":"4","journalAbbreviation":"J Clin Sleep Med","language":"eng","note":"PMID: 29609722\nPMCID: PMC5886436","page":"585-591","source":"PubMed","title":"The Validity of a New Consumer-Targeted Wrist Device in Sleep Measurement: An Overnight Comparison Against Polysomnography in Children and Adolescents","title-short":"The Validity of a New Consumer-Targeted Wrist Device in Sleep Measurement","volume":"14","author":[{"family":"Pesonen","given":"Anu-Katriina"},{"family":"Kuula","given":"Liisa"}],"issued":{"date-parts":[["2018"]],"season":"15"}}}],"schema":""} 14 17 healthy children Age: 11.0 ± 0.8 yearsRange: 9.9 to 12.6 years17 healthy adolescents Age: 17.8 ± 1.8 years Range 14.4 to 19.8 yearsTwo classIn children Accuracy = 90%Specificity* = 68%In adolescents Accuracy = 89%Specificity* = 58%AW-64 2013Marino et al. ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"AMK7gv2z","properties":{"formattedCitation":"\\super 8\\nosupersub{}","plainCitation":"8","noteIndex":0},"citationItems":[{"id":247,"uris":[""],"uri":[""],"itemData":{"id":247,"type":"article-journal","abstract":"We validated actigraphy for detecting sleep and wakefulness versus polysomnography (PSG).Actigraphy and polysomnography were simultaneously collected during sleep laboratory admissions. All studies involved 8.5 h time in bed, except for sleep restriction studies. Epochs (30-sec; n = 232,849) were characterized for sensitivity (actigraphy = sleep when PSG = sleep), specificity (actigraphy = wake when PSG = wake), and accuracy (total proportion correct); the amount of wakefulness after sleep onset (WASO) was also assessed. A generalized estimating equation (GEE) model included age, gender, insomnia diagnosis, and daytime/nighttime sleep timing factors.Controlled sleep laboratory conditions.Young and older adults, healthy or chronic primary insomniac (PI) patients, and daytime sleep of 23 night-workers (n = 77, age 35.0 ± 12.5, 30F, mean nights = 3.2).N/A.Overall, sensitivity (0.965) and accuracy (0.863) were high, whereas specificity (0.329) was low; each was only slightly modified by gender, insomnia, day/night sleep timing (magnitude of change &lt; 0.04). Increasing age slightly reduced specificity. Mean WASO/night was 49.1 min by PSG compared to 36.8 min/night by actigraphy (β = 0.81; CI = 0.42, 1.21), unbiased when WASO &lt; 30 min/night, and overestimated when WASO &gt; 30 min/night.This validation quantifies strengths and weaknesses of actigraphy as a tool measuring sleep in clinical and population studies. Overall, the participant-specific accuracy is relatively high, and for most participants, above 80%. We validate this finding across multiple nights and a variety of adults across much of the young to midlife years, in both men and women, in those with and without insomnia, and in 77 participants. We conclude that actigraphy is overall a useful and valid means for estimating total sleep time and wakefulness after sleep onset in field and workplace studies, with some limitations in specificity.","container-title":"Sleep","DOI":"10.5665/sleep.3142","ISSN":"0161-8105","issue":"11","journalAbbreviation":"Sleep","page":"1747-1755","source":"Silverchair","title":"Measuring Sleep: Accuracy, Sensitivity, and Specificity of Wrist Actigraphy Compared to Polysomnography","title-short":"Measuring Sleep","volume":"36","author":[{"family":"Marino","given":"Miguel"},{"family":"Li","given":"Yi"},{"family":"Rueschman","given":"Michael N."},{"family":"Winkelman","given":"J. W."},{"family":"Ellenbogen","given":"J. M."},{"family":"Solet","given":"J. M."},{"family":"Dulin","given":"Hilary"},{"family":"Berkman","given":"Lisa F."},{"family":"Buxton","given":"Orfeu M."}],"issued":{"date-parts":[["2013",11,1]]}}}],"schema":""} 8Total of 77 participants including: people with insomnia, healthy participants, older adults, healthy sleep-restricted participants and night workers Age: 45.0 ± 12.5 years17 patients with insomniaAge: 40.5 ± 8.2 yearsTwo classIn participants without insomniaAccuracy = 86.9%Specificity* = 33.1%In insomnia patientsAccuracy = 83.3%Specificity* = 34.7%Actiwatch SpectrumConsumer wearablesaFatigue Science Readiband2021Chinoy et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"dHQlTpCZ","properties":{"formattedCitation":"\\super 15\\nosupersub{}","plainCitation":"15","noteIndex":0},"citationItems":[{"id":243,"uris":[""],"uri":[""],"itemData":{"id":243,"type":"article-journal","abstract":"Consumer sleep-tracking devices are widely used and becoming more technologically advanced, creating strong interest from researchers and clinicians for their possible use as alternatives to standard actigraphy. We therefore tested the performance of many of the latest consumer sleep-tracking devices, alongside actigraphy, versus the gold-standard sleep assessment technique, polysomnography (PSG).In total, 34 healthy young adults (22 women; 28.1 ± 3.9 years, mean ± SD) were tested on three consecutive nights (including a disrupted sleep condition) in a sleep laboratory with PSG, along with actigraphy (Philips Respironics Actiwatch 2) and a subset of consumer sleep-tracking devices. Altogether, four wearable (Fatigue Science Readiband, Fitbit Alta HR, Garmin Fenix 5S, Garmin Vivosmart 3) and three non-wearable (EarlySense Live, ResMed S+, SleepScore Max) devices were tested. Sleep/wake summary and epoch-by-epoch agreement measures were compared with PSG.Most devices (Fatigue Science Readiband, Fitbit Alta HR, EarlySense Live, ResMed S+, SleepScore Max) performed as well as or better than actigraphy on sleep/wake performance measures, while the Garmin devices performed worse. Overall, epoch-by-epoch sensitivity was high (all ≥0.93), specificity was low-to-medium (0.18-0.54), sleep stage comparisons were mixed, and devices tended to perform worse on nights with poorer/disrupted sleep.Consumer sleep-tracking devices exhibited high performance in detecting sleep, and most performed equivalent to (or better than) actigraphy in detecting wake. Device sleep stage assessments were inconsistent. Findings indicate that many newer sleep-tracking devices demonstrate promising performance for tracking sleep and wake. Devices should be tested in different populations and settings to further examine their wider validity and utility.","container-title":"Sleep","DOI":"10.1093/sleep/zsaa291","ISSN":"0161-8105","journalAbbreviation":"Sleep","source":"Silverchair","title":"Performance of Seven Consumer Sleep-Tracking Devices Compared with Polysomnography","URL":"","author":[{"family":"Chinoy","given":"Evan D"},{"family":"Cuellar","given":"Joseph A"},{"family":"Huwa","given":"Kirbie E"},{"family":"Jameson","given":"Jason T"},{"family":"Watson","given":"Catherine H"},{"family":"Bessman","given":"Sara C"},{"family":"Hirsch","given":"Dale A"},{"family":"Cooper","given":"Adam D"},{"family":"Drummond","given":"Sean P A"},{"family":"Markwald","given":"Rachel R"}],"accessed":{"date-parts":[["2021",1,14]]},"issued":{"date-parts":[["2021"]]}}}],"schema":""} 15 34 healthy adultsAge: 28.1 ± 3.9Two classAccuracy = 88%Specificity = 45%Fitbit Alta HRAccuracy = 90%Specificity = 54%Garmin Fenix 5SAccuracy = 88%Specificity = 18%Garmin Vivosmart3Accuracy = 88%Specificity = 19%Fitbit Alta HR2020Kahawage et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"UEy0hR42","properties":{"formattedCitation":"\\super 7\\nosupersub{}","plainCitation":"7","noteIndex":0},"citationItems":[{"id":181,"uris":[""],"uri":[""],"itemData":{"id":181,"type":"article-journal","abstract":"Consumer activity trackers claiming to measure sleep/wake patterns are ubiquitous within clinical and consumer settings. However, validation of these devices in sleep disorder populations are lacking. We examined 1?night of sleep in 42 individuals with insomnia (mean = 49.14 ± 17.54 years) using polysomnography, a wrist actigraph (Actiwatch Spectrum Pro: AWS) and a consumer activity tracker (Fitbit Alta HR: FBA). Epoch‐by‐epoch analysis and Bland?Altman methods evaluated each device against polysomnography for sleep/wake detection, total sleep time, sleep efficiency, wake after sleep onset and sleep latency. FBA sleep stage classification of light sleep (N1 + N2), deep sleep (N3) and rapid eye movement was also compared with polysomnography. Compared with polysomnography, both activity trackers displayed high accuracy (81.12% versus 82.80%, AWS and FBA respectively; ns) and sensitivity (sleep detection; 96.66% versus 96.04%, respectively; ns) but low specificity (wake detection; 39.09% versus 44.76%, respectively; p = .037). Both trackers overestimated total sleep time and sleep efficiency, and underestimated sleep latency and wake after sleep onset. FBA demonstrated sleep stage sensitivity and specificity, respectively, of 79.39% and 58.77% (light), 49.04% and 95.54% (deep), 65.97% and 91.53% (rapid eye movement). Both devices were more accurate in detecting sleep than wake, with equivalent sensitivity, but statistically different specificity. FBA provided equivalent estimates as AWS for all traditional actigraphy sleep parameters. FBA also showed high specificity when identifying N3, and rapid eye movement, though sensitivity was modest. Thus, it underestimates these sleep stages and overestimates light sleep, demonstrating more shallow sleep than actually obtained. Whether FBA could serve as a low‐cost substitute for actigraphy in insomnia requires further investigation.","container-title":"Journal of Sleep Research","DOI":"10.1111/jsr.12931","ISSN":"0962-1105, 1365-2869","issue":"1","journalAbbreviation":"J Sleep Res","language":"en","source":" (Crossref)","title":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I: In‐lab validation against polysomnography","title-short":"Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I","URL":"","volume":"29","author":[{"family":"Kahawage","given":"Piyumi"},{"family":"Jumabhoy","given":"Ria"},{"family":"Hamill","given":"Kellie"},{"family":"Zambotti","given":"Massimiliano"},{"family":"Drummond","given":"Sean P. A."}],"accessed":{"date-parts":[["2020",10,20]]},"issued":{"date-parts":[["2020",2]]}}}],"schema":""} 7 42 adults with insomnia Age: 49.14 ± 17.54 yearsRange 19 to 82 yearsTwo classκ** = 0.58Accuracy = 82.8%Specificity* = 44.76%Fitbit Charge HR2020Godino et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"fqHAZgMt","properties":{"formattedCitation":"\\super 16\\nosupersub{}","plainCitation":"16","noteIndex":0},"citationItems":[{"id":171,"uris":[""],"uri":[""],"itemData":{"id":171,"type":"article-journal","abstract":"Purpose This study sought to assess the performance of the Fitbit Charge HR, a consumer-level multi-sensor activity tracker, to measure physical activity and sleep in children. Methods 59 healthy boys and girls aged 9–11 years old wore a Fitbit Charge HR, and accuracy of physical activity measures were evaluated relative to research-grade measures taken during a combination of 14 standardized laboratory- and field-based assessments of sitting, stationary cycling, treadmill walking or jogging, stair walking, outdoor walking, and agility drills. Accuracy of sleep measures were evaluated relative to polysomnography (PSG) in 26 boys and girls during an at-home unattended PSG overnight recording. The primary analyses included assessment of the agreement (biases) between measures using the Bland-Altman method, and epoch-by-epoch (EBE) analyses on a minute-by-minute basis. Results Fitbit Charge HR underestimated steps (~11.8 steps per minute), heart rate (~3.58 bpm), and metabolic equivalents (~0.55 METs per minute) and overestimated energy expenditure (~0.34 kcal per minute) relative to research-grade measures (p< 0.05). The device showed an overall accuracy of 84.8% for classifying moderate and vigorous physical activity (MVPA) and sedentary and light physical activity (SLPA) (sensitivity MVPA: 85.4%; specificity SLPA: 83.1%). Mean estimates of bias for measuring total sleep time, wake after sleep onset, and heart rate during sleep were 14 min, 9 min, and 1.06 bpm, respectively, with 95.8% sensitivity in classifying sleep and 56.3% specificity in classifying wake epochs. Conclusions Fitbit Charge HR had adequate sensitivity in classifying moderate and vigorous intensity physical activity and sleep, but had limitations in detecting wake, and was more accurate in detecting heart rate during sleep than during exercise, in healthy children. Further research is needed to understand potential challenges and limitations of these consumer devices.","container-title":"PLOS ONE","DOI":"10.1371/journal.pone.0237719","ISSN":"1932-6203","issue":"9","journalAbbreviation":"PLOS ONE","language":"en","note":"publisher: Public Library of Science","page":"e0237719","source":"PLoS Journals","title":"Performance of a commercial multi-sensor wearable (Fitbit Charge HR) in measuring physical activity and sleep in healthy children","volume":"15","author":[{"family":"Godino","given":"Job G."},{"family":"Wing","given":"David"},{"family":"Zambotti","given":"Massimiliano","dropping-particle":"de"},{"family":"Baker","given":"Fiona C."},{"family":"Bagot","given":"Kara"},{"family":"Inkelis","given":"Sarah"},{"family":"Pautz","given":"Carina"},{"family":"Higgins","given":"Michael"},{"family":"Nichols","given":"Jeanne"},{"family":"Brumback","given":"Ty"},{"family":"Chevance","given":"Guillaume"},{"family":"Colrain","given":"Ian M."},{"family":"Patrick","given":"Kevin"},{"family":"Tapert","given":"Susan F."}],"issued":{"date-parts":[["2020",9,4]]}}}],"schema":""} 16 19 healthy children Age: 9.9 ± 0.7 yearsRange 9 to 11 yearsTwo classκ = 0.53Accuracy = 92.1%Specificity* = 56.9%Fitbit Charge 2 and Fitbit Charge HR2019Moreno-Pino et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"5Dl6YwrF","properties":{"formattedCitation":"\\super 17\\nosupersub{}","plainCitation":"17","noteIndex":0},"citationItems":[{"id":239,"uris":[""],"uri":[""],"itemData":{"id":239,"type":"article-journal","abstract":"Study Objectives:\nConsumer wearable devices may be a helpful method of assessing sleep, but validation is required for their use in clinical practice. Our aim was to validate two models of Fitbit sleep trackers that rely on both accelerometer and heart rate sensors against polysomnography in participants with obstructive sleep apnea (OSA).\n\nMethods:\nParticipants were adults presenting with symptoms of OSA and attending our outpatient sleep clinic. A polysomnography (PSG) was applied to all participants at the same time they were wearing a Fitbit sleep tracker. Using paired t tests and Bland-Altman plots, we compared the sleep measures provided by the wearable devices with those obtained by PSG. Since Fitbit devices’ automatic detection of sleep start time can cause bias, we performed a correction using Huber loss function-based linear regression and a leave-one-out strategy.\n\nResults:\nOur sample consisted of 65 patients. Diagnosis of OSA was confirmed on 55 (84.6%). There were statistically significant differences between PSG and Fitbit measures for all sleep outcomes but rapid eye movement sleep. Fitbit devices overestimated total sleep time, and underestimated wake after sleep onset and sleep onset latency. After correction of bias, Fitbit-delivered measures of sleep onset latency did not significantly differ of those provided by PSG.\n\nConclusions:\nFitbit wearable devices showed an acceptable sensitivity but poor specificity. Consumer sleep trackers still have insufficient accuracy for clinical settings, especially in clinical populations. Solving technical issues and optimizing clinically-oriented features could make them apt for their use in clinical practice in a nondistant future.\n\nCitation:\nMoreno-Pino F, Porras-Segovia A, López-Esteban P, Artés A, Baca-García E. Validation of Fitbit Charge 2 and Fitbit Alta HR against polysomnography for assessing sleep in adults with obstructive sleep apnea. J Clin Sleep Med. 2019;15(11):1645–1653.","container-title":"Journal of Clinical Sleep Medicine : JCSM : Official Publication of the American Academy of Sleep Medicine","DOI":"10.5664/jcsm.8032","ISSN":"1550-9389","issue":"11","journalAbbreviation":"J Clin Sleep Med","note":"PMID: 31739855\nPMCID: PMC6853383","page":"1645-1653","source":"PubMed Central","title":"Validation of Fitbit Charge 2 and Fitbit Alta HR Against Polysomnography for Assessing Sleep in Adults With Obstructive Sleep Apnea","volume":"15","author":[{"family":"Moreno-Pino","given":"Fernando"},{"family":"Porras-Segovia","given":"Alejandro"},{"family":"López-Esteban","given":"Pilar"},{"family":"Artés","given":"Antonio"},{"family":"Baca-García","given":"Enrique"}],"issued":{"date-parts":[["2019",11,15]]}}}],"schema":""} 1765 adults from which 55 with OSAAge: 58.84 ± 13.84 yearsRange 22 to 85 yearsTwo classSpecificity* = 43.85%Polar Fitness Tracker2018Pesonen et al ADDIN ZOTERO_ITEM CSL_CITATION {"citationID":"PLX3xFfy","properties":{"formattedCitation":"\\super 14\\nosupersub{}","plainCitation":"14","noteIndex":0},"citationItems":[{"id":168,"uris":[""],"uri":[""],"itemData":{"id":168,"type":"article-journal","abstract":"STUDY OBJECTIVES: The validity of consumer-targeted wrist-worn sleep measurement systems has been little studied in children and adolescents. We examined the validity of a new fitness tracker (PFT) manufactured by Polar Electro Oy and the previously validated Actiwatch 2 (AW2) from Philips Respironics against polysomnography (PSG) in children and adolescents.\nMETHODS: Seventeen children (age 11.0 ± 0.8 years) and 17 adolescents (age 17.8 ± 1.8 years) wore the PFT and AW2 concurrently with an ambulatory PSG in their own home for 1 night. We compared sleep onset, offset, sleep interval (time from sleep on to offset), actual sleep time (time scored as sleep between sleep on and offset), and wake after sleep onset (WASO) between accelerometers and PSG. Sensitivity, specificity, and accuracy were calculated from the epoch-by-epoch data.\nRESULTS: Both devices performed adequately against PSG, with excellent sensitivity for both age groups (> 0.91). In terms of specificity, the PFT was adequate in both groups (> 0.77), and AW2 adequate in children (0.68) and poor in adolescents (0.58). In the younger group, the PFT underestimated actual sleep time by 29.9 minutes and AW2 underestimated actual sleep time by 43.6 minutes. Both overestimated WASO, PFT by 24.4 minutes and AW2 by 20.9 minutes. In the older group, both devices underestimated actual sleep time (PFT by 20.6 minutes and AW2 by 26.8 minutes) and overestimated WASO (PFT by 12.5 minutes and AW2 by 14.3 minutes). Both devices were accurate in defining sleep onset.\nCONCLUSIONS: This study suggests that this consumer-targeted wrist-worn device performs as well as, or even better than, the previously validated AW2 against PSG in children and adolescents. Both devices underestimated sleep but to a lesser extent than seen in many previous validation studies on research-targeted accelerometers.","container-title":"Journal of clinical sleep medicine: JCSM: official publication of the American Academy of Sleep Medicine","DOI":"10.5664/jcsm.7050","ISSN":"1550-9397","issue":"4","journalAbbreviation":"J Clin Sleep Med","language":"eng","note":"PMID: 29609722\nPMCID: PMC5886436","page":"585-591","source":"PubMed","title":"The Validity of a New Consumer-Targeted Wrist Device in Sleep Measurement: An Overnight Comparison Against Polysomnography in Children and Adolescents","title-short":"The Validity of a New Consumer-Targeted Wrist Device in Sleep Measurement","volume":"14","author":[{"family":"Pesonen","given":"Anu-Katriina"},{"family":"Kuula","given":"Liisa"}],"issued":{"date-parts":[["2018"]],"season":"15"}}}],"schema":""} 14 17 healthy children Age: 11.0 ± 0.8 yearsRange: 9.9 to 12.6 years17 healthy adolescents Age: 17.8 ± 1.8 years Range 14.4 to 19.8 yearsTwo classIn children Accuracy = 91%Specificity* = 77%In adolescentsAccuracy = 90%Specificity* = 83%*Please note that the referenced papers used the term ‘specificity’ with wake defined as the negative class. In our results we use term ‘sensitivity’ with wake as the positive class, so the terms are referring to the same outcome.** Weighted kappa scoreaThis includes only a limited selection of recent published literatureBland-Altman plots for the sleep statistics for both adults and children/adolescentsFigure S1 Bland-Altman plots for the sleep statistics for both adults and children/adolescents.SOL: sleep onset latency; WASO: wake after sleep onset; TST: total sleep time; SE: sleep efficiencyReferences supplementary data ADDIN ZOTERO_BIBL {"uncited":[],"omitted":[],"custom":[]} CSL_BIBLIOGRAPHY 1. Somers VK, Dyken ME, Clary MP, Abboud FM. Sympathetic neural mechanisms in obstructive sleep apnea. J Clin Invest. 1995;96(4):1897-1904. doi:10.1172/JCI1182352. Fonseca P, Gilst MM van, Radha M, et al. Automatic sleep staging using heart rate variability, body movements and recurrent neural networks in a sleep disordered population. Sleep. Published online 2020. doi:10.1093/sleep/zsaa0483. Bliwise DL, Williams ML, Irbe D, Ansari FP, Rye DB. Inter-rater Reliability for Identification of REM Sleep in Parkinson’s Disease. Sleep. 2000;23(5):1-6. doi:10.1093/sleep/23.5.1j4. Hauri PJ, Wisbey J. Wrist actigraphy in insomnia. Sleep. 1992;15(4):293-301. doi:10.1093/sleep/15.4.2935. Sadeh A, Hauri PJ, Kripke DF, Lavie P. The Role of Actigraphy in the Evaluation of Sleep Disorders. Sleep. 1995;18(4):288-302. doi:10.1093/sleep/18.4.2886. Long X, Fonseca P, Haakma R, Aarts RM. Actigraphy-based sleep/wake detection for insomniacs. In: 2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks (BSN). ; 2017:1-4. doi:10.1109/BSN.2017.79357117. Kahawage P, Jumabhoy R, Hamill K, Zambotti M, Drummond SPA. Validity, potential clinical utility, and comparison of consumer and research‐grade activity trackers in Insomnia Disorder I: In‐lab validation against polysomnography. J Sleep Res. 2020;29(1). doi:10.1111/jsr.129318. Marino M, Li Y, Rueschman MN, et al. Measuring Sleep: Accuracy, Sensitivity, and Specificity of Wrist Actigraphy Compared to Polysomnography. Sleep. 2013;36(11):1747-1755. doi:10.5665/sleep.31429. Walch O, Huang Y, Forger D, Goldstein C. Sleep stage prediction with raw acceleration and photoplethysmography heart rate data derived from a consumer wearable device. Sleep. 2019;42(12). doi:10.1093/sleep/zsz18010. Fonseca P, Weysen T, Goelema MS, et al. Validation of Photoplethysmography-Based Sleep Staging Compared With Polysomnography in Healthy Middle-Aged Adults. Sleep. 2017;40(7). doi:10.1093/sleep/zsx09711. Beattie Z, Oyang Y, Statan A, et al. Estimation of sleep stages in a healthy adult population from optical plethysmography and accelerometer signals. Physiol Meas. 2017;38(11):1968-1979. doi:10.1088/1361-6579/aa904712. Korkalainen H, Aakko J, Duce B, et al. Deep learning enables sleep staging from photoplethysmogram for patients with suspected sleep apnea. Sleep. doi:10.1093/sleep/zsaa09813. Molkkari M, Tenhunen M, Tarniceriu A, Vehkaoja A, Himanen S-L, R?s?nen E. Non-Linear Heart Rate Variability Measures in Sleep Stage Analysis with Photoplethysmography. In: ; 2019. doi:10.22489/CinC.2019.28714. Pesonen A-K, Kuula L. The Validity of a New Consumer-Targeted Wrist Device in Sleep Measurement: An Overnight Comparison Against Polysomnography in Children and Adolescents. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2018;14(4):585-591. doi:10.5664/jcsm.705015. Chinoy ED, Cuellar JA, Huwa KE, et al. Performance of Seven Consumer Sleep-Tracking Devices Compared with Polysomnography. Sleep. Published online 2021. doi:10.1093/sleep/zsaa29116. Godino JG, Wing D, Zambotti M de, et al. Performance of a commercial multi-sensor wearable (Fitbit Charge HR) in measuring physical activity and sleep in healthy children. PLOS ONE. 2020;15(9):e0237719. doi:10.1371/journal.pone.023771917. Moreno-Pino F, Porras-Segovia A, López-Esteban P, Artés A, Baca-García E. Validation of Fitbit Charge 2 and Fitbit Alta HR Against Polysomnography for Assessing Sleep in Adults With Obstructive Sleep Apnea. J Clin Sleep Med JCSM Off Publ Am Acad Sleep Med. 2019;15(11):1645-1653. doi:10.5664/jcsm.8032 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download