Prospective Testing of Two Models Based on Clinical and Oximetric Variables for Prediction of Obstructive Sleep Apnea

Roche, N.; Herer, Bertrand; Roig, Catherine; Huchon, G.

doi:10.1378/chest.121.3.747

Cited by 25 publications

(22 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in the present study, there was no significant difference between group 1 and group 2 in the prevalence of OSA and characteristics of the populations. In addition, the overall accuracy of our prediction model in the development cohort (group 1; 79.2%) was much higher than that of the previous study (66%) [27]. Variation in symptom perception and reporting over time, problems with internal validity, and reproducibility of prediction models are plausible explanations for our finding.…”

Section: Discussioncontrasting

confidence: 64%

“…The small number of clinical variables and the absence of morphometric and oximetric measurements in our prediction models may have affected its diagnostic accuracy and reproducibility. However, our prediction model achieved relatively good overall diagnostic accuracy in the development cohort (group 1) compared with other studies using small numbers of variables [27]. Furthermore, our primary aim was to investigate the diagnostic characteristics of clinical models that do not involve complex morphometric or oximetric measurements that are suitable for routine clinical use.…”

Section: Discussionmentioning

confidence: 87%

“…Roche et al found that even in a stable setting with unchanged mode of recruitment, the diagnostic characteristics of mathematical models based on clinical and oximetric variables varied markedly over time [27]. The authors postulated that variation in the prevalence of OSA due to changes in referral pattern and low diagnostic performance of the mathematical model caused by the limited numbers of variables used were the possible reasons.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Diagnostic characteristics of clinical prediction models for obstructive sleep apnea in different clinic populations

et al. 2010

View full text Add to dashboard Cite

show abstract

Section: Discussioncontrasting

confidence: 64%

Section: Discussionmentioning

confidence: 87%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Diagnostic characteristics of clinical prediction models for obstructive sleep apnea in different clinic populations

et al. 2010

View full text Add to dashboard Cite

show abstract

“…These results were improved by using the Δ index together with the other conventional indices, which yielded a sensitivity of 90% and a specificity of 70%. Finally, Roche et al (2002) suggested to combine information from oximetry data with clinical features using logistic regression. This model achieved an accuracy of 62.1%.…”

Section: Discussionmentioning

confidence: 99%

“…They include the oxygen desaturation index over 2% (ODI2), 3% (ODI3) and 4% (ODI4), and the cumulative time spent below a given level of saturation. Typically, a saturation level of 90% (CT90) is applied (Lévy et al 1996, Roche et al 2002, Vázquez et al 2000, Netzer et al 2001. Recently, signal processing techniques have been also used for automated analysis of oximetry recordings through spectral and nonlinear methods (Zamarrón et al 2003.…”

Section: Introductionmentioning

confidence: 99%

The classification of oximetry signals using Bayesian neural networks to assist in the detection of obstructive sleep apnoea syndrome

et al. 2010

View full text Add to dashboard Cite

Nowadays, polysomnography (PSG) is the gold-standard to diagnose obstructive sleep apnoea syndrome (OSAS). However, it is complex, time-consuming and expensive. Nocturnal pulse oximetry, which provides oxygen saturation (SaO 2 ) recordings, allows us to overcome these difficulties and could be an alternative to PSG. In the present study, multilayer perceptron (MLP) neural networks were applied to help in OSAS diagnosis using information from SaO 2 signals. We performed time and spectral analysis of these recordings to extract 14 features related to OSAS. According to the principle used for network optimisation, wWe compared the performance of two different MLP classifiers: maximum likelihood (ML) and Bayesian (BY) MLP networks. A total of 187 subjects suspected of suffering from OSAS took part in the study. Their SaO 2 signals were divided into a training set with 74 recordings and a test set with 113 recordings to develop and validate the classifiers. BY-MLP networks achieved the best performance on the test set with 85.58% accuracy (87.76% sensitivity and 82.39% specificity). These results were substantially better than those provided by ML-MLP networks, which were affected by overfitting and achieved an accuracy of 76.81% (86.42% sensitivity and 62.83% specificity). Our results suggest that the Bayesian framework is preferred to implement our MLP classifiers. The proposed BY-MLP networks could be used for early OSAS detection, contributing to and thus reduce the number of required PSGs.

show abstract

Radial basis function classifiers to help in the diagnosis of the obstructive sleep apnoea syndrome from nocturnal oximetry

Marcos

Hornero

Álvarez

et al. 2007

Med Biol Eng Comput

View full text Add to dashboard Cite

The aim of this study is to assess the ability of radial basis function (RBF) classifiers as an assistant tool for the diagnosis of the obstructive sleep apnoea syndrome (OSAS). A total of 187 subjects suspected of suffering from OSAS were available for our research. The initial population was divided into training, validation and test sets for deriving and testing our neural classifiers. We used nonlinear features from nocturnal oxygen saturation (SaO(2)) to perform patients' classification. We evaluated three different RBF construction techniques based on the following algorithms: k-means (KM), fuzzy c-means (FCM) and orthogonal least squares (OLS). A diagnostic accuracy of 86.1, 84.7 and 85.5% was provided by the networks developed with KM, FCM and OLS, respectively. The three proposed networks achieved an area under the receiver operating characteristic (ROC) curve over 0.90. Our results showed that a useful non-invasive method could be applied to diagnose OSAS from nonlinear features of SaO(2) with RBF classifiers.

show abstract

Prospective Testing of Two Models Based on Clinical and Oximetric Variables for Prediction of Obstructive Sleep Apnea

Cited by 25 publications

References 18 publications

Diagnostic characteristics of clinical prediction models for obstructive sleep apnea in different clinic populations

Diagnostic characteristics of clinical prediction models for obstructive sleep apnea in different clinic populations

The classification of oximetry signals using Bayesian neural networks to assist in the detection of obstructive sleep apnoea syndrome

Radial basis function classifiers to help in the diagnosis of the obstructive sleep apnoea syndrome from nocturnal oximetry

Contact Info

Product

Resources

About