Abstract:Rationale: More than a million polysomnograms (PSGs) are performed annually in the United States to diagnose obstructive sleep apnea (OSA). Third-party payers now advocate a home sleep test (HST), rather than an in-laboratory PSG, as the diagnostic study for OSA regardless of clinical probability, but the economic benefit of this approach is not known.
Objectives:We determined the diagnostic performance of OSA prediction tools including the newly developed OSUNet, based on an artificial neural network, and per… Show more
“…As a preprocessing step first, all cases with missing data were excluded. Second, BMI was categorized as normal (20.0 ≤ BMI < 25), overweight (25.0 ≤ BMI < 30.0), and obese (BMI ≥ 30) according to the World (14) 127 42154 5113 410 313 (4) 87 (29) 153 (50) 51 (17) 259 (85) 45 (15) 51 (17) 111 (36) 96 (32) 46 (15) 103 (34) Health Organization guidelines [23]. Age was categorized as young adult (18-35 years), middle-aged adult (36-55 years) and older adult (56 years and older) [24].…”
Section: Pre-processingmentioning
confidence: 99%
“…Since there is no widely approved predefined classification of neck circumference, it was categorized in four equally sized groupings using the quintile values of their distribution. A cut-off score of 10 was used for the Epworth sleepiness scale (ESS), with categorization as normal (0-9) and abnormal (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24) [25].…”
Section: Pre-processingmentioning
confidence: 99%
“…Several strategies have been proposed and investigated for the detection of obstructive sleep apnea diagnoses from nonpolysomnographic features to decrease the number of PSG studies [4][5][6][7][8][9][10][11][12][13][14]. In those studies, the clinical features that are associated with the OSA have been combined using statistical modeling or machine learning techniques into a clinical prediction model.…”
Study results showed that machine learning methods can be used to estimate the probabilities of no, mild, moderate, and severe obstructive sleep apnea and such approaches may improve accurate initial OSA screening and help referring only the suspected moderate or severe OSA patients to sleep laboratories for the expensive tests.
“…As a preprocessing step first, all cases with missing data were excluded. Second, BMI was categorized as normal (20.0 ≤ BMI < 25), overweight (25.0 ≤ BMI < 30.0), and obese (BMI ≥ 30) according to the World (14) 127 42154 5113 410 313 (4) 87 (29) 153 (50) 51 (17) 259 (85) 45 (15) 51 (17) 111 (36) 96 (32) 46 (15) 103 (34) Health Organization guidelines [23]. Age was categorized as young adult (18-35 years), middle-aged adult (36-55 years) and older adult (56 years and older) [24].…”
Section: Pre-processingmentioning
confidence: 99%
“…Since there is no widely approved predefined classification of neck circumference, it was categorized in four equally sized groupings using the quintile values of their distribution. A cut-off score of 10 was used for the Epworth sleepiness scale (ESS), with categorization as normal (0-9) and abnormal (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24) [25].…”
Section: Pre-processingmentioning
confidence: 99%
“…Several strategies have been proposed and investigated for the detection of obstructive sleep apnea diagnoses from nonpolysomnographic features to decrease the number of PSG studies [4][5][6][7][8][9][10][11][12][13][14]. In those studies, the clinical features that are associated with the OSA have been combined using statistical modeling or machine learning techniques into a clinical prediction model.…”
Study results showed that machine learning methods can be used to estimate the probabilities of no, mild, moderate, and severe obstructive sleep apnea and such approaches may improve accurate initial OSA screening and help referring only the suspected moderate or severe OSA patients to sleep laboratories for the expensive tests.
“…There are several studies that addressed this problem, since additional tests and procedures do not always result in definitive diagnosis or termination of illness. There are several studies that suggest the use of ANN models in reducing overall costs for proper treatment of clinical conditions (Walczak, 2000;Liew, 2007, Abbod, 2011Teferra, 2014).…”
Digital Agenda in Serbia involves the introduction of an electronic system for monitoring of the main characteristics of patients, disease progression and treatment outcomes through EHR (Electronic Health Record). Internationally standardized data set contains more than 150 variables, with a tendency to introduce new frequently. In addition to the increased demand for treatment, there are also demands for optimizing the health care system. In order to predict the likelihood of diagnosis, course and outcome of treatment, classically multivariate regression linear logistic model is being used. In recent years, studies indicate that the use of Artificial Neural Networks (ANN) may provide improved results in terms of likelihood of final diagnosis and outcomes that include input variables which, by their nature, have a non-linear interdependence. We reviewed current ANN models, their advantages and disadvantages compared to common regression models and their applicability in clinical practice. Also, we analyzed and suggested models that could possibly optimize the process of diagnosis, predict the cost and duration of treatment and rationalize medical and other resources by reducing the cost/ benefit coefficient per patient.
“…Other screening tools, such as the multivariable apnea prediction 15 and OSUNet, 16 treat BMI and age as continuous variables and may offer greater utility. 3 Therefore, a binary, highthreshold value for BMI limits the discriminatory power of this key risk factor.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.