Machine learning for image-based detection of patients with obstructive sleep apnea: an exploratory study.
Ontology highlight
ABSTRACT: Purpose
In 2-dimensional lateral cephalometric radiographs, patients with severe obstructive sleep apnea (OSA) exhibit a more crowded oropharynx in comparison with non-OSA. We tested the hypothesis that machine learning, an application of artificial intelligence (AI), could be used to detect patients with severe OSA based on 2-dimensional images.Methods
A deep convolutional neural network was developed (n = 1258; 90%) and tested (n = 131; 10%) using data from 1389 (100%) lateral cephalometric radiographs obtained from individuals diagnosed with severe OSA (n = 867; apnea hypopnea index > 30 events/h sleep) or non-OSA (n = 522; apnea hypopnea index < 5 events/h sleep) at a single center for sleep disorders. Three kinds of data sets were prepared by changing the area of interest using a single image: the original image without any modification (full image), an image containing a facial profile, upper airway, and craniofacial soft/hard tissues (main region), and an image containing part of the occipital region (head only). A radiologist also performed a conventional manual cephalometric analysis of the full image for comparison.Results
The sensitivity/specificity was 0.87/0.82 for full image, 0.88/0.75 for main region, 0.71/0.63 for head only, and 0.54/0.80 for the manual analysis. The area under the receiver-operating characteristic curve was the highest for main region 0.92, for full image 0.89, for head only 0.70, and for manual cephalometric analysis 0.75.Conclusions
A deep convolutional neural network identified individuals with severe OSA with high accuracy. Future research on this concept using AI and images can be further encouraged when discussing triage of OSA.
SUBMITTER: Tsuiki S
PROVIDER: S-EPMC8590647 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundObstructive sleep apnea (OSA) is a highly prevalent sleep disorder associated with daytime sleepiness, fatigue, and increased all-cause mortality risk in patients with cancer. Existing screening tools for OSA do not account for the interaction of cancer-related features that may increase OSA risk.Study design and methodsThis is a retrospective study of patients with cancer at a single tertiary cancer institution who underwent a home sleep apnea test (HSAT) to evaluate for OSA. Unsupervised machine learning (ML) was used to reduce the dimensions and extract significant features associated with OSA. ML classifiers were applied to principal components and model hyperparameters were optimized using k-fold cross-validation. Training models for OSA were subsequently tested and compared with the STOP-Bang questionnaire on a prospective unseen test set of patients who underwent an HSAT.ResultsFrom a training dataset of 249 patients, kernel principal component analysis (PCA) extracted eight components through dimension reduction to explain the maximum variance with OSA at 98%. Predictors of OSA were smoking, asthma, chronic kidney disease, STOP-Bang score, race, diabetes, radiation to head/neck/thorax (RT-HNT), type of cancer, and cancer metastases. Of the ML models, PCA + RF had the highest sensitivity (96.8%), specificity (92.3%), negative predictive value (92%), F1 score (0.93), and ROC-AUC score (0.88). The PCA + RF screening algorithm also performed better than the STOP-Bang questionnaire alone when tested on a prospective unseen test set.ConclusionsThe PCA + RF ML model had the highest accuracy in screening for OSA in patients with cancer. History of RT-HNT, cancer metastases, and type of cancer were identified as cancer-related risk factors for OSA.
| S-EPMC10735319 | biostudies-literature
Project description:BackgroundObstructive sleep apnea (OSA) is a globally prevalent disease closely associated with hypertension. To date, no predictive model for OSA-related hypertension has been established. We aimed to use machine learning (ML) to construct a model to analyze risk factors and predict OSA-related hypertension.Materials and methodsWe retrospectively collected the clinical data of OSA patients diagnosed by polysomnography from October 2019 to December 2021 and randomly divided them into training and validation sets. A total of 1,493 OSA patients with 27 variables were included. Independent risk factors for the risk of OSA-related hypertension were screened by the multifactorial logistic regression models. Six ML algorithms, including the logistic regression (LR), the gradient boosting machine (GBM), the extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), bootstrapped aggregating (Bagging), and the multilayer perceptron (MLP), were used to develop the model on the training set. The validation set was used to tune the model hyperparameters to determine the final prediction model. We compared the accuracy and discrimination of the models to identify the best machine learning algorithm for predicting OSA-related hypertension. In addition, a web-based tool was developed to promote its clinical application. We used permutation importance and Shapley additive explanations (SHAP) to determine the importance of the selected features and interpret the ML models.ResultsA total of 18 variables were selected for the models. The GBM model achieved the most extraordinary discriminatory ability (area under the receiver operating characteristic curve = 0.873, accuracy = 0.885, sensitivity = 0.713), and on the basis of this model, an online tool was built to help clinicians optimize OSA-related hypertension patient diagnosis. Finally, age, family history of hypertension, minimum arterial oxygen saturation, body mass index, and percentage of time of SaO2 < 90% were revealed by the SHAP method as the top five critical variables contributing to the diagnosis of OSA-related hypertension.ConclusionWe established a risk prediction model for OSA-related hypertension patients using the ML method and demonstrated that among the six ML models, the gradient boosting machine model performs best. This prediction model could help to identify high-risk OSA-related hypertension patients, provide early and individualized diagnoses and treatment plans, protect patients from the serious consequences of OSA-related hypertension, and minimize the burden on society.
| S-EPMC9760810 | biostudies-literature
Project description:Rationale: Obstructive sleep apnea (OSA) has been associated with metabolic dysregulation and systemic inflammation. This may be due to pathophysiologic effects of OSA on visceral adipose tissue. We sought to assess the transcriptional consequences of OSA on adipocytes by utilizing pathway-focused analyses. Methods: Patients scheduled to undergo ventral hernia repair surgery were recruited to wear a portable home sleep monitor for two nights prior to surgery. Visceral fat biopsies were obtained intra-operatively. RNA was extracted and whole-genome expression profiling was performed. Gene Set Enrichment Analysis (GSEA) was used to identify curated gene sets that were differentially enriched in OSA subjects. Network analysis was applied to a select set of highly enriched pathways. Results: 10 patients with OSA and 8 control subjects were recruited. There were no differences in age, gender, body mass index between the two groups, but the OSA subjects had a significantly higher respiratory disturbance index (19.2 vs. 0.6, P-value 0.05) and worse hypoxemia (minimum oxygen saturation 79.7% vs. 87.8%, P-value < 0.001). GSEA identified a number of gene sets up-regulated in adipose tissue of OSA patients including the pro-inflammatory NF-M-NM-:B pathway and the proteolytic ubiquitin/proteasome module. A critical metabolic pathway, the peroxisome proliferator-activated receptor (PPAR), was down-regulated in subjects with OSA. Network analysis linked members of these modules together and identified regulatory hubs. Conclusions: OSA is associated with alterations in visceral fat gene expression. Pathway-based network analysis highlighted perturbations in several key pathways whose coordinated interactions may contribute to the metabolic dysregulation observed in this complex disorder. Total RNA from visceral fat of 18 subjects (10 OSA, 8 Control) was hybridized to 18 Affymetrix Genechip Human Gene 1.0 ST microarrays.
2012-09-01 | E-GEOD-38792 | biostudies-arrayexpress
Project description:Rationale: Obstructive sleep apnea (OSA) has been associated with metabolic dysregulation and systemic inflammation. This may be due to pathophysiologic effects of OSA on visceral adipose tissue. We sought to assess the transcriptional consequences of OSA on adipocytes by utilizing pathway-focused analyses. Methods: Patients scheduled to undergo ventral hernia repair surgery were recruited to wear a portable home sleep monitor for two nights prior to surgery. Visceral fat biopsies were obtained intra-operatively. RNA was extracted and whole-genome expression profiling was performed. Gene Set Enrichment Analysis (GSEA) was used to identify curated gene sets that were differentially enriched in OSA subjects. Network analysis was applied to a select set of highly enriched pathways. Results: 10 patients with OSA and 8 control subjects were recruited. There were no differences in age, gender, body mass index between the two groups, but the OSA subjects had a significantly higher respiratory disturbance index (19.2 vs. 0.6, P-value 0.05) and worse hypoxemia (minimum oxygen saturation 79.7% vs. 87.8%, P-value < 0.001). GSEA identified a number of gene sets up-regulated in adipose tissue of OSA patients including the pro-inflammatory NF-κB pathway and the proteolytic ubiquitin/proteasome module. A critical metabolic pathway, the peroxisome proliferator-activated receptor (PPAR), was down-regulated in subjects with OSA. Network analysis linked members of these modules together and identified regulatory hubs. Conclusions: OSA is associated with alterations in visceral fat gene expression. Pathway-based network analysis highlighted perturbations in several key pathways whose coordinated interactions may contribute to the metabolic dysregulation observed in this complex disorder.
2012-09-01 | GSE38792 | GEO
Project description:The ability of computers to learn from and annotate large databases of crystallization-trial images provides not only the ability to reduce the workload of crystallization studies, but also an opportunity to annotate crystallization trials as part of a framework for improving screening methods. Here, a system is presented that scores sets of images based on the likelihood of containing crystalline material as perceived by a machine-learning algorithm. The system can be incorporated into existing crystallization-analysis pipelines, whereby specialists examine images as they normally would with the exception that the images appear in rank order according to a simple real-valued score. Promising results are shown for 319 112 images associated with 150 structures solved by the Joint Center for Structural Genomics pipeline during the 2006-2007 year. Overall, the algorithm achieves a mean receiver operating characteristic score of 0.919 and a 78% reduction in human effort per set when considering an absolute score cutoff for screening images, while incurring a loss of five out of 150 structures.
| S-EPMC2585161 | biostudies-literature
Project description:Obstructive sleep apnea (OSA) can aggravate blood pressure and increase the risk of cardiovascular diseases in hypertensive individuals, yet the underlying pathophysiological process is still incompletely understood. More importantly, OSA remains a significantly undiagnosed condition. In this study, a total of 559 hypertensive patients with and without OSA were included. Metabolome and lipidome-wide analyses were performed to explore the pathophysiological processes of hypertension comorbid OSA and derive potential biomarkers for diagnosing OSA in hypertensive subjects. Compared to non-OSA hypertensive patients (discovery set = 120; validation set = 116), patients with OSA (discovery set = 165; validation set = 158) demonstrated a unique sera metabolic phenotype dominated by abnormalities in biological processes of oxidative stress and inflammation. By integrating three machine learning algorithms, six discriminatory metabolites (including 5-hydroxyeicosatetraenoic acid, taurine, histidine, lysophosphatidic acid 16:0, lysophosphatidylcholine 18:0, and dihydrosphingosine) were selected for constructing diagnostic and classified model. Notably, the established multivariate-model could accurately identify OSA subjects. The corresponding area under the curve values and the correct classification rates were 0.995 and 96.8% for discovery sets, 0.997 and 99.1% for validation sets. This work updates the molecular insights of hypertension comorbid OSA and paves the way for the use of metabolomics for the diagnosis of OSA in hypertensive individuals.
| S-EPMC9598902 | biostudies-literature
Project description:Importance:Given the high prevalence of obstructive sleep apnea (OSA), there is a need for simpler and automated diagnostic approaches. Objective:To evaluate whether mandibular movement (MM) monitoring during sleep coupled with an automated analysis by machine learning is appropriate for OSA diagnosis. Design, Setting, and Participants:Diagnostic study of adults undergoing overnight in-laboratory polysomnography (PSG) as the reference method compared with simultaneous MM monitoring at a sleep clinic in an academic institution (Sleep Laboratory, Centre Hospitalier Universitaire Université Catholique de Louvain Namur Site Sainte-Elisabeth, Namur, Belgium). Patients with suspected OSA were enrolled from July 5, 2017, to October 31, 2018. Main Outcomes and Measures:Obstructive sleep apnea diagnosis required either evoking signs or symptoms or related medical or psychiatric comorbidities coupled with a PSG-derived respiratory disturbance index (PSG-RDI) of at least 5 events/h. A PSG-RDI of at least 15 events/h satisfied the diagnosis criteria even in the absence of associated symptoms or comorbidities. Patients who did not meet these criteria were classified as not having OSA. Agreement analysis and diagnostic performance were assessed by Bland-Altman plot comparing PSG-RDI and the Sunrise system RDI (Sr-RDI) with diagnosis threshold optimization via receiver operating characteristic curves, allowing for evaluation of the device sensitivity and specificity in detecting OSA at 5 events/h and 15 events/h. Results:Among 376 consecutive adults with suspected OSA, the mean (SD) age was 49.7 (13.2) years, the mean (SD) body mass index was 31.0 (7.1), and 207 (55.1%) were men. Reliable agreement was found between PSG-RDI and Sr-RDI in patients without OSA (n = 46; mean difference, 1.31; 95% CI, -1.05 to 3.66 events/h) and in patients with OSA with a PSG-RDI of at least 5 events/h with symptoms (n = 107; mean difference, -0.69; 95% CI, -3.77 to 2.38 events/h). An Sr-RDI underestimation of -11.74 (95% CI, -20.83 to -2.67) events/h in patients with OSA with a PSG-RDI of at least 15 events/h was detected and corrected by optimization of the Sunrise system diagnostic threshold. The Sr-RDI showed diagnostic capability, with areas under the receiver operating characteristic curve of 0.95 (95% CI, 0.92-0.96) and 0.93 (95% CI, 0.90-0.93) for corresponding PSG-RDIs of 5 events/h and 15 events/h, respectively. At the 2 optimal cutoffs of 7.63 events/h and 12.65 events/h, Sr-RDI had accuracy of 0.92 (95% CI, 0.90-0.94) and 0.88 (95% CI, 0.86-0.90) as well as posttest probabilities of 0.99 (95% CI, 0.99-0.99) and 0.89 (95% CI, 0.88-0.91) at PSG-RDIs of at least 5 events/h and at least 15 events/h, respectively, corresponding to positive likelihood ratios of 14.86 (95% CI, 9.86-30.12) and 5.63 (95% CI, 4.92-7.27), respectively. Conclusions and Relevance:Automatic analysis of MM patterns provided reliable performance in RDI calculation. The use of this index in OSA diagnosis appears to be promising.
| S-EPMC6991283 | biostudies-literature
Project description:Unsupervised clustering models have been widely used for multimetric phenotyping of complex and heterogeneous diseases such as diabetes and obstructive sleep apnea (OSA) to more precisely characterize the disease beyond simplistic conventional diagnosis standards. However, the number of clusters and key phenotypic features have been subjectively selected, reducing the reliability of the phenotyping results. Here, to minimize such subjective decisions for highly confident phenotyping, we develop a multimetric phenotyping framework by combining supervised and unsupervised machine learning. This clusters 2277 OSA patients to six phenotypes based on their multidimensional polysomnography (PSG) data. Importantly, these new phenotypes show statistically different comorbidity development for OSA-related cardio-neuro-metabolic diseases, unlike the conventional single-metric apnea-hypopnea index-based phenotypes. Furthermore, the key features of highly comorbid phenotypes were identified through supervised learning rather than subjective choice. These results can also be used to automatically phenotype new patients and predict their comorbidity risks solely based on their PSG data. The phenotyping framework based on the combination of unsupervised and supervised machine learning methods can also be applied to other complex, heterogeneous diseases for phenotyping patients and identifying important features for high-risk phenotypes.
| S-EPMC7904925 | biostudies-literature
Project description:ObjectivesTo evaluate the role of clinical scores assessing the risk of disease severity in patients with clinical suspicion of obstructive sleep apnea syndrome (OSA). The hypothesis was tested by applying artificial intelligence (AI) to demonstrate its effectiveness in distinguishing between mild-moderate OSA and severe OSA risk.MethodsA support vector machine model (SVM) was developed from the samples included in the analysis (N = 498), and they were split into 75% for training (N = 373) with the remaining for testing (N = 125). Two diagnostic thresholds were selected for OSA severity: mild to moderate (apnea-hypopnea index (AHI) ≥ 5 events/h and AHI < 30 events/h) and severe (AHI ≥ 30 events/h). The algorithms were trained and tested to predict OSA patient severity.ResultsThe sensitivity and specificity for the SVM model were 0.93 and 0.80 with an accuracy of 0.86; instead, the logistic regression full mode reported a value of 0.74 and 0.63, respectively, with an accuracy of 0.68. After backward stepwise elimination for features selection, the reduced logistic regression model demonstrated a sensitivity and specificity of 0.79 and 0.56, respectively, and an accuracy of 0.67.ConclusionArtificial intelligence could be applied to patients with symptoms related to OSA to identify individuals with a severe OSA risk with clinical-based algorithms in the OSA framework.
| S-EPMC10056063 | biostudies-literature
Project description:Carpal tunnel syndrome (CTS) is the most common entrapment neuropathy of the upper extremity. It is usually associated with the compression of the median nerve in the median groove. Because the main symptoms of CTS pain and numbness worsen at night, sleep disorders in CTS patients and the impact of preferred sleeping position on CTS development have been formerly studied. However, to the best of our knowledge, this is the first study assessing the frequency of CTS in obstructive sleep apnea (OSA) patients. This study aimed to determine the frequency of CTS in OSA patients and evaluate the causative relation between the two diseases.Records of individuals who were admitted to our sleep laboratory were retrospectively scanned. Eighty patients who were diagnosed with OSA and did not have comorbidities that might cause OSA (e.g., diabetes mellitus, hypothyroiditis, rheumatic diseases, and cervical radiculopathy) were included in the study along with 80 healthy controls who matched for age, sex, and BMI of OSA patients. To maintain observer blindness, patients were not questioned regarding their symptoms or the clinical data that would be used in the study. All participants underwent nerve conduction studies. Those who were diagnosed with CTS were questioned regarding CTS symptoms and the preferred sleeping position. Subsequently, patients were given the Boston CTS questionnaire.CTS frequency in OSA patients was found to be 27.5%. There was no significant relation between preferred sleeping position or being a manual worker and having CTS.CTS frequency in OSA patients is significantly higher than that in healthy individuals. In contrast to previous studies that have been performed in the absence of polysomnographic and electrophysiological data, in our study biomechanical factors were not associated with CTS presence. Therefore, we conclude that intermittent hypoxemia is the main etiological factor for CTS in OSA patients. Inflammation may be a common factor for etiopathogenesis for both diseases, but this hypothesis needs further investigation.
| S-EPMC5758072 | biostudies-other