Diagnosis of obstructive sleep apnea in children based on the XGBoost algorithm using nocturnal heart rate and blood oxygen feature

Ye, Pengfei; Qin, Han; Zhan, Xiaojun; Wang, Zhan; Liu, Chang; Song, Biao; Kong, Yaru; Jia, Xinbei; Qi, Yuwei; Ji, Jie; Li, Chang; Ni, Xin; Tai, Jun

doi:10.1016/j.amjoto.2022.103714

Cited by 5 publications

(3 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While it ranked as the less accurate and sensitive among the five models in this study, its higher AUC and specificity make it a promising candidate for future investigations. The study by Ye et al ( 2023 ) included involving 3,139 children with suspected OSA used age, sex, BMI, hypoxia index, mean nocturnal heart rate, and fastest heart rate as predictive features for diagnosing mild, moderate, and severe OSA. XGBoost demonstrated AUCs of 0.95, 0.88, and 0.88, with classification accuracies of 90.45%, 85.67%, and 89.81%, respectively.…”

Section: Discussionmentioning

confidence: 99%

Development and application of a machine learning-based predictive model for obstructive sleep apnea screening

Liu,

Geng,

Shen

et al. 2024

Front. Big Data

View full text Add to dashboard Cite

ObjectiveTo develop a robust machine learning prediction model for the automatic screening and diagnosis of obstructive sleep apnea (OSA) using five advanced algorithms, namely Extreme Gradient Boosting (XGBoost), Logistic Regression (LR), Support Vector Machine (SVM), Light Gradient Boosting Machine (LightGBM), and Random Forest (RF) to provide substantial support for early clinical diagnosis and intervention.MethodsWe conducted a retrospective analysis of clinical data from 439 patients who underwent polysomnography at the Affiliated Hospital of Xuzhou Medical University between October 2019 and October 2022. Predictor variables such as demographic information [age, sex, height, weight, body mass index (BMI)], medical history, and Epworth Sleepiness Scale (ESS) were used. Univariate analysis was used to identify variables with significant differences, and the dataset was then divided into training and validation sets in a 4:1 ratio. The training set was established to predict OSA severity grading. The validation set was used to assess model performance using the area under the curve (AUC). Additionally, a separate analysis was conducted, categorizing the normal population as one group and patients with moderate-to-severe OSA as another. The same univariate analysis was applied, and the dataset was divided into training and validation sets in a 4:1 ratio. The training set was used to build a prediction model for screening moderate-to-severe OSA, while the validation set was used to verify the model's performance.ResultsAmong the four groups, the LightGBM model outperformed others, with the top five feature importance rankings of ESS total score, BMI, sex, hypertension, and gastroesophageal reflux (GERD), where Age, ESS total score and BMI played the most significant roles. In the dichotomous model, RF is the best performer of the five models respectively. The top five ranked feature importance of the best-performing RF models were ESS total score, BMI, GERD, age and Dry mouth, with ESS total score and BMI being particularly pivotal.ConclusionMachine learning-based prediction models for OSA disease grading and screening prove instrumental in the early identification of patients with moderate-to-severe OSA, revealing pertinent risk factors and facilitating timely interventions to counter pathological changes induced by OSA. Notably, ESS total score and BMI emerge as the most critical features for predicting OSA, emphasizing their significance in clinical assessments. The dataset will be publicly available on my Github.

show abstract

Section: Discussionmentioning

confidence: 99%

Development and application of a machine learning-based predictive model for obstructive sleep apnea screening

Liu,

Geng,

Shen

et al. 2024

Front. Big Data

View full text Add to dashboard Cite

show abstract

“…Ye et al used a machine learning diagnostic model based on the XGBoost algorithm for accurate prediction of children with different OSA severities, using heart rate and blood oxygen data as the major features. When compared to PSG, this diagnostic modality reduces the number of signals and uses a simpler diagnostic process which may be helpful for those with suspected OSA but does not have the opportunity of obtaining a diagnostic PSG [ 62 ].…”

Section: Discussionmentioning

confidence: 99%

Comorbidity Profile and Predictors of Obstructive Sleep Apnea Severity and Mortality in Non-Obese Obstructive Sleep Apnea Patients

2023

View full text Add to dashboard Cite

Backgrounds and Objectives: Obstructive sleep apnea (OSA) is associated with increased morbidity and mortality. OSA is an independent risk factor for many different conditions, especially cardiovascular diseases. The purpose of this study was to ascertain the comorbidity profile of non-obese patients with newly diagnosed OSA and evaluate the risk for cardiovascular disease and mortality. The present study also aimed to establish predictors for OSA severity. Materials and Methods: This study included 138 newly diagnosed patients who underwent polysomnographic analysis. The 10-year risk for cardiovascular disease was assessed using a newly validated prediction model: Systematic Coronary Risk Evaluation (SCORE-2). In addition, the Charlson Comorbidity Index (CCI) was assessed as a widely-used example of a mortality comorbidity index. Results: The study population included 138 patients: 86 males and 52 females. Patients were stratified, according to AHI (apnea/hypopnea index), into four groups: 33 patients had mild OSA (5 ≤ AHI < 15), 33 patients had moderate OSA (15 ≤ AHI < 30), 31 patients had severe OSA (AHI ≥ 30), and 41 individuals had AHI < 5, which were a part of the control group. SCORE-2 increased in line with OSA severity and was higher in OSA groups compared to the control group (H = 29.913; DF = 3; p < 0.001). Charlson Index was significantly higher in OSA patients compared to controls (p = 0.001), with a higher prevalence of total comorbidities in the OSA group of patients. Furthermore, CCI 10-year survival score was significantly lower in the OSA group, suggesting a shorter survival of those patients with a more severe form of OSA. We also examined the prediction model for OSA severity. Conclusions: Determining the comorbidity profile and estimation of the 10-year risk score of OSA patients could be used to classify these patients into various mortality risk categories and, according to that, provide them with adequate treatment.

show abstract

“… Biddle & Fallah (2021) used SVM to detect and identify faults in the sensors of autonomous vehicle control systems. Ye et al (2022) established a diagnostic model of OSA in children based on the XGBoost algorithm. Using heart rate and blood oxygen data as the main features, a machine learning diagnostic model based on the XGBoost algorithm can accurately identify children with OSA at different severities.…”

Section: Literary Reviewmentioning

confidence: 99%

A new method to identifying optimal adjustment strategy when the car cockpit is uncomfortable: optimal state distance method

Chen

Shi

Yang

et al. 2023

PeerJ Computer Science

View full text Add to dashboard Cite

With the rapid development of the automobile industry, the comfort of the cockpit has become the standard for judging the quality of the car. People have also put forward higher requirements for cockpit comfort. In the process of driving, the cockpit environment will constantly change, and the comfort will also change. When the comprehensive comfort level of the cockpit decreases and the occupants feel uncomfortable, the cockpit comfort should be adjusted. In this article, a cockpit comfort evaluation model is established to realize the evaluation of cockpit comfort. In addition, we elaborate the theory of optimal state distance, where the numerical magnitude of the optimal state distance is used to reflect the extent to which an indicator deviates from its optimal state. Also, a cockpit optimal adjustment strategy identification model is established based on the theory, which can obtain the optimal adjustment strategy in a certain cockpit operating environment, facilitate the timely adjustment of the corresponding actuator, and realize the dynamic monitoring and adjustment of cockpit comfort. This project provides a reference direction for cockpit comfort adjustment, which is of great significance for future research and development of automotive cockpit comfort.

show abstract

Diagnosis of obstructive sleep apnea in children based on the XGBoost algorithm using nocturnal heart rate and blood oxygen feature

Cited by 5 publications

References 27 publications

Development and application of a machine learning-based predictive model for obstructive sleep apnea screening

Development and application of a machine learning-based predictive model for obstructive sleep apnea screening

Comorbidity Profile and Predictors of Obstructive Sleep Apnea Severity and Mortality in Non-Obese Obstructive Sleep Apnea Patients

A new method to identifying optimal adjustment strategy when the car cockpit is uncomfortable: optimal state distance method

Contact Info

Product

Resources

About