Journal of clinical sleep medicine : JCSM : official publication of the American Academy of Sleep Medicine · 2025
Predictive modeling of obstructive sleep apnea using pharyngeal magnetic resonance imaging radiomics and clinical data.
СтатьяПоддержка исследования: вне правительства СШАНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Chen Yibin, Xiao Heng, Huang Min, Zheng Yingying, Dong Xiaoyu, Chen Guohao
Аннотация
STUDY OBJECTIVES: This study aims to assess the predictive performance of models combining pharyngeal magnetic resonance imaging radiomics and clinical data for distinguishing severe and nonsevere obstructive sleep apnea.
METHODS: A total of 106 patients were included in the study, with 48 patients having an apnea-hypopnea index < 30 events/h and 58 patients having an apnea-hypopnea index ≥ 30 events/h. Radiomics features were extracted from magnetic resonance imaging images. After applying minimum redundancy and maximum relevance and least absolute shrinkage and selection operator with cross-validation for dimensionality reduction, radiomics models were developed using logistic regression, support vector machine, random forest, and gradient boosting machine. Age and body mass index were used as clinical features to construct a combined model with radiomics features. The performance of the models was evaluated using F1 scores and the area under the receiver operating characteristic curve (AUC).
RESULTS: A total of 129 radiomics features were extracted from magnetic resonance imaging images. Following dimensionality reduction and feature selection, 2 radiomics features with significant predictive value were identified. The combined model, incorporating support vector machine (AUC = 0.78, F1 = 0.75), random forest (AUC = 0.78, F1 = 0.74), gradient boosting machine (AUC = 0.79, F1 = 0.75), and logistic regression (AUC = 0.82, F1 = 0.80), demonstrated superior performance compared to models based solely on radiomics features. The radiomics-only models included support vector machine (AUC = 0.76, F1 = 0.72), random forest (AUC = 0.73, F1 = 0.67), gradient boosting machine (AUC = 0.76, F1 = 0.73), and logistic regression (AUC = 0.78, F1 = 0.76). Among the combined models, logistic regression achieved the highest predictive accuracy and classification performance.
CONCLUSIONS: The combined model, integrating radiomics features with clinical characteristics, demonstrates a superior ability to distinguish between severe and nonsevere obstructive sleep apnea. This approach offers a noninvasive and effective new perspective for clinical decision-making.
CITATION: Chen Y, Xiao H, Huang M, Zheng Y, Dong X, Chen G. Predictive modeling of obstructive sleep apnea using pharyngeal magnetic resonance imaging radiomics and clinical data. J Clin Sleep Med. 2025;21(8):1363-1369.
METHODS: A total of 106 patients were included in the study, with 48 patients having an apnea-hypopnea index < 30 events/h and 58 patients having an apnea-hypopnea index ≥ 30 events/h. Radiomics features were extracted from magnetic resonance imaging images. After applying minimum redundancy and maximum relevance and least absolute shrinkage and selection operator with cross-validation for dimensionality reduction, radiomics models were developed using logistic regression, support vector machine, random forest, and gradient boosting machine. Age and body mass index were used as clinical features to construct a combined model with radiomics features. The performance of the models was evaluated using F1 scores and the area under the receiver operating characteristic curve (AUC).
RESULTS: A total of 129 radiomics features were extracted from magnetic resonance imaging images. Following dimensionality reduction and feature selection, 2 radiomics features with significant predictive value were identified. The combined model, incorporating support vector machine (AUC = 0.78, F1 = 0.75), random forest (AUC = 0.78, F1 = 0.74), gradient boosting machine (AUC = 0.79, F1 = 0.75), and logistic regression (AUC = 0.82, F1 = 0.80), demonstrated superior performance compared to models based solely on radiomics features. The radiomics-only models included support vector machine (AUC = 0.76, F1 = 0.72), random forest (AUC = 0.73, F1 = 0.67), gradient boosting machine (AUC = 0.76, F1 = 0.73), and logistic regression (AUC = 0.78, F1 = 0.76). Among the combined models, logistic regression achieved the highest predictive accuracy and classification performance.
CONCLUSIONS: The combined model, integrating radiomics features with clinical characteristics, demonstrates a superior ability to distinguish between severe and nonsevere obstructive sleep apnea. This approach offers a noninvasive and effective new perspective for clinical decision-making.
CITATION: Chen Y, Xiao H, Huang M, Zheng Y, Dong X, Chen G. Predictive modeling of obstructive sleep apnea using pharyngeal magnetic resonance imaging radiomics and clinical data. J Clin Sleep Med. 2025;21(8):1363-1369.