Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features
Obstructive sleep apnea (OSA) is a global health concern and is typically diagnosed using in-laboratory polysomnography (PSG). However, PSG is highly time-consuming and labor-intensive. We, therefore, developed machine learning models based on easily accessed anthropometric features to screen for th...
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/22/8630 |
| _version_ | 1827643755990089728 |
|---|---|
| author | Cheng-Yu Tsai Huei-Tyng Huang Hsueh-Chien Cheng Jieni Wang Ping-Jung Duh Wen-Hua Hsu Marc Stettler Yi-Chun Kuan Yin-Tzu Lin Chia-Rung Hsu Kang-Yun Lee Jiunn-Horng Kang Dean Wu Hsin-Chien Lee Cheng-Jung Wu Arnab Majumdar Wen-Te Liu |
| author_facet | Cheng-Yu Tsai Huei-Tyng Huang Hsueh-Chien Cheng Jieni Wang Ping-Jung Duh Wen-Hua Hsu Marc Stettler Yi-Chun Kuan Yin-Tzu Lin Chia-Rung Hsu Kang-Yun Lee Jiunn-Horng Kang Dean Wu Hsin-Chien Lee Cheng-Jung Wu Arnab Majumdar Wen-Te Liu |
| author_sort | Cheng-Yu Tsai |
| collection | DOAJ |
| description | Obstructive sleep apnea (OSA) is a global health concern and is typically diagnosed using in-laboratory polysomnography (PSG). However, PSG is highly time-consuming and labor-intensive. We, therefore, developed machine learning models based on easily accessed anthropometric features to screen for the risk of moderate to severe and severe OSA. We enrolled 3503 patients from Taiwan and determined their PSG parameters and anthropometric features. Subsequently, we compared the mean values among patients with different OSA severity and considered correlations among all participants. We developed models based on the following machine learning approaches: logistic regression, k-nearest neighbors, naïve Bayes, random forest (RF), support vector machine, and XGBoost. Collected data were first independently split into two data sets (training and validation: 80%; testing: 20%). Thereafter, we adopted the model with the highest accuracy in the training and validation stage to predict the testing set. We explored the importance of each feature in the OSA risk screening by calculating the Shapley values of each input variable. The RF model achieved the highest accuracy for moderate to severe (84.74%) and severe (72.61%) OSA. The level of visceral fat was found to be a predominant feature in the risk screening models of OSA with the aforementioned levels of severity. Our machine learning models can be employed to screen for OSA risk in the populations in Taiwan and in those with similar craniofacial structures. |
| first_indexed | 2024-03-09T18:01:32Z |
| format | Article |
| id | doaj.art-e86bd102a07141a3bf73b0352cb36050 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:01:32Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-e86bd102a07141a3bf73b0352cb360502023-11-24T09:53:16ZengMDPI AGSensors1424-82202022-11-012222863010.3390/s22228630Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric FeaturesCheng-Yu Tsai0Huei-Tyng Huang1Hsueh-Chien Cheng2Jieni Wang3Ping-Jung Duh4Wen-Hua Hsu5Marc Stettler6Yi-Chun Kuan7Yin-Tzu Lin8Chia-Rung Hsu9Kang-Yun Lee10Jiunn-Horng Kang11Dean Wu12Hsin-Chien Lee13Cheng-Jung Wu14Arnab Majumdar15Wen-Te Liu16Centre for Transport Studies, Department of Civil and Environmental Engineering, Imperial College London, London SW7 2AZ, UKDepartment of Medical Physics and Biomedical Engineering, University College London, London WC1E 6BT, UKParasites and Microbes Programme, Wellcome Sanger Institute, Hinxton CB10 1RQ, UKChemical Engineering and Biotechnology, University of Cambridge, Cambridge CB3 0AS, UKCognitive Neuroscience, Division of Psychology and Language Science, University College London, London WC1H 0AP, UKSchool of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei 110301, TaiwanCentre for Transport Studies, Department of Civil and Environmental Engineering, Imperial College London, London SW7 2AZ, UKSleep Center, Shuang Ho Hospital, Taipei Medical University, New Taipei City 235041, TaiwanDepartment of Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou, Taoyuan 33305, TaiwanDepartment of Neurology, Shuang Ho Hospital, Taipei Medical University, New Taipei City 235041, TaiwanDivision of Pulmonary Medicine, Department of Internal Medicine, Shuang Ho Hospital, Taipei Medical University, New Taipei City 235041, TaiwanDepartment of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei 110301, TaiwanSleep Center, Shuang Ho Hospital, Taipei Medical University, New Taipei City 235041, TaiwanDepartment of Psychiatry, Taipei Medical University Hospital, Taipei 110301, TaiwanDepartment of Otolaryngology, Shuang Ho Hospital, Taipei Medical University, New Taipei City 235041, TaiwanCentre for Transport Studies, Department of Civil and Environmental Engineering, Imperial College London, London SW7 2AZ, UKSchool of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei 110301, TaiwanObstructive sleep apnea (OSA) is a global health concern and is typically diagnosed using in-laboratory polysomnography (PSG). However, PSG is highly time-consuming and labor-intensive. We, therefore, developed machine learning models based on easily accessed anthropometric features to screen for the risk of moderate to severe and severe OSA. We enrolled 3503 patients from Taiwan and determined their PSG parameters and anthropometric features. Subsequently, we compared the mean values among patients with different OSA severity and considered correlations among all participants. We developed models based on the following machine learning approaches: logistic regression, k-nearest neighbors, naïve Bayes, random forest (RF), support vector machine, and XGBoost. Collected data were first independently split into two data sets (training and validation: 80%; testing: 20%). Thereafter, we adopted the model with the highest accuracy in the training and validation stage to predict the testing set. We explored the importance of each feature in the OSA risk screening by calculating the Shapley values of each input variable. The RF model achieved the highest accuracy for moderate to severe (84.74%) and severe (72.61%) OSA. The level of visceral fat was found to be a predominant feature in the risk screening models of OSA with the aforementioned levels of severity. Our machine learning models can be employed to screen for OSA risk in the populations in Taiwan and in those with similar craniofacial structures.https://www.mdpi.com/1424-8220/22/22/8630obstructive sleep apneapolysomnographyanthropometric featuresrandom forestvisceral fat level |
| spellingShingle | Cheng-Yu Tsai Huei-Tyng Huang Hsueh-Chien Cheng Jieni Wang Ping-Jung Duh Wen-Hua Hsu Marc Stettler Yi-Chun Kuan Yin-Tzu Lin Chia-Rung Hsu Kang-Yun Lee Jiunn-Horng Kang Dean Wu Hsin-Chien Lee Cheng-Jung Wu Arnab Majumdar Wen-Te Liu Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features Sensors obstructive sleep apnea polysomnography anthropometric features random forest visceral fat level |
| title | Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features |
| title_full | Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features |
| title_fullStr | Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features |
| title_full_unstemmed | Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features |
| title_short | Screening for Obstructive Sleep Apnea Risk by Using Machine Learning Approaches and Anthropometric Features |
| title_sort | screening for obstructive sleep apnea risk by using machine learning approaches and anthropometric features |
| topic | obstructive sleep apnea polysomnography anthropometric features random forest visceral fat level |
| url | https://www.mdpi.com/1424-8220/22/22/8630 |
| work_keys_str_mv | AT chengyutsai screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT hueitynghuang screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT hsuehchiencheng screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT jieniwang screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT pingjungduh screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT wenhuahsu screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT marcstettler screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT yichunkuan screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT yintzulin screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT chiarunghsu screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT kangyunlee screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT jiunnhorngkang screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT deanwu screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT hsinchienlee screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT chengjungwu screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT arnabmajumdar screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures AT wenteliu screeningforobstructivesleepapneariskbyusingmachinelearningapproachesandanthropometricfeatures |