Wun-Hao Cheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wun-Hao Cheng has a total of 47 co-authors. Here are the top co-authors by number of shared publications:
- Cheng-Yu Tsai
Research Center of Sleep Medicine, College of Medicine, Taipei Medical University, Taipei 11031, Taiwan
2 shared publications - Jie‐Syuan Wu
School of Medicine, College of Medicine Taipei Medical University Taipei Taiwan
2 shared publications - Lee‐Yuan Lin
Department of Medical Education Chang Gung Memorial Hospital Keelung Taiwan
2 shared publications - Wen-Hua Hsu
School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei 110301, Taiwan
2 shared publications Ying-Jung Wu
School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
2 shared publications- Arnab Majumdar
Imperial College London
1 shared publication
Latest Publications
- 5 September 2026
- Published by European Respiratory Society (ERS)
- Vol. 68 (suppl 70) , PA1113
Myristoylated, alanine-rich C-kinase substrate (MARCKS) mediates anti-apoptosis of airway fibroblast and airway remodeling in obese asthma
Backgroud: Obesity is a crucial phenotype of asthma. Myristoylated alanine-rich C-kinase substrate (MARCKS) is known to play a crucial role in mucin secretion in airway epithelial cells. However, the mechanisms through which obesity exacerbates airway remodeling in asthma remain understood. Methods: Bronchial tissues were obtained from healthy subjects, non-obese asthma patients, and obese asthma patients. MARCKS expression in bronchial tissues was analyzed by immunohistochemistry and western blotting. Normal human lung fibroblasts (NHLFs) and fibroblasts from patients with severe asthma were treated with oxidized low-density lipoprotein (ox-LDL), with or without TLR4 or MARCKS shRNA knockdown. Mitochondrial membrane potential (mΔψ) was assessed to evaluate mitochondrial dysfunction and MOMP, and fibrotic and anti-apoptotic protein expression was analyzed by western blot. Results: Our data showed elevated MARCKS expression in the subepithelial layer of bronchial tissues from obese asthma patients compared to non-obese asthma patients and healthy subject. Additionally, ox-LDL exacerbated fibrotic protein expression and induced MARCKS expression in fibroblasts from patients with severe asthma. ox-LDL-induced minority MOMP and anti-apoptosis through the TLR4/MARCKS/NF-κB signaling pathway in NHLFs, Conclusion: MARCKS exacerbates airway fibrosis in obese asthma by promoting minority MOMP and anti-apoptosis.
- 5 September 2026
- Published by European Respiratory Society (ERS)
- Vol. 68 (suppl 70) , PA1113
Myristoylated, alanine-rich C-kinase substrate (MARCKS) mediates anti-apoptosis of airway fibroblast and airway remodeling in obese asthma
Backgroud: Obesity is a crucial phenotype of asthma. Myristoylated alanine-rich C-kinase substrate (MARCKS) is known to play a crucial role in mucin secretion in airway epithelial cells. However, the mechanisms through which obesity exacerbates airway remodeling in asthma remain understood. Methods: Bronchial tissues were obtained from healthy subjects, non-obese asthma patients, and obese asthma patients. MARCKS expression in bronchial tissues was analyzed by immunohistochemistry and western blotting. Normal human lung fibroblasts (NHLFs) and fibroblasts from patients with severe asthma were treated with oxidized low-density lipoprotein (ox-LDL), with or without TLR4 or MARCKS shRNA knockdown. Mitochondrial membrane potential (mΔψ) was assessed to evaluate mitochondrial dysfunction and MOMP, and fibrotic and anti-apoptotic protein expression was analyzed by western blot. Results: Our data showed elevated MARCKS expression in the subepithelial layer of bronchial tissues from obese asthma patients compared to non-obese asthma patients and healthy subject. Additionally, ox-LDL exacerbated fibrotic protein expression and induced MARCKS expression in fibroblasts from patients with severe asthma. ox-LDL-induced minority MOMP and anti-apoptosis through the TLR4/MARCKS/NF-κB signaling pathway in NHLFs, Conclusion: MARCKS exacerbates airway fibrosis in obese asthma by promoting minority MOMP and anti-apoptosis.
- Open Access
- 22 May 2026
- Published by Springer Nature in Journal of Biomedical Science
- Vol. 33 (1) , 52
Doublecortin-like kinase 1 promotes fibroblast activation and fibrotic progression through Smad3 binding in idiopathic pulmonary fibrosis
Background: Idiopathic pulmonary fibrosis (IPF) is a progressive, fatal lung disease with limited treatment options. Although the results of a gene expression analysis revealed that lung tissues in individuals with IPF contain high levels of doublecortin-like kinase 1 (DCLK1), the role of DCLK1 in fibroblast activation remains unclear. The present study examined the function of DCLK1 in IPF and bleomycin-induced pulmonary fibrosis, exploring its potential as a therapeutic target warranting further investigation. Methods: We investigated the role of DCLK1 in fibroblast activation and pulmonary fibrosis in lung tissues from patients with IPF, a bleomycin-induced pulmonary fibrosis mouse model, and cultured normal human lung fibroblasts. DCLK1 knockout mice and mice treated with the selective DCLK1 inhibitor DCLK1-IN-1 were used to evaluate the effects of genetic DCLK1 deficiency and pharmacological DCLK1 inhibition on fibrotic progression and lung function. Results: DCLK1 was markedly upregulated in the IPF lung tissues and in bleomycin-induced fibrotic lung tissues. Global deletion of Dclk1 considerably attenuated fibrotic remodeling and preserved lung function in mice. In addition, transforming growth factor β (TGF-β) induced DCLK1 expression in normal human lung fibroblasts through Smad3 and NF-κB signaling, while Akt/DCLK1/Smad3 signaling was associated with fibroblast activation and profibrotic marker expression. Moreover, DCLK1 was associated with Smad3, and these findings were consistent with DCLK1–Smad3-associated signaling linked to connective tissue growth factor expression. Finally, oral administration of DCLK1-IN-1 slowed fibrotic progression and preserved lung function in bleomycin-treated mice. Conclusions: DCLK1 is associated with fibroblast activation and pulmonary fibrosis, with findings consistent with DCLK1–Smad3-associated signaling linked to profibrotic marker expression. Genetic deletion or pharmacological inhibition of DCLK1 attenuated fibrotic progression and preserved lung function, suggesting that DCLK1 warrants further investigation as a potential therapeutic target in pulmonary fibrosis.
- 22 May 2026
- Published by Springer Nature in Journal of Biomedical Science
- Vol. 33 (1) , 52
Open AccessDoublecortin-like kinase 1 promotes fibroblast activation and fibrotic progression through Smad3 binding in idiopathic pulmonary fibrosis
Background: Idiopathic pulmonary fibrosis (IPF) is a progressive, fatal lung disease with limited treatment options. Although the results of a gene expression analysis revealed that lung tissues in individuals with IPF contain high levels of doublecortin-like kinase 1 (DCLK1), the role of DCLK1 in fibroblast activation remains unclear. The present study examined the function of DCLK1 in IPF and bleomycin-induced pulmonary fibrosis, exploring its potential as a therapeutic target warranting further investigation. Methods: We investigated the role of DCLK1 in fibroblast activation and pulmonary fibrosis in lung tissues from patients with IPF, a bleomycin-induced pulmonary fibrosis mouse model, and cultured normal human lung fibroblasts. DCLK1 knockout mice and mice treated with the selective DCLK1 inhibitor DCLK1-IN-1 were used to evaluate the effects of genetic DCLK1 deficiency and pharmacological DCLK1 inhibition on fibrotic progression and lung function. Results: DCLK1 was markedly upregulated in the IPF lung tissues and in bleomycin-induced fibrotic lung tissues. Global deletion of Dclk1 considerably attenuated fibrotic remodeling and preserved lung function in mice. In addition, transforming growth factor β (TGF-β) induced DCLK1 expression in normal human lung fibroblasts through Smad3 and NF-κB signaling, while Akt/DCLK1/Smad3 signaling was associated with fibroblast activation and profibrotic marker expression. Moreover, DCLK1 was associated with Smad3, and these findings were consistent with DCLK1–Smad3-associated signaling linked to connective tissue growth factor expression. Finally, oral administration of DCLK1-IN-1 slowed fibrotic progression and preserved lung function in bleomycin-treated mice. Conclusions: DCLK1 is associated with fibroblast activation and pulmonary fibrosis, with findings consistent with DCLK1–Smad3-associated signaling linked to profibrotic marker expression. Genetic deletion or pharmacological inhibition of DCLK1 attenuated fibrotic progression and preserved lung function, suggesting that DCLK1 warrants further investigation as a potential therapeutic target in pulmonary fibrosis.
- Open Access
- 18 February 2026
- Published by Springer Nature in Inflammation
- Vol. 49 (1)
FNDC5/Irisin Attenuates TGF-β-induced Fibroblast Differentiation Via AMPK/Nrf2 Signaling and Mitochondrial Stabilization in Subepithelial Fibrosis
Severe asthma is characterized by subepithelial fibrosis and mitochondrial dysfunction in bronchial fibroblasts. Fibronectin type III domain-containing 5 (FNDC5)/irisin modulates metabolism and protects mitochondria during inflammation, but its role in fibroblast differentiation remains unclear. In this study, we assessed FNDC5 expression in bronchial tissues from house dust mite (HDM)–induced asthmatic mice and controls by immunohistochemistry, analyzed fibrotic proteins by Western blotting, performed RNA sequencing to identify antioxidant pathway–associated genes, evaluated mitochondrial function using JC-1 staining, and measured airway hyperresponsiveness (AHR) with the FlexiVent system. FNDC5 levels were reduced in asthma-derived lung fibroblasts and in HDM-induced asthmatic mice compared to controls. Treatment with irisin decreased α-SMA expression, improved mitochondrial function, and suppressed TGF-β–induced fibrotic protein expression and Smad3 phosphorylation in human lung fibroblasts. RNA sequencing confirmed that irisin enhanced antioxidant-related gene expression and activated the AMPK pathway, which in turn reduced TGF-β–induced Smad3 phosphorylation and fibronectin expression. Moreover, irisin induced Nrf2 phosphorylation, and Nrf2 knockdown diminished its protective effects on fibrosis markers. In vivo, irisin treatment alleviated HDM-induced airway fibrosis, inflammation, and lung function impairments. Together, these findings demonstrate that FNDC5/irisin protects against airway fibrosis in patients with asthma by activating AMPK and Nrf2 signaling pathways, thereby preserving mitochondrial function and reducing fibroblast differentiation, highlighting FNDC5/irisin as a potential therapeutic target for airway remodeling in asthma.
- 18 February 2026
- Published by Springer Nature in Inflammation
- Vol. 49 (1)
Open AccessFNDC5/Irisin Attenuates TGF-β-induced Fibroblast Differentiation Via AMPK/Nrf2 Signaling and Mitochondrial Stabilization in Subepithelial Fibrosis
Severe asthma is characterized by subepithelial fibrosis and mitochondrial dysfunction in bronchial fibroblasts. Fibronectin type III domain-containing 5 (FNDC5)/irisin modulates metabolism and protects mitochondria during inflammation, but its role in fibroblast differentiation remains unclear. In this study, we assessed FNDC5 expression in bronchial tissues from house dust mite (HDM)–induced asthmatic mice and controls by immunohistochemistry, analyzed fibrotic proteins by Western blotting, performed RNA sequencing to identify antioxidant pathway–associated genes, evaluated mitochondrial function using JC-1 staining, and measured airway hyperresponsiveness (AHR) with the FlexiVent system. FNDC5 levels were reduced in asthma-derived lung fibroblasts and in HDM-induced asthmatic mice compared to controls. Treatment with irisin decreased α-SMA expression, improved mitochondrial function, and suppressed TGF-β–induced fibrotic protein expression and Smad3 phosphorylation in human lung fibroblasts. RNA sequencing confirmed that irisin enhanced antioxidant-related gene expression and activated the AMPK pathway, which in turn reduced TGF-β–induced Smad3 phosphorylation and fibronectin expression. Moreover, irisin induced Nrf2 phosphorylation, and Nrf2 knockdown diminished its protective effects on fibrosis markers. In vivo, irisin treatment alleviated HDM-induced airway fibrosis, inflammation, and lung function impairments. Together, these findings demonstrate that FNDC5/irisin protects against airway fibrosis in patients with asthma by activating AMPK and Nrf2 signaling pathways, thereby preserving mitochondrial function and reducing fibroblast differentiation, highlighting FNDC5/irisin as a potential therapeutic target for airway remodeling in asthma.
- 29 August 2025
- Published by Elsevier in Sleep Medicine
- Vol. 136, 106768
Sleep apnea in interstitial lung disease: A systematic review and meta-analysis of prevalence, severity, and risk factors
- 29 August 2025
- Published by Elsevier in Sleep Medicine
- Vol. 136, 106768
Sleep apnea in interstitial lung disease: A systematic review and meta-analysis of prevalence, severity, and risk factors
- Open Access
- 17 October 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13 (20) , 6190
Predicting Survival Status in COVID-19 Patients: Machine Learning Models Development with Ventilator-Related and Biochemical Parameters from Early Stages: A Pilot Study
Objective: Coronavirus disease 2019 (COVID-19) can cause intubation and ventilatory support due to respiratory failure, and extubation failure increases mortality risk. This study, therefore, aimed to explore the feasibility of using specific biochemical and ventilator parameters to predict survival status among COVID-19 patients by using machine learning. Methods: This study included COVID-19 patients from Taipei Medical University-affiliated hospitals from May 2021 to May 2022. Sequential data on specific biochemical and ventilator parameters from days 0–2, 3–5, and 6–7 were analyzed to explore differences between the surviving (successfully weaned off the ventilator) and non-surviving groups. These data were further used to establish separate survival prediction models using random forest (RF). Results: The surviving group exhibited significantly lower mean C-reactive protein (CRP) levels and mean potential of hydrogen ions levels (pH) levels on days 0–2 compared to the non-surviving group (CRP: non-surviving group: 13.16 ± 5.15 ng/mL, surviving group: 10.23 ± 5.15 ng/mL; pH: non-surviving group: 7.32 ± 0.07, survival group: 7.37 ± 0.07). Regarding the survival prediction performanace, the RF model trained solely with data from days 0–2 outperformed models trained with data from days 3–5 and 6–7. Subsequently, CRP, the partial pressure of carbon dioxide in arterial blood (PaCO2), pH, and the arterial oxygen partial pressure to fractional inspired oxygen (P/F) ratio served as primary indicators in survival prediction in the day 0–2 model. Conclusions: The present developed models confirmed that early biochemical and ventilatory parameters—specifically, CRP levels, pH, PaCO2, and P/F ratio—were key predictors of survival for COVID-19 patients. Assessed during the initial two days, these indicators effectively predicted the likelihood of successful weaning of from ventilators, emphasizing their importance in early management and improved outcomes in COVID-19-related respiratory failure.
- 17 October 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13 (20) , 6190
Open AccessPredicting Survival Status in COVID-19 Patients: Machine Learning Models Development with Ventilator-Related and Biochemical Parameters from Early Stages: A Pilot Study
Objective: Coronavirus disease 2019 (COVID-19) can cause intubation and ventilatory support due to respiratory failure, and extubation failure increases mortality risk. This study, therefore, aimed to explore the feasibility of using specific biochemical and ventilator parameters to predict survival status among COVID-19 patients by using machine learning. Methods: This study included COVID-19 patients from Taipei Medical University-affiliated hospitals from May 2021 to May 2022. Sequential data on specific biochemical and ventilator parameters from days 0–2, 3–5, and 6–7 were analyzed to explore differences between the surviving (successfully weaned off the ventilator) and non-surviving groups. These data were further used to establish separate survival prediction models using random forest (RF). Results: The surviving group exhibited significantly lower mean C-reactive protein (CRP) levels and mean potential of hydrogen ions levels (pH) levels on days 0–2 compared to the non-surviving group (CRP: non-surviving group: 13.16 ± 5.15 ng/mL, surviving group: 10.23 ± 5.15 ng/mL; pH: non-surviving group: 7.32 ± 0.07, survival group: 7.37 ± 0.07). Regarding the survival prediction performanace, the RF model trained solely with data from days 0–2 outperformed models trained with data from days 3–5 and 6–7. Subsequently, CRP, the partial pressure of carbon dioxide in arterial blood (PaCO2), pH, and the arterial oxygen partial pressure to fractional inspired oxygen (P/F) ratio served as primary indicators in survival prediction in the day 0–2 model. Conclusions: The present developed models confirmed that early biochemical and ventilatory parameters—specifically, CRP levels, pH, PaCO2, and P/F ratio—were key predictors of survival for COVID-19 patients. Assessed during the initial two days, these indicators effectively predicted the likelihood of successful weaning of from ventilators, emphasizing their importance in early management and improved outcomes in COVID-19-related respiratory failure.
- 14 September 2024
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 64 (suppl 68) , PA830
FNDC5/irisin regulates mitochondrial function in airway subepithelial fibrosis of severe asthma
Severe asthma is characterized by persistent airway inflammation and subepithelial fibrosis. Mitochondrial dysfunction plays a crucial role in bronchial fibroblast differentiation and resistance to apoptosis, exacerbating subepithelial fibrosis. Fibronectin type III domain containing 5 (FNDC5)/irisin plays a crucial role in modulating metabolism and protecting mitochondria during inflammation through AMP-activated protein kinase (AMPK) phosphorylation, potentially alleviating the progression of chronic diseases. AMPK activation induces osteoblast differentiation via the Smad1/5/9 pathway. However, whether TGF-β reduces FNDC5 via the AMPK/smad1/5/9 pathway in normal human lung fibroblasts (NHLFs) remains unknown. This research project aims to explore the role of FNDC5/irisin in alleviating airway subepithelial fibrosis by regulating mitochondrial function in severe asthma. Preliminary data indicated reduced FNDC5 expression in severe asthma bronchial tissue and fibroblasts. Irisin downregulated α-SMA expression dose-dependently, alleviating mitochondrial dysfunction in asthma fibroblasts. FNDC5 expression was decreased in OVA-induced airway fibrosis mice. RNA-sequence analysis showed reduced Fndc5 expression in OVA-induced asthmatic mice. TGF-β decreased FNDC5 expression and AMPK phosphorylation in a time-dependent manner in NHLFs. Moreover, irisin decreased TGF-β-induced α-SMA and fibronectin expression and Smad3 phosphorylation in NHLFs. Additionally, irisin affected mitochondrial-associated cellular components, biological processes, and molecular functions compared to TGF-β-treated NHLFs. Therefore, this research project's central hypothesis is that FNDC5/irisin down-regulates TGF-β-induced mitochondrial dysfunction in subepithelial fibrosis in severe asthma.
- 14 September 2024
- Published by European Respiratory Society (ERS) in European Respiratory Journal
- Vol. 64 (suppl 68) , PA830
FNDC5/irisin regulates mitochondrial function in airway subepithelial fibrosis of severe asthma
Severe asthma is characterized by persistent airway inflammation and subepithelial fibrosis. Mitochondrial dysfunction plays a crucial role in bronchial fibroblast differentiation and resistance to apoptosis, exacerbating subepithelial fibrosis. Fibronectin type III domain containing 5 (FNDC5)/irisin plays a crucial role in modulating metabolism and protecting mitochondria during inflammation through AMP-activated protein kinase (AMPK) phosphorylation, potentially alleviating the progression of chronic diseases. AMPK activation induces osteoblast differentiation via the Smad1/5/9 pathway. However, whether TGF-β reduces FNDC5 via the AMPK/smad1/5/9 pathway in normal human lung fibroblasts (NHLFs) remains unknown. This research project aims to explore the role of FNDC5/irisin in alleviating airway subepithelial fibrosis by regulating mitochondrial function in severe asthma. Preliminary data indicated reduced FNDC5 expression in severe asthma bronchial tissue and fibroblasts. Irisin downregulated α-SMA expression dose-dependently, alleviating mitochondrial dysfunction in asthma fibroblasts. FNDC5 expression was decreased in OVA-induced airway fibrosis mice. RNA-sequence analysis showed reduced Fndc5 expression in OVA-induced asthmatic mice. TGF-β decreased FNDC5 expression and AMPK phosphorylation in a time-dependent manner in NHLFs. Moreover, irisin decreased TGF-β-induced α-SMA and fibronectin expression and Smad3 phosphorylation in NHLFs. Additionally, irisin affected mitochondrial-associated cellular components, biological processes, and molecular functions compared to TGF-β-treated NHLFs. Therefore, this research project's central hypothesis is that FNDC5/irisin down-regulates TGF-β-induced mitochondrial dysfunction in subepithelial fibrosis in severe asthma.
- 6 May 2024
- Published by Elsevier in Respiratory Medicine
- Vol. 227, 107657
Oxygen therapy for exercise capacity in fibrotic interstitial lung disease: A systematic review and meta-analysis of randomised controlled trials
- 6 May 2024
- Published by Elsevier in Respiratory Medicine
- Vol. 227, 107657
Oxygen therapy for exercise capacity in fibrotic interstitial lung disease: A systematic review and meta-analysis of randomised controlled trials
- Open Access
- 14 October 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Screening the Risk of Obstructive Sleep Apnea by Utilizing Supervised Learning Techniques Based on Anthropometric Features and Snoring Events (Preprint)
BACKGROUND Obstructive sleep apnea (OSA) is sleep-disordered breathing and is typically diagnosed by polysomnography (PSG). However, PSG is time-consuming and has some clinical limitations. OBJECTIVE This study thus aimed to establish machine learning models to screen for the risk of having moderate-to-severe and severe OSA based on easily acquired features (e.g., symptoms or risk factors of OSA). METHODS This retrospective study collected data on 3629 patients from Taiwan, who had undergone PSG for symptoms of OSA. Their baseline characteristics, anthropometric measures, and PSG data were obtained. The number of snoring events of PSG was further derived, and correlations among the collected variables were investigated. Next, this study utilized six common supervised machine learning techniques to establish OSA risk screening models, including random forest (RF), XGBoost, k-nearest neighbors, support vector machine, logistic regression, and naïve Bayes. First, data were independently separated into a training and validation dataset (80%) and a test dataset (20%). The approach which had the highest accuracy in the training and validation phase was employed to perform the classification for the test dataset. Moreover, the feature importance of employed models was determined by calculating the Shapley value of every factor, which represented the impact on OSA risk screening. RESULTS RF models manifested the highest accuracy and area under the receiver operating characteristic curve (AUC) in the training and validation phase in screening for both OSA severities (over 70% of accuracy and over 80% of AUC). Hence, we employed the RF to perform the classification of the independent test dataset, and results showed 79.32% accuracy for moderate-to-severe OSA and 74.37% accuracy for severe OSA. Next, snoring events and the visceral fat level were the most and second-most essential features of screening for OSA risk. CONCLUSIONS Snoring events and the visceral fat level were essential features of screening for OSA risk. Based on these easily assessed variables, this study established models that can be considered to apply for screening OSA risk in populations with similar craniofacial features.
- 14 October 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessScreening the Risk of Obstructive Sleep Apnea by Utilizing Supervised Learning Techniques Based on Anthropometric Features and Snoring Events (Preprint)
BACKGROUND Obstructive sleep apnea (OSA) is sleep-disordered breathing and is typically diagnosed by polysomnography (PSG). However, PSG is time-consuming and has some clinical limitations. OBJECTIVE This study thus aimed to establish machine learning models to screen for the risk of having moderate-to-severe and severe OSA based on easily acquired features (e.g., symptoms or risk factors of OSA). METHODS This retrospective study collected data on 3629 patients from Taiwan, who had undergone PSG for symptoms of OSA. Their baseline characteristics, anthropometric measures, and PSG data were obtained. The number of snoring events of PSG was further derived, and correlations among the collected variables were investigated. Next, this study utilized six common supervised machine learning techniques to establish OSA risk screening models, including random forest (RF), XGBoost, k-nearest neighbors, support vector machine, logistic regression, and naïve Bayes. First, data were independently separated into a training and validation dataset (80%) and a test dataset (20%). The approach which had the highest accuracy in the training and validation phase was employed to perform the classification for the test dataset. Moreover, the feature importance of employed models was determined by calculating the Shapley value of every factor, which represented the impact on OSA risk screening. RESULTS RF models manifested the highest accuracy and area under the receiver operating characteristic curve (AUC) in the training and validation phase in screening for both OSA severities (over 70% of accuracy and over 80% of AUC). Hence, we employed the RF to perform the classification of the independent test dataset, and results showed 79.32% accuracy for moderate-to-severe OSA and 74.37% accuracy for severe OSA. Next, snoring events and the visceral fat level were the most and second-most essential features of screening for OSA risk. CONCLUSIONS Snoring events and the visceral fat level were essential features of screening for OSA risk. Based on these easily assessed variables, this study established models that can be considered to apply for screening OSA risk in populations with similar craniofacial features.