Predicting Obstructive Sleep Apnea Based on Computed Tomography Scan Using Deep Learning Models.

RATIONALE: The incidence of clinically undiagnosed obstructive sleep apnea (OSA) is high among the general population due to limited access to polysomnography. Computed tomography (CT) of craniofacial regions obtained for other purposes can be beneficial in predicting OSA and its severity. OBJECTIVES: To predict OSA and its severity based on paranasal CT using a 3-dimensional deep learning algorithm. METHODS: One internal dataset (n=798) and two external datasets (n=135 and 85) were used in this study. In the internal dataset, 92 normal, 159 mild, 201 moderate, and 346 severe OSA participants were enrolled to derive the deep learning model. A multimodal deep learning model was elicited from the connection between a 3-dimensional convolutional neural network (CNN)-based part treating unstructured data (CT images) and a multi-layer perceptron (MLP)-based part treating structured data (age, sex, and body mass index) to predict OSA and its severity. MEASUREMENTS AND MAIN RESULTS: In four-class classification for predicting the severity of OSA, the AirwayNet-MM-H model (multimodal model with airway-highlighting preprocessing algorithm) showed an average accuracy of 87.6% (95% confidence interval [CI] 86.8-88.6) in the internal dataset and 84.0% (95% CI 83.0-85.1) and 86.3% (95% CI 85.3-87.3) in the two external datasets, respectively. In the two-class classification for predicting significant OSA (moderate to severe OSA), The area under the receiver operating characteristics (AUROC), accuracy, sensitivity, specificity, and F1 score were 0.910 (95% CI 0.899-0.922), 91.0% (95% CI 90.1-91.9), 89.9% (95% CI 88.8-90.9), 93.5% (95% CI 92.7-94.3), and 93.2% (95% CI 92.5-93.9), respectively, in the internal dataset. Furthermore, the diagnostic performance of the Airway Net-MM-H model outperformed that of the other six state-of-the-art deep learning models in terms of accuracy for both four- and two-class classifications and AUROC for two-class classification (p<0.001). CONCLUSIONS: A novel deep learning model, including a multimodal deep learning model and an airway-highlighting preprocessing algorithm from CT images obtained for other purposes, can provide significantly precise outcomes for OSA diagnosis.

PSMD1 as a prognostic marker and potential target in oropharyngeal cancer.

BACKGROUND: Despite the diverse genetic mutations in head and neck cancer, the chemotherapy outcome for this cancer has not improved for decades. It is urgent to select prognostic factors and therapeutic targets for oropharyngeal cancer to establish precision medicine. Recent studies have identified PSMD1 as a potential prognostic marker in several cancers. We aimed to assess the prognostic significance of PSMD1 expression in oropharyngeal squamous cell carcinoma (OPSCC) patients using immunohistochemistry. METHODS: We studied 64 individuals with OPSCC tissue from surgery at Seoul National University Bundang Hospital between April 2008 and August 2017. Immunostaining analysis was conducted on the tissue microarray (TMA) sections (4 µm) for p16 and PSMD1. H-score, which scale from 0 to 300, was calculated from each nucleus, cytoplasm, and cellular expression. Clinicopathological data were compared with Chi-squared test, Fisher's exact test, t-test, and logistic regression. Survival data until 2021 were achieved from national statistical office of Korea. Kaplan-Meier method and cox-regression model were used for disease-specific survival (DSS) analysis. RESULTS: H-score of 90 in nucleus was appropriate cutoff value for 'High PSMD1 expression' in OPSCC. Tonsil was more frequent location in low PSMD1 group (42/52, 80.8%) than in high PSMD1 group (4/12, 33.3%; P = .002). Early-stage tumor was more frequent in in low PSMD1 group (45/52, 86.5%) than in high PSMD1 group (6/12, 50%; P = .005). HPV was more positive in low PSMD1 group (43/52, 82.7%) than in high PSMD1 group (5/12, 41.7%; P = .016). Patients with PSMD1 high expression showed poorer DSS than in patients with PSMD1 low expression (P = .006 in log rank test). In multivariate analysis, PSMD1 expression, pathologic T staging, and specimen age were found to be associated with DSS (P = .011, P = .025, P = .029, respectively). CONCLUSIONS: In our study, we established PSMD1 as a negative prognostic factor in oropharyngeal squamous cell carcinoma, indicating its potential as a target for targeted therapy and paving the way for future in vitro studies on drug repositioning.

The clinical influence of nasal surgery on PAP compliance and optimal application among OSA subjects uncomfortable with PAP device wear.

This study aimed to evaluate the alteration of PAP compliance after nasal surgery and to determine the optimal indications of nasal surgery in obstructive sleep apnea (OSA) subjects. Among OSA subjects using PAP devices, 29 subjects who underwent septoturbinoplasty due to nasal obstruction were included and their pre- and postoperative medical and PAP records were reviewed retrospectively. Postoperative autoPAP usage data was further assessed by grouping the compliance (the percentage of days with usage ≥ 4 h) data (group 1: the good compliance group; group 2: the poor compliance group). The data showed that 56% of subjects in group 1 complained of nasal obstruction as the only barrier to using a PAP device and about 89% reported experiencing the efficacy of PAP usage. Both the mean and peak average PAP pressures were significantly reduced in group 1 following nasal surgery. Group 2 had multiple subjective problems that interfered with wearing a PAP device and reported a lack of experiencing the efficacy of PAP usage. Preoperative nasal cavity volume values were smaller and absolute blood eosinophil counts were significantly lower in group 1. The current data demonstrate that nasal surgery might increase the compliance of PAP device wear in OSA subjects who complained of only nasal obstruction as a barrier to wearing PAP and who had small nasal cavity volumes combined with allergic inflammation.

Automation of generative adversarial network-based synthetic data-augmentation for maximizing the diagnostic performance with paranasal imaging.

Thus far, there have been no reported specific rules for systematically determining the appropriate augmented sample size to optimize model performance when conducting data augmentation. In this paper, we report on the feasibility of synthetic data augmentation using generative adversarial networks (GAN) by proposing an automation pipeline to find the optimal multiple of data augmentation to achieve the best deep learning-based diagnostic performance in a limited dataset. We used Waters' view radiographs for patients diagnosed with chronic sinusitis to demonstrate the method developed herein. We demonstrate that our approach produces significantly better diagnostic performance parameters than models trained using conventional data augmentation. The deep learning method proposed in this study could be implemented to assist radiologists in improving their diagnosis. Researchers and industry workers could overcome the lack of training data by employing our proposed automation pipeline approach in GAN-based synthetic data augmentation. This is anticipated to provide new means to overcome the shortage of graphic data for algorithm training.

The effect of complex training on the children with all of the deformities including forward head, rounded shoulder posture, and lumbar lordosis.

The purpose of this study is to investigate the effect of complex training on children with the deformities including forward head, rounded shoulder posture, and lumbar lordosis. The complex training program was performed for 6 month three times per week. The complex training improved posture as measured by forward head angle (FHA), forward shoulder angle (FSA), and angle between anterior superior iliac spine and posterior superior iliac spine (APA). In the present results, complex training might overcome vertebral deformity through decreasing forward head, rounded shoulder posture, and lumbar lordosis and increasing flexibility in the children.