Development of Pericardial Fat Count Images Using a Combination of Three Different Deep-Learning Models: Image Translation Model From Chest Radiograph Image to Projection Image of Three-Dimensional Computed Tomography.

RATIONALE AND OBJECTIVES: Pericardial fat (PF)-the thoracic visceral fat surrounding the heart-promotes the development of coronary artery disease by inducing inflammation of the coronary arteries. To evaluate PF, we generated pericardial fat count images (PFCIs) from chest radiographs (CXRs) using a dedicated deep-learning model. MATERIALS AND METHODS: We reviewed data of 269 consecutive patients who underwent coronary computed tomography (CT). We excluded patients with metal implants, pleural effusion, history of thoracic surgery, or malignancy. Thus, the data of 191 patients were used. We generated PFCIs from the projection of three-dimensional CT images, wherein fat accumulation was represented by a high pixel value. Three different deep-learning models, including CycleGAN were combined in the proposed method to generate PFCIs from CXRs. A single CycleGAN-based model was used to generate PFCIs from CXRs for comparison with the proposed method. To evaluate the image quality of the generated PFCIs, structural similarity index measure (SSIM), mean squared error (MSE), and mean absolute error (MAE) of (i) the PFCI generated using the proposed method and (ii) the PFCI generated using the single model were compared. RESULTS: The mean SSIM, MSE, and MAE were 8.56 × 10-1, 1.28 × 10-2, and 3.57 × 10-2, respectively, for the proposed model, and 7.62 × 10-1, 1.98 × 10-2, and 5.04 × 10-2, respectively, for the single CycleGAN-based model. CONCLUSION: PFCIs generated from CXRs with the proposed model showed better performance than those generated with the single model. The evaluation of PF without CT may be possible using the proposed method.

Improvement of sleep parameters by titration polysomnography could predict adherence to positive airway pressure therapy in obstructive sleep apnea.

STUDY OBJECTIVES: Poor adherence to continuous positive airway pressure (CPAP) has been a critical issue in treating obstructive sleep apnea. Because long-term CPAP adherence may be established shortly after treatment begins, early intervention is essential. This study aimed to identify the potential factors affecting CPAP therapy adherence during diagnostic polysomnography and auto CPAP titration polysomnography. METHODS: This retrospective observational study included 463 patients with obstructive sleep apnea who underwent consecutive diagnostic polysomnography and titration polysomnography. We recorded their demographic, anthropometric, and lifestyle factors and obtained self-reported comments regarding their sleep status following both polysomnography evaluations. CPAP adherence was evaluated following 3 months of treatment. RESULTS: A total of 312 patients (67.4%) fulfilled the criteria for good adherence. Each patient's CPAP adherence was categorized as "poor" (< 4 hours/night or <70% of nights), "good" (≥ 4 hours/night and ≥ 70% of nights), or "excellent" (≥ 6 hours/night and ≥ 80% of nights). There were no significant differences in arterial oxyhemoglobin saturation measured by pulse oximetry and apnea-hypopnea index during diagnostic polysomnography among 3 groups. The polysomnographic evaluations indicated that patients with better adherence displayed more significant improvements in sleep parameters, including apnea-hypopnea index, sleep efficacy, sleep latency, and sleep architecture, which were correlated with an improvement in self-reported sleep quality. CONCLUSIONS: Polysomnographic evaluations enabled CPAP adherence prediction and a comparison of self-reported sleep quality with and without CPAP; CPAP adherence led to improvements in polysomnographic parameters. Our findings suggest that titration polysomnography and self-reported sleep improvement with CPAP could be used for adherence prediction in clinical practice. CITATION: Shirahata T, Uchida Y, Uchida T, et al. Improvement of sleep parameters by titration polysomnography could predict adherence to positive airway pressure therapy in obstructive sleep apnea. J Clin Sleep Med. 2023;19(8):1465-1473.

The efficacy of 18F-FDG-PET-based radiomic and deep-learning features using a machine-learning approach to predict the pathological risk subtypes of thymic epithelial tumors.

OBJECTIVE: To examine whether the machine-learning approach using 18-fludeoxyglucose positron emission tomography (18F-FDG-PET)-based radiomic and deep-learning features is useful for predicting the pathological risk subtypes of thymic epithelial tumors (TETs). METHODS: This retrospective study included 79 TET [27 low-risk thymomas (types A, AB and B1), 31 high-risk thymomas (types B2 and B3) and 21 thymic carcinomas] patients who underwent pre-therapeutic 18F-FDG-PET/CT. High-risk TETs (high-risk thymomas and thymic carcinomas) were 52 patients. The 107 PET-based radiomic features, including SUV-related parameters [maximum SUV (SUVmax), metabolic tumor volume (MTV), and total lesion glycolysis (TLG)] and 1024 deep-learning features extracted from the convolutional neural network were used to predict the pathological risk subtypes of TETs using six different machine-learning algorithms. The area under the curves (AUCs) were calculated to compare the predictive performances. RESULTS: SUV-related parameters yielded the following AUCs for predicting thymic carcinomas: SUVmax 0.713, MTV 0.442, and TLG 0.479 or high-risk TETs: SUVmax 0.673, MTV 0.533, and TLG 0.539. The best-performing algorithm was the logistic regression model for predicting thymic carcinomas (AUC 0.900, accuracy 81.0%), and the random forest (RF) model for high-risk TETs (AUC 0.744, accuracy 72.2%). The AUC was significantly higher in the logistic regression model than three SUV-related parameters for predicting thymic carcinomas, and in the RF model than MTV and TLG for predicting high-risk TETs (each; p < 0.05). CONCLUSION: 18F-FDG-PET-based radiomic analysis using a machine-learning approach may be useful for predicting the pathological risk subtypes of TETs. ADVANCES IN KNOWLEDGE: Machine-learning approach using 18F-FDG-PET-based radiomic features has the potential to predict the pathological risk subtypes of TETs.

Predicting the presence and severity of obstructive sleep apnea based on mandibular measurements using quantitative analysis of facial profiles via three-dimensional photogrammetry.

BACKGROUND: In obstructive sleep apnea (OSA), the upper airway is obstructed during sleep due to obesity and/or posterior collapse of the tongue root. Maxillofacial morphological abnormalities increase the risk of OSA in the Asian population. This study sought to elucidate whether three-dimensional (3D) photogrammetry measurements correlate with the severity of OSA irrespective of sex and degree of obesity. METHODS: A prospective pilot study was performed, in which 37 consecutive adult patients (M/F = 28/9) underwent polysomnography and 3D photogrammetry in the supine position for the diagnosis of OSA. Measurements obtained from 3D photogrammetry included mandibular width (Mw), mandibular length (Ml), mandibular depth (Md), mandibular width-length angle (Mwla), and mandibular area (Ma). The effects of sex and body mass index (BMI) on the measurements and their association with the apnea-hypopnea index (AHI) were statistically analyzed. The inter-rater reliability of the measurements was evaluated using intraclass correlation coefficients (ICC). RESULTS: Mwla (R = 0.73, p < 0.01), Mw (R = 0.39, p < 0.05), and Md (R = -0.34, p < 0.05) were significantly correlated with the severity of OSA. On multivariate analysis, Mwla (p < 0.01) and Md (p < 0.05) remained independent factors for AHI after adjusting for sex, age, BMI, and neck circumference. In addition, diagnosability analysis revealed that Mwla was useful for identifying the presence of OSA (AHI ≥5) (cutoff: 78.6°, sensitivity: 0.938, specificity: 0.800, area under the curve: 0.931). The ICC was >0.9, showing high reliability. CONCLUSIONS: This study suggests that Mwla measured using 3D photogrammetry can predict the presence of OSA and correlates with the severity of OSA, independent of obesity and sex.