The effect of pharyngeal structures on the severity of obstructive sleep apnea.

BACKGROUND/OBJECTIVE: Obstructive sleep apnea (OSA) is characterized by complete or partial cessation of breathing during sleep. The tongue is suggested as a possible anatomical site causing airway obstruction. However, the role of other pharyngeal structures in the development of OSA remains unclear. We designed a study using both the apnea-hypopnea index (AHI) and the oxygen saturation measurements to assess the severity of OSA. We aimed to identify critical anatomical structures of the upper airway that correlate with the severity of OSA and to evaluate the utility of magnetic resonance imaging (MRI) markers to detect possible OSA in patients without overt symptoms. MATERIALS AND METHODS: The study included participants referred to the neurology outpatient clinic from the check-up unit. Participants were grouped as controls, mild, moderate, or severe OSA according to the AHI. A cranial MRI with a field of view (FOV) encompassing the upper airway structures was obtained from all participants. The areas of the tongue and the uvula were measured on the sagittal images by drawing the boundaries of the tissues manually. The posterior air space (PAS) area was evaluated from regions of interest in five parallel planes. RESULTS: Of 105 participants, 30 were controls, 27 had mild, 25 had moderate, and 23 had severe OSA. The moderate and severe OSA groups did not differ in oxygen saturation levels during sleep. Therefore, patients with moderate and severe OSA were combined into one group (moderate/severe OSA). The area of the tongue was significantly larger in the moderate/severe OSA group compared to the control group. Both the tongue and the uvula areas showed a significant positive correlation with the AHI. CONCLUSION: Our findings suggest that the tongue and uvula have prominent roles in the severity of OSAS. It may be useful to measure these structures with MRI to screen for at-risk individuals without overt OSA symptoms.

Radiogenomics of lower-grade gliomas: machine learning-based MRI texture analysis for predicting 1p/19q codeletion status.

OBJECTIVE: To evaluate the potential value of the machine learning (ML)-based MRI texture analysis for predicting 1p/19q codeletion status of lower-grade gliomas (LGG), using various state-of-the-art ML algorithms. MATERIALS AND METHODS: For this retrospective study, 107 patients with LGG were included from a public database. Texture features were extracted from conventional T2-weighted and contrast-enhanced T1-weighted MRI images, using LIFEx software. Training and unseen validation splits were created using stratified 10-fold cross-validation technique along with minority over-sampling. Dimension reduction was done using collinearity analysis and feature selection (ReliefF). Classifications were done using adaptive boosting, k-nearest neighbours, naive Bayes, neural network, random forest, stochastic gradient descent, and support vector machine. Friedman test and pairwise post hoc analyses were used for comparison of classification performances based on the area under the curve (AUC). RESULTS: Overall, the predictive performance of the ML algorithms were statistically significantly different, χ2(6) = 26.7, p < 0.001. There was no statistically significant difference among the performance of the neural network, naive Bayes, support vector machine, random forest, and stochastic gradient descent, adjusted p > 0.05. The mean AUC and accuracy values of these five algorithms ranged from 0.769 to 0.869 and from 80.1 to 84%, respectively. The neural network had the highest mean rank with mean AUC and accuracy values of 0.869 and 83.8%, respectively. CONCLUSIONS: The ML-based MRI texture analysis might be a promising non-invasive technique for predicting the 1p/19q codeletion status of LGGs. Using this technique along with various ML algorithms, more than four-fifths of the LGGs can be correctly classified. KEY POINTS: • More than four-fifths of the lower-grade gliomas can be correctly classified with machine learning-based MRI texture analysis. Satisfying classification outcomes are not limited to a single algorithm. • A few-slice-based volumetric segmentation technique would be a valid approach, providing satisfactory predictive textural information and avoiding excessive segmentation duration in clinical practice. • Feature selection is sensitive to different patient data set samples so that each sampling leads to the selection of different feature subsets, which needs to be considered in future works.

Evaluation of sporadic intracranial cavernous malformations for detecting associated developmental venous anomalies: added diagnostic value of C-arm contrast-enhanced cone-beam CT to routine contrast-enhanced MRI.

OBJECTIVE: Our purpose was to investigate the added diagnostic value of C-arm contrast-enhanced cone-beam CT (CE-CBCT) to routine contrast-enhanced MRI (CE-MRI) in detecting associated developmental venous anomalies (DVAs) in patients with sporadic intracranial cavernous malformations (ICMs). METHODS: Fifty-six patients (53 with single and three with double ICMs) met the inclusion criteria. All patients had routine CE-MRI scans performed at 1.5 Tesla. The imaging studies (CE-MRI and CE-CBCT) were retrospectively and independently reviewed by two observers, with consensus by a third. Group difference, intra- and interobserver agreement, and diagnostic performance of the modalities in detecting associated DVAs were calculated. Reference standard was CE-MRI. RESULTS: On CE-MRI and CE-CBCT, 37 (66%; of 56) and 47 patients (84%; of 56) had associated DVAs, respectively. In 10 patients (52.6%; of CE-MRI negatives [n=19]), CE-CBCT improved the diagnosis. Nine patients (16%; of 56) had no DVA on both imaging techniques. Difference in proportions of associated DVAs on CE-MRI and CE-CBCT was statistically significant, p < 0.05. Sensitivity, specificity, positive likelihood ratio, and area under the curve of CE-CBCT were 100% (95% confidence interval [CI]: 90.5-100%), 47.3% (95% CI: 24.4-71.1%), 1.9 (95%CI: 1.240-2.911), 0.737 (95%CI: 0.602-0.845), respectively. Intraobserver agreement was excellent for CE-MRI, kappa (κ) coefficient = 0.960, and CE-CBCT, κ=0.931. Interobserver agreement was substantial for CE-MRI, κ=0.803, and excellent for CE-CBCT, κ=0.810. CONCLUSIONS: CE-CBCT is a useful imaging technique especially in patients with negative routine CE-MRI in terms of detecting associated DVAs. In nearly half of these particular patients, it reveals an associated DVA as a new diagnosis. KEY POINTS: • Although it is known to be the gold standard, some of the DVAs associated with ICMs are underdiagnosed with CE-MRI. • In nearly half of the patients with negative routine CE-MRI, CE-CBCT reveals an associated DVA as a new diagnosis. • Intra- and interobserver agreement on CE-CBCT is excellent in terms of detecting associated DVAs.

Preoperative evaluation of tumour consistency in pituitary macroadenomas: a machine learning-based histogram analysis on conventional T2-weighted MRI.

PURPOSE: To evaluate the potential value of machine learning (ML)-based histogram analysis (or first-order texture analysis) on T2-weighted magnetic resonance imaging (MRI) for predicting consistency of pituitary macroadenomas (PMA) and to compare it with that of signal intensity ratio (SIR) evaluation. METHODS: Fifty-five patients with 13 hard and 42 soft PMAs were included in this retrospective study. Histogram features were extracted from coronal T2-weighted original, filtered and transformed MRI images by manual segmentation. To achieve balanced classes (38 hard vs 42 soft), multiple samples were obtained from different slices of the PMAs with hard consistency. Dimension reduction was done with reproducibility analysis, collinearity analysis and feature selection. ML classifier was artificial neural network (ANN). Reference standard for the classifications was based on surgical and histopathological findings. Predictive performance of histogram analysis was compared with that of SIR evaluation. The main metric for comparisons was the area under the receiver operating characteristic curve (AUC). RESULTS: Only 137 of 162 features had excellent reproducibility. Collinearity analysis yielded 20 features. Feature selection algorithm provided six texture features. For histogram analysis, the ANN correctly classified 72.5% of the PMAs regarding consistency with an AUC value of 0.710. For SIR evaluation, accuracy and AUC values were 74.5% and 0.551, respectively. Considering AUC values, ML-based histogram analysis performed better than SIR evaluation (z = 2.312, p = 0.021). CONCLUSION: ML-based T2-weighted MRI histogram analysis might be a useful technique in predicting the consistency of PMAs, with a better predictive performance than that of SIR evaluation.

CT Findings of the Massive Air Embolism in a Case with Thoracic Gunshot Wound: A Special Case Report as if a Whole Body Pneumo- Angiography was Performed.

BACKGROUND: Massive Systemic Arterial Air Embolism (SAAE) associated with penetrating trauma is a rare condition. A few cases were reported for massive arterial air embolism in the literature. Computed tomography is a fast and easily accessible modality for detecting air in the vasculature. We report CT findings of a rare case with a thoracic gunshot wound, which demonstrate air almost in all systemic vessels like ''full body pneumoangiography''. CASE PRESENTATION: A 42-year-old male patient with a thoracic gunshot wound was admitted to the Accident and Emergency (A&E) unit in a state of cardiac arrest. Postmortem Computed Tomography (CT) was performed and extensive air was revealed in several great vessels. CONCLUSION: We conclude that the underline causes of massive air embolism in our case are two main mechanisms: firstly, massive air enters the vasculature via bronchovascular fistula as there was bilateral lung contusion and directly through cardiac truncus, secondly while CPR was being conducted, massive air was pumped to the systemic circulation.