Introduction: Obstructive sleep apnea syndrome has become an important public health concern. Polysomnography is traditionally considered an established and effective diagnostic tool providing information on the severity of obstructive sleep apnea syndrome and the degree of sleep fragmentation. However, the numerous steps in the polysomnography test to diagnose obstructive sleep apnea syndrome are costly and time consuming. This study aimed to test the efficacy and clinical applicability of different machine learning methods based on demographic information and questionnaire data to predict obstructive sleep apnea syndrome severity. Materials and methods: We collected data about demographic characteristics, spirometry values, gas exchange (PaO2, PaCO2) and symptoms (Epworth Sleepiness Scale, snoring, etc.) of 313 patients with previous diagnosis of obstructive sleep apnea syndrome. After principal component analysis, we selected 19 variables which were used for further preprocessing and to eventually train seven types of classification models and five types of regression models to evaluate the prediction ability of obstructive sleep apnea syndrome severity, represented either by class or by apnea–hypopnea index. All models are trained with an increasing number of features and the results are validated through stratified 10-fold cross validation. Results: Comparative results show the superiority of support vector machine and random forest models for classification, while support vector machine and linear regression are better suited to predict apnea–hypopnea index. Also, a limited number of features are enough to achieve the maximum predictive accuracy. The best average classification accuracy on test sets is 44.7 percent, with the same average sensitivity (recall). In only 5.7 percent of cases, a severe obstructive sleep apnea syndrome (class 4) is misclassified as mild (class 2). Regression results show a minimum achieved root mean squared error of 22.17. Conclusion: The problem of predicting apnea–hypopnea index or severity classes for obstructive sleep apnea syndrome is very difficult when using only data collected prior to polysomnography test. The results achieved with the available data suggest the use of machine learning methods as tools for providing patients with a priority level for polysomnography test, but they still cannot be used for automated diagnosis.
The role of colony stimulating factors (CSFs) in cystic fibrosis (CF) circulating neutrophils has not been thoroughly evaluated, considering that the neutrophil burden of lung inflammation in these subjects is very high. The aim of this study was to assess granulocyte-CSF (G-CSF) and granulocyte-macrophage-CSF (GM-CSF) levels in CF patients in various clinical conditions and how these cytokines impact on activation and priming of neutrophils. G-CSF and GM-CSF levels were measured in sputum and serum samples of stable CF patients (n = 21) and in CF patients with acute exacerbation before and after a course of antibiotic therapy (n = 19). CSFs were tested on non CF neutrophils to investigate their effects on reactive oxygen species (ROS) production, degranulation (CD66b, elastase, lactoferrin, MMP-9), and chemotaxis. At very low concentrations found in CF patients (0.005–0.1 ng/ml), both cytokines inhibited ROS production, while higher concentrations (1–5 ng/ml) exerted a stimulatory effect. While either CSF induced elastase and MMP-9 secretion, lactoferrin levels were increased only by G-CSF. Chemotaxis was inhibited by GM-CSF, but was increased by G-CSF. However, when present together at low concentrations, CSFs increased basal and fMLP-stimulated ROS production and chemotaxis. These results suggest the CSF levels that circulating neutrophils face before extravasating into the lungs of CF patients may enhance their function contributing to the airway damage.
16 January 2018 in The Clinical Respiratory Journal
Background: The accuracy of surgical guides is a relevant factor in both surgical safety and prosthetic implications. The impact of widespread fabrication technologies (milling and 3D printing) was investigated. Methods: Surgical guides manufactured by means of two specific milling and 3D-printing systems were digitized and compared in a 3D analysis with the digital file of the designed guides. The surface mean 3D distance (at the surface where the teeth and mucosa made contact) and the axial and linear deviations of the sleeves’ housings were measured by means of a metrological software program. Univariate and multivariate statistical analyses were used to investigate the effects of the fabrication technology, type of support, and arch type on the surgical guides’ accuracy. Results: The median deviations of the intaglio surface in contact with the mucosa were significantly (p < 0.001) lower for the milled surgical guides (0.05 mm) than for the 3D-printed guides (−0.07 mm), in comparison with the reference STL file. The generalized estimated equation models showed that the axial deviations of the sleeves’ housings (a median of 0.82 degrees for the milling, and 1.37 degrees for the 3D printing) were significantly affected by the fabrication technology (p = 0.011) (the milling exhibited better results), the type of support (p < 0.001), and the combined effect of the fabrication technology and the sleeve-to-crest angle (p = 0.003). The linear deviation (medians of 0.12 mm for the milling and 0.21 mm for the 3D printing) of their center points was significantly affected by the type of support (p = 0.001), with the milling performing slightly better than the 3D printing. Conclusions: The magnitude of the difference might account for a limited clinical significance.
1 January 2017 in Journal of Clinical and Experimental Dentistry
To evaluate if changes in lower incisor position following orthodontic treatment are correlated with development of gingival recessions. Pre- and post-treatment digital models and lateral cephalograms of 22 subjects were collected retrospectively. The clinical crown length, gingival scallop, and papilla height of the central lower incisor were measured along with the cephalometric incisor's inclination, the distance from the mandibular plane, and the distance between the Infradentale and Menton points. Statistical correlations between gingival and cephalometric variables were studied. In addition, two groups were defined based on the post-treatment incisor inclination value ('normal' or 'proclined') and compared. The incisor inclination was correlated with the change in gingival scallop and papilla height. Moreover, there was a statistically significant difference in clinical crown height and gingival scallop between the 'normal' group and the 'proclined' group. Changes in lower incisor position, especially an excessive proclination, after orthodontic treatment may play a role in the development of gingival recession. Key words:Orthodontic treatment, Incisor inclination, IMPA, Gingival recession, Alveolar bone.
About Me
Researcher and Professor of Human biomedic statistic
Researcher and Professor of Human biomedic statistic