@article{VollmerVollmerLangetal.2022,
  author    = {Vollmer, Andreas and Vollmer, Michael and Lang, Gernot and Straub, Anton and Shavlokhova, Veronika and K{\"u}bler, Alexander and Gubik, Sebastian and Brands, Roman and Hartmann, Stefan and Saravi, Babak},
  title     = {Associations between periodontitis and COPD: An artificial intelligence-based analysis of NHANES III},
  series = {Journal of Clinical Medicine},
  volume    = {11},
  journal   = {Journal of Clinical Medicine},
  number    = {23},
  issn      = {2077-0383},
  doi       = {10.3390/jcm11237210},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-312713},
  year      = {2022},
  abstract  = {A number of cross-sectional epidemiological studies suggest that poor oral health is associated with respiratory diseases. However, the number of cases within the studies was limited, and the studies had different measurement conditions. By analyzing data from the National Health and Nutrition Examination Survey III (NHANES III), this study aimed to investigate possible associations between chronic obstructive pulmonary disease (COPD) and periodontitis in the general population. COPD was diagnosed in cases where FEV (1)/FVC ratio was below 70\% (non-COPD versus COPD; binary classification task). We used unsupervised learning utilizing k-means clustering to identify clusters in the data. COPD classes were predicted with logistic regression, a random forest classifier, a stochastic gradient descent (SGD) classifier, k-nearest neighbors, a decision tree classifier, Gaussian naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), a multilayer perceptron artificial neural network (MLP), and a radial basis function neural network (RBNN) in Python. We calculated the accuracy of the prediction and the area under the curve (AUC). The most important predictors were determined using feature importance analysis. Results: Overall, 15,868 participants and 19 feature variables were included. Based on k-means clustering, the data were separated into two clusters that identified two risk characteristic groups of patients. The algorithms reached AUCs between 0.608 (DTC) and 0.953\% (CNN) for the classification of COPD classes. Feature importance analysis of deep learning algorithms indicated that age and mean attachment loss were the most important features in predicting COPD. Conclusions: Data analysis of a large population showed that machine learning and deep learning algorithms could predict COPD cases based on demographics and oral health feature variables. This study indicates that periodontitis might be an important predictor of COPD. Further prospective studies examining the association between periodontitis and COPD are warranted to validate the present results.},
  language  = {en}
}
@article{KarnatiSeimetzKleefeldtetal.2021,
  author    = {Karnati, Srikanth and Seimetz, Michael and Kleefeldt, Florian and Sonawane, Avinash and Madhusudhan, Thati and Bachhuka, Akash and Kosanovic, Djuro and Weissmann, Norbert and Kr{\"u}ger, Karsten and Erg{\"u}n, S{\"u}leyman},
  title     = {Chronic Obstructive Pulmonary Disease and the Cardiovascular System: Vascular Repair and Regeneration as a Therapeutic Target},
  series = {Frontiers in Cardiovascular Medicine},
  volume    = {8},
  journal   = {Frontiers in Cardiovascular Medicine},
  issn      = {2297-055X},
  doi       = {10.3389/fcvm.2021.649512},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-235631},
  year      = {2021},
  abstract  = {Chronic obstructive pulmonary disease (COPD) is a major cause of morbidity and mortality worldwide and encompasses chronic bronchitis and emphysema. It has been shown that vascular wall remodeling and pulmonary hypertension (PH) can occur not only in patients with COPD but also in smokers with normal lung function, suggesting a causal role for vascular alterations in the development of emphysema. Mechanistically, abnormalities in the vasculature, such as inflammation, endothelial dysfunction, imbalances in cellular apoptosis/proliferation, and increased oxidative/nitrosative stress promote development of PH, cor pulmonale, and most probably pulmonary emphysema. Hypoxemia in the pulmonary chamber modulates the activation of key transcription factors and signaling cascades, which propagates inflammation and infiltration of neutrophils, resulting in vascular remodeling. Endothelial progenitor cells have angiogenesis capabilities, resulting in transdifferentiation of the smooth muscle cells via aberrant activation of several cytokines, growth factors, and chemokines. The vascular endothelium influences the balance between vaso-constriction and -dilation in the heart. Targeting key players affecting the vasculature might help in the development of new treatment strategies for both PH and COPD. The present review aims to summarize current knowledge about vascular alterations and production of reactive oxygen species in COPD. The present review emphasizes on the importance of the vasculature for the usually parenchyma-focused view of the pathobiology of COPD.},
  language  = {en}
}
@article{AlmadeJongJelusicetal.2016,
  author    = {Alma, Harma and de Jong, Corina and Jelusic, Danijel and Wittmann, Michael and Schuler, Michael and Flokstra-de Blok, Bertine and Kocks, Janwillem and Schultz, Konrad and van der Molen, Thys},
  title     = {Health status instruments for patients with COPD in pulmonary rehabilitation: defining a minimal clinically important difference},
  series = {npj Primary Care Respiration Medicine},
  volume    = {26},
  journal   = {npj Primary Care Respiration Medicine},
  number    = {16041},
  doi       = {10.1038/npjpcrm.2016.41},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-166327},
  year      = {2016},
  abstract  = {The minimal clinically important difference (MCID) defines to what extent change on a health status instrument is clinically relevant, which aids scientists and physicians in measuring therapy effects. This is the first study that aimed to establish the MCID of the Clinical chronic obstructive pulmonary disease (COPD) Questionnaire (CCQ), the COPD Assessment Test (CAT) and the St George's Respiratory Questionnaire (SGRQ) in the same pulmonary rehabilitation population using multiple approaches. In total, 451 COPD patients participated in a 3-week Pulmonary Rehabilitation (PR) programme (58 years, 65\% male, 43 pack-years, GOLD stage II/III/IV 50/39/11\%). Techniques used to assess the MCID were anchor-based approaches, including patient-referencing, criterion-referencing and questionnaire-referencing, and the distribution-based methods standard error of measurement (SEM), 1.96SEM and half standard deviation (0.5s.d.). Patient- and criterion-referencing led to MCID estimates of 0.56 and 0.62 (CCQ); 3.12 and 2.96 (CAT); and 8.40 and 9.28 (SGRQ). Questionnaire-referencing suggested MCID ranges of 0.28-0.61 (CCQ), 1.46-3.08 (CAT) and 6.86-9.47 (SGRQ). The SEM, 1.96SEM and 0.5s.d. were 0.29, 0.56 and 0.46 (CCQ); 3.28, 6.43 and 2.80 (CAT); 5.20, 10.19 and 6.06 (SGRQ). Pooled estimates were 0.52 (CCQ), 3.29 (CAT) and 7.91 (SGRQ) for improvement. MCID estimates differed depending on the method used. Pooled estimates suggest clinically relevant improvements needing to exceed 0.40 on the CCQ, 3.00 on the CAT and 7.00 on the SGRQ for moderate to very severe COPD patients. The MCIDs of the CAT and SGRQ in the literature might be too low, leading to overestimation of treatment effects for patients with COPD.},
  language  = {en}
}