Background: The synergistic association between oral cavity disorders and airway disorders in smokers has long been recognized. Periodontal disease and airway obstruction are 20 times more likely in smokers. Smoking causes increased inflammatory cytokines in the oral mucosa; generally, airway obstruction has been associated with increased inflammatory markers in the airway mucosa. This study developed a prototype to visualize smokers' oral mucosa to identify potential airway obstruction disease.Methods: This study collected many types of oral mucosal lesions that are typically found in smokers, such as leukoplakia, nicotinic stomatitis, black hairy tongue, oral cancer, and smoker melanosis, from various literature and images of the mucosa of patients with a history of smoking who were treated at the hospital. The data is divided into a training, validation, and testing set and then using the PyTorch framework and the UltraLytics library.Results: This study created a prototype of an endoscope that can detect lesions on the oral mucosa-related airway obstruction disease. Sixty-three percent of the respondents who underwent prototype testing were between the ages of twenty-one and thirty. Of those who smoked, 86% had done so for five to ten years. Sixty percent of the respondents had no COPD diagnosis. The sensitivity of the prototype demonstrated a high rate of 84%. However, the specificity exhibited 57.14%.Conclusion: Endoscopic detection of the oral mucosa can be used for early screening of suspected obstructive airway disorders in smokers. This tool could enhance screening for smoking's effects on the mouth and prevent early obstructive airway diseases.
Read full abstract