Evaluation of the Applicability of Artificial Intelligence for the Prediction of Obstructive Sleep Apnoea

Background Due to the large number of undiagnosed obstructive sleep apnoea (OSA) patients, our aim was investigate applicability artificial intelligence (AI) in preliminary screening, based on simple anthropometric, demographic and questionnaire parameters. Methods Based results polysomnography performed, 100 patients study were grouped as follows: non-OSA, mild OSA moderately severe–severe OSA. Anthropometric measurements Berlin Epworth questionnaires completed. Results prediction body mass index (BMI), gender age accurate 81% cases. With completion questionnaires, accuracy rose 83%. The alone yielded a correct 75%, while 62% all best for categorization by severity obtained combining BMI, parameters, together with responses (71%). Supplemented neck circumference, this result improves slightly (73%). Conclusion results, it can be concluded that effectively easily categorized using AI anthropometric well data.