"PREDICTION OF OBSTRUCTION SLEEPAPNEA USING CRANIOFACIAL IMAGES Table of ContentsLiterature Review....................................................................................................................... 2Introduction ............................................................................................................................ 2Aim ......................................................................................................................................... 3Obstructive Sleep Apnea ........................................................................................................ 3Methods and the machine learning languages........................................................................ 6Conclusion ............................................................................................................................... 10References ................................................................................................................................ 111 | P a g e Literature Review IntroductionThe Obstructive Sleep Apnea (OSA) is considered as one of the most prevalent disorderslinked with the snoring, collapse of repetitive upper airways throughout the sleep, oxygendesaturation plus sleep fragmentation. It is related with incremented cardiovascular morbidityaccident risk of the motor vehicles and the overall mortality. The diagnosis process of OSA isburdensome and awkward as it requires specialist assessments plus overnight observation in asleep laboratory. The subsequent is expensive, labor intensive plus resources restricted. As aconsequence, the detection of OSA is the society is low plus the majority of the patients whosuffer from this disorders are as yet undetected and undiagnosed. Therefore, there is a vitalclinical requirement to explore a process to enhance detection plus diagnosis of the OSA inthe society (Kushida, 2007). Forecast algorithms have been formed in order to the risks stratification plus a screening ofthe topics for the OSA. These algorithms are on the basis of basically on data like patientdemographics, symptoms as well as rates of obesity or measurement of the obesity inpatients. Despite the fact that the obesity is usually deemed the key risk factor for the OSA,the craniofacial morphology is rapidly recognized as a vital interacting factor in OSApathogenesis. However, the Craniofacial risk factors are added in the minority cases of theOSA pathogenesis expectation algorithms. This connects to the impractical characteristicnature of the presently available craniofacial assessments techniques. Moreover, thesuboptimal level of accuracy of these types of clinical algorithms plus the complexity ofsome assessments techniques constraint their routine application in the clinical diagnosisprocess of OSA (Pascualy, 2010). 2 | P a g e "