Sonographic Phenotyping of the Upper Airway in OSA With Backscattered Imaging Analysis by Machine Learning
Otolaryngology - Head and Neck Surgery
; 167(1 Supplement):P159, 2022.
Article
in English
| EMBASE | ID: covidwho-2064479
ABSTRACT
Introduction:
Anatomic assessment of the upper airway remains important in directing and monitoring of care for patients with obstructive sleep apnea (OSA). Nasopharyngoscopy is routine in clinical practice, but it can be invasive and potentially less attractive in the post-COVID-19 care setting. It also only allows subjective assessment. Ultrasound imaging of the upper airway with backscattered imaging analyzed via machine learning algorithm is investigated as a potential alternative. Method(s) Sixty-three subjects (14 female) with a mean age of 39.4 (12.6) years, body mass index (BMI) of 26.4 (4.6) kg/m2, and apnea-hypopnea index (AHI) of 19.0 (16.1) were consented from Stanford sleep surgery (July 2020 to May 2021). A standardized ultrasound protocol was used to image the soft palate, oropharynx, tongue base, and epiglottis. Via ultrasound device cleared by US Food and Drug Administation, backscattered ultrasound imaging (BUI) of the upper airway was performed and analyzed with machinelearning algorithms. Combined with B-mode measurements of airway muscular cross-sections, a logistic regression model was built to correlate with OSA severity. Result(s) The BUI of subjects with mild OSA was different from moderate-severe (AHI>=15) OSA at the soft palate (P=.0007). The axial-to-lateral ratio of upper airway length was reduced in the lower soft palate of the moderate-severe group (P=.0207). The logistic regression model with BUI, axial-to-lateral ratio at the soft palate, and BMI showed an area under the receiver-operating characteristic curve of 0.84 (95% CI, 0.726-0.920) in moderate-severe OSA. Conclusion(s) A noninvasive yet replicable technique to visualize and phenotype the upper airway is critical in the management of patients with sleep-disordered breathing. Sonographic BUI combined with B-mode airway measurements analyzed by machine learning show promise in characterizing the upper airway in patients with moderate-severe OSA.
adult; algorithm; apnea hypopnea index; body mass; conference abstract; controlled study; echography; epiglottis; female; human; human cell; machine learning; major clinical study; male; oropharynx; phenotype; receiver operating characteristic; SJSA-1 cell line; sleep disordered breathing; soft palate; tongue; upper respiratory tract
Full text:
Available
Collection:
Databases of international organizations
Database:
EMBASE
Language:
English
Journal:
Otolaryngology - Head and Neck Surgery
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS