Accomplishing successful endotracheal intubation (ETI) both rapidly and with a minimum number of attempts is a crucial skill for clinicians in order to manage patient airways and acute care. Repeated attempts may cause airway injury and increase the occurrence of hypoxemia and unstable hemodynamics. Usually, ETI is performed under visualization and recognition of the glottis. Initial success is highly dependent on clinician experience. Most current devices help confirm endotracheal tube position after intubation, but are not useful in reducing unnecessary attempts. This study develops an acoustically guided tracheal intubation system to help ETI providers identify the glottis during the procedure. By insufflating oxygen into the hypopharynx, the acoustic responses of non-glottic structures and the glottis are analyzed using the proposed two-stage segment-based algorithm to recognize the glottis and determine the glottic boundary. The first stage uses a delta Bayesian information criterion with a sliding window process to track potential boundaries with computational efficiency. In the second stage, a Gaussian mixture model classifier with a majority vote filter determines an acoustically evaluated boundary. Performance was assessed by comparing the acoustically evaluated boundary with the actual boundary obtained from synchronized fiberoptic images. Non-glottic and glottic sounds were recorded from 9 anesthetized adults receiving ETI. The success rate of glottic boundary recognition using the proposed method on this dataset is 77.78 %. For the successful cases, corresponding time differences between the acoustically evaluated boundary and the actual boundary were less than 400 ms. This robust method is feasible to help practitioners determine the glottis position rapidly and may reduce unnecessary attempts during ETI.
All Science Journal Classification (ASJC) codes
- Biomedical Engineering