Ex vivo investigation of a smart endotracheal tube for identifying esophageal intubation.

SIGNIFICANCE: Unrecognized intubation of the esophagus instead of the trachea results in rapid and severe consequences for the patient. Utilizing the spectral properties of the tissues could reduce incidents of these events.

AIM: We aim to investigate the design and implementation of a smart endotracheal tube (ETT) with integrated optical fiber sensors to distinguish esophageal and tracheal tissues.

APPROACH: Computational methods are investigated to characterize and classify nine pairs of ex vivo porcine organs using spectral properties. Two classifiers [ K -nearest neighbor and linear discriminant analysis (LDA)] are investigated.

RESULTS

Of the tissues sampled, 100% are correctly distinguished, with LDA being the preferred choice when considering both performance and applicability.

CONCLUSIONS

In clinical practice, this approach offers a method for confirming correct tracheal intubation using the spectral properties of the tissues, performed in a single step with no other invasive medical device than the ETT required to detect the spectral measurements.