Expert-level aspiration and penetration detection during flexible endoscopic evaluation of swallowing with artificial intelligence-assisted diagnosis

Abstract

Flexible endoscopic evaluation of swallowing (FEES) is considered the gold standard in diagnosing oropharyngeal dysphagia. Recent advances in deep learning have led to a resurgence of artificial intelligence-assisted computer-aided diagnosis (AI-assisted CAD) for a variety of applications. AI-assisted CAD would be a remarkable benefit in providing medical services to populations with inadequate access to dysphagia experts, especially in aging societies. This paper presents an AI-assisted CAD named FEES-CAD for aspiration and penetration detection on video recording during FEES. FEES-CAD segments the input FEES video and classifies penetration, aspiration, residue in the vallecula, and residue in the hypopharynx based on the segmented FEES video. We collected and annotated FEES videos from 199 patients to train the network and tested the performance of FEES-CAD using FEES videos from other 40 patients. These patients consecutively underwent FEES between December 2016 and August 2019 at Fukushima Medical University Hospital. FEES videos were deidentified, randomized, and rated by FEES-CAD and laryngologists with over 15 years of experience in performing FEES. FEES-CAD achieved an average Dice similarity coefficient of 98.6$$\%$$ % . FEES-CAD achieved expert-level accuracy performance on penetration (92.5$$\%$$ % ), aspiration (92.5$$\%$$ % ), residue in the vallecula (100$$\%$$ % ), and residue in the hypopharynx (87.5$$\%$$ % ) classification tasks. To the best of our knowledge, FEES-CAD is the first CNN-based system that achieves expert-level performance in detecting aspiration and penetration.

Findings of a videofluoroscopic swallowing study in patients with dysphagia

“…The pharyngeal area at rest can be used as another objective index for evaluating the coordination of the pharyngeal phase of swallowing. The pharyngeal cavity contraction rate reflects the degree of contraction during swallowing in the pharyngeal phase ( 35 , 36 ). In this stage, the hyoid bone on the larynx moves upward, the arytenoepiglottis and thyrohyoid muscles contract, and the base of the tongue inclines backward to ensure the epiglottis forms a proper cover; while the epiglottis valley on both sides is oriented close to the midline, the muscle group in the larynx contracts, the vocal cord and the ventricular band retracts, the glottis closes, and the pharyngeal constrictor retracts.…”

Machine learning in the evaluation of voice and swallowing in the head and neck cancer patient

“…ML algorithms may help clinicians read FEES and VFSS more accurately. ML has been used to automatically localize key anatomical landmarks needed for a routine swallowing assessment in real-time [60] as well as detect the presence of penetration or aspiration during VFSS [61,62] and FEES [63 ▪▪ ]. Other studies have applied DL for automated temporal (pharyngeal phase, pharyngeal delay time) [64,65,66,67 ▪▪ ] and spatial (hyoid bone movement, airway invasion) [68,69] analysis of VFSS.…”

Deep Learning Techniques and Imaging in Otorhinolaryngology—A State-of-the-Art Review

“…The automated segmentation of cancer tissue has been successfully attempted in the nasopharynx (DSC 0.78) [74], oropharynx (DSC 0.76) [75], and laryngeal lesion (DSC 0.814) [76]. Aside from cancer pathology, segmentation may be used to select the region of interest for automated functional laryngeal analysis, such us the identification of the glottis angle [77,78], glottal midline [79], vocal cord paralysis [80], postintubation granuloma [81], vocal cord dynamics [82,83], or in the endoscopic evaluation of aspiration and penetration risk in dysphagia (FESS-CAD, DSC 0.92.5) [84].…”