Deep Learning Analysis to Automatically Detect the Presence of Penetration or Aspiration in Videofluoroscopic Swallowing Study

Background: Videofluoroscopic swallowing study (VFSS) is currently considered the gold standard to precisely diagnose and quantitatively investigate dysphagia. However, VFSS interpretation is complex and requires consideration of several factors. Purpose: Therefore, considering the expected impact on dysphagia management, this study aimed to apply deep learning to detect the presence of penetration or aspiration in VFSS of patients with dysphagia automatically.Materials and Methods: The VFSS data of 190 participants with dysphagia were collected. A total of 10 frame images from one swallowing process were selected (five high-peak images and five low-peak images) for the application of deep learning in a VFSS video of a patient with dysphagia. We applied a convolutional neural network (CNN) for deep learning using the Python programming language. For the classification of VFSS findings (normal swallowing, penetration, and aspiration), the classification was determined in both high-peak and low-peak images. Thereafter, the two classifications determined through high-peak and low-peak images were integrated into a final classification.Results: The area under the curve (AUC) for the validation dataset of the VFSS image for the CNN model was 0.946 for normal findings, 0.885 for penetration, and 1.000 for aspiration. The average AUC was 0.962.Conclusion: This study demonstrated that deep learning algorithms, particularly the CNN, could be applied for detecting the presence of penetration and aspiration in VFSS of patients with dysphagia.

Impact of nasogastric tube feeding on swallowing function in patients with dysphagia: a pilot study

During dysphagia treatment, direct oral swallowing therapy is applied to some patients temporarily fed via nasogastric tube. However, the risk of aspiration in oral swallowing while nasogastric tube in situ may be disregarded in a standard videofluoroscopic swallowing study performed without a nasogastric tube. To evaluate the diagnostic significance of nasogastric tube in situ videofluoroscopic swallowing study of nectar and pureed diet compared to the standard videofluoroscopic swallowing study without nasogastric tube. Videofluoroscopic swallowing study records of dysphagia patients conducted between June and August 2017 in a university hospital were collected for review. Rosenbek’s penetration-aspiration scale, diagnostic criteria of aspiration were used to define aspiration. videofluoroscopic dysphagia scale for videofluoroscopic swallowing study with or without nasogastric tube were compared for nectar and pureed diet swallowing. Patients had various duration of nasogastric tube feeding. Paired T-test comparing the videofluoroscopic dysphagia scales for videofluoroscopic swallowing study with or without nasogastric tube revealed significant aggravation of swallowing dysfunction in nectar drinking while nasogastric tube in situ. This aggravation was noted in 19% (n=4) of patients who suffered from stroke regardless of nasogastric tube duration. Nasogastric tube in situ videofluoroscopic swallowing study, at least of nectar drinking could be beneficial in selecting a safe candidate for direct oral swallowing therapy in conjunction with the conventional nasogastric tube removed videofluoroscopic swallowing study.

Videofluoroscopic Swallowing Study Findings Associated With Subsequent Pneumonia in Patients With Dysphagia Due to Frailty

Dysphagia in frailty or deconditioning without specific diagnosis that may cause dysphagia such as stroke, traumatic brain injury, or laryngeal pathology, has been reported in previous studies; however, little is known about which findings of the videofluoroscopic swallowing study (VFSS) are associated with subsequent pneumonia and how many patients actually develop subsequent pneumonia in this population. In this study, we followed 190 patients with dysphagia due to frailty or deconditioning without specific diagnosis that may cause dysphagia for 3 months after VFSS and analyzed VFSS findings for the risk of developing pneumonia. During the study period, the incidence of subsequent pneumonia was 24.74%; regarding the VFSS findings, (1) airway penetration (PAS 3) and aspiration (PAS 7 and 8) were associated with increased risk of developing pneumonia, and (2) the functional dysphagia scale (FDS) scores of the patients who developed subsequent pneumonia were higher than those of the patients who did not develop subsequent pneumonia. Our study findings might assist clinicians in making clinical decisions based on the VFSS findings in this population.

Safety and Efficacy of Specially Designed Texture-Modified Foods for Patients with Dysphagia Due to Brain Disorders: A Prospective Study

Providing texture-modified food for patients with dysphagia is a cornerstone of dysphagia treatment. This study aimed to evaluate the safety and efficacy of a specially designed texture-modified food that can be easily swallowed while maintaining the unique taste by adjusting hardness and adhesiveness in patients with brain disorders using a videofluoroscopic swallowing study. We included 101 patients with oropharyngeal dysphagia due to brain disorders who were referred to the rehabilitation department. To evaluate the safety and efficacy of a specially designed texture-modified food, rice gruel was compared with a regular instant rice porridge, and bulgogi mousse was compared with ground bulgogi, which normally serves as a texture-modified diet for patients with dysphagia in our hospital during the videofluoroscopic swallowing study. The Penetration–Aspiration Scale score, oropharyngeal transit time, number of swallows required to maximally eliminate food materials from the oropharyngeal space, and vallecular and pyriform sinus residue after swallowing scale score were compared. Rice gruel required a shorter oropharyngeal transit time and fewer number of swallowing per the given amount of food than regular instant rice porridge; however, no statistical difference was found in the vallecular and pyriform sinus residue after swallowing scale scores and the Penetration–Aspiration Scale scores. Bulgogi mousse required more swallowing and had lower Penetration–Aspiration Scale scores than ground bulgogi; however, no significant difference was found in the oropharyngeal transit time and the vallecular and pyriform sinus residue after swallowing scale scores. The study foods were safe and efficacious compared to control foods usually provided for patients with dysphagia from various brain disorders.

Usefulness of the Modified Videofluoroscopic Dysphagia Scale in Choosing the Feeding Method for Stroke Patients with Dysphagia

Introduction: The Videofluoroscopic Dysphagia Scale (VDS) is used to predict the long-term prognosis of dysphagia in patients with strokes. However, the inter-rater reliability of the VDS was low in a previous study. To overcome the mentioned limitations of the VDS, the modified version of the VDS (mVDS) was created and clinically applied to evaluate its usefulness in choosing the feeding method for stroke patients with dysphagia. Methods: The videofluoroscopic swallowing study (VFSS) data of 56 stroke patients with dysphagia were collected retrospectively. We investigated the presence of aspiration pneumonia and the selected feeding method. We also evaluated the correlations between the mVDS and the selected feeding method, and between the mVDS and the presence of aspiration pneumonia after stroke. Univariate logistic regression and receiver operating characteristic analyses were used in the data analysis. Results: The inter-rater reliability (Cronbach α value) of the total score of the mVDS was 0.886, which was consistent with very good inter-rater reliability. In all patients with dysphagia, the supratentorial stroke subgroup, and the infratentorial stroke subgroup, the mVDS scores were statistically correlated with the feeding method selected (p < 0.05) and the presence of aspiration pneumonia (p < 0.05). Conclusions: The mVDS can be a useful scale for quantifying the severity of dysphagia, and it can be a useful tool in the clinical setting and in studies for interpreting the VFSS findings in stroke patients with dysphagia. Further studies with a greater number of patients and various stroke etiologies are required for more generalized applications of the mVDS.