Artificial Intelligence in Videofluoroscopy Swallow Study Analysis: A Comprehensive Review

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Medicine
• Speech-Language Pathology

Background:

• Videofluoroscopic Swallowing Study (VFSS) is the gold standard for diagnosing dysphagia.
• VFSS interpretation faces challenges due to human bias and variability.
• Artificial intelligence (AI) offers potential solutions to enhance VFSS analysis.

Purpose of the Study:

• To review current AI applications in analyzing VFSS for swallowing disorders.
• To assess AI's role in supporting clinical decision-making for dysphagia.
• To identify progress and limitations in AI-driven VFSS analysis.

Main Methods:

• Comprehensive literature review of AI techniques applied to VFSS.
• Analysis of AI's performance in specific VFSS tasks like phase detection and abnormality identification.
• Evaluation of AI model generalizability and integration challenges.

Main Results:

• Significant AI advancements in pharyngeal phase detection, bolus/hyoid bone segmentation, and penetration-aspiration detection.
• AI shows potential for analyzing clinical relevance and expanding VFSS scope.
• An end-to-end automated AI tool for VFSS analysis is not yet available.

Conclusions:

• AI holds considerable potential to improve the objectivity and efficiency of VFSS analysis.
• Further research is needed on dataset availability, model generalizability, and clinical integration for speech-language pathologists.
• AI can enhance diagnostic accuracy and clinical decision support in swallowing disorder assessment.