Related Experiment Video
Updated: Aug 7, 2026
07:45
Minimally Invasive Murine Laryngoscopy for Close-Up Imaging of Laryngeal Motion During Breathing and Swallowing
Published on: December 1, 2023
Large Language Model-Driven Throat-Wearable Sensing System for Real-Time Recognition and Evaluation of Swallowing
Zitang Yuan1, Yihan Lin2, Xiyao Zhao1
1School of Instrumentation and Optoelectronic Engineering, Beihang University, Beijing, China.
Advanced Healthcare Materials
|August 6, 2026
Summary
A new throat-wearable sensing system (TWSS) uses a large language model (LLM) to monitor laryngeal activity for assessing swallowing function after stroke. This wearable technology offers continuous, objective evaluation, improving upon traditional methods.
Area of Science:
- Biomedical Engineering
- Wearable Technology
- Artificial Intelligence in Healthcare
Background:
- Swallowing disorders (dysphagia) are frequent post-stroke complications.
- Current assessment methods lack continuous, objective evaluation.
- Clinical observation and subjective screening limit accurate diagnosis.
Purpose of the Study:
- To develop a large language model (LLM)-driven throat-wearable sensing system (TWSS) for real-time monitoring of laryngeal activity.
- To enable quantitative assessment of swallowing function.
- To provide a continuous and objective evaluation tool for dysphagia.
Main Methods:
- Development of TWSS comprising a flexible sensing patch and a structured signal sequence-based LLM framework (S3-LLM).
- Integration of a stretchable sensor sensitive to pressure and strain for capturing laryngeal movements.
- Utilizing S3-LLM with temporal encoding and parameter-efficient fine-tuning for few-shot learning.
Main Results:
- TWSS achieved 92.4% accuracy in recognizing normal laryngeal activities.
- TWSS demonstrated 87.6% accuracy in evaluating swallowing function.
- Performance showed an approximate 20% improvement over conventional assessment models.
Conclusions:
- TWSS offers a promising wearable platform for continuous, objective, and quantitative assessment of swallowing disorders.
- The system demonstrates the potential of LLM-driven wearables in personalized healthcare.
- This technology can aid in the management of post-stroke dysphagia.
