A Phlegm Stagnation Monitoring Based on VDS Algorithm

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Computer Science

Background:

• Phlegm stagnation poses a significant care challenge, particularly in nonmedical settings.
• Effective real-time monitoring is crucial for timely intervention and patient well-being.

Purpose of the Study:

• To propose a practical, nonmedical sputum monitoring system for phlegm stagnation care.
• To develop and evaluate a novel algorithm for real-time detection and classification of phlegm-related throat movements.

Main Methods:

• Real-time video acquisition of the laryngeal area.
• A three-frame differential method for detecting throat movement targets.
• A Video-based Detection and Speed (VDS) algorithm combined with Dynamic Time Warping (DTW) and Support Vector Machine (SVM) for sequence image classification.

Main Results:

• The proposed VDS algorithm demonstrated effective identification of throat movement anomalies.
• The system achieved a higher recognition rate compared to existing methods like DTW, SVM, CTM, and HMM.
• The method proved practical for implementation in nonmedical monitoring systems.

Conclusions:

• The developed VDS algorithm offers a practical and effective solution for nonmedical sputum monitoring.
• Real-time analysis of laryngeal movements can aid nursing personnel in managing sputum blocking issues.
• This approach facilitates timely identification of phlegm stagnation symptoms for improved patient care.