Artificial Intelligence-Enabled Novel Atrial Fibrillation Diagnosis System Using 3D Pulse Perception Flexible...

Area of Science:

• Cardiovascular Medicine
• Biomedical Engineering
• Artificial Intelligence

Background:

• Atrial fibrillation (AF) is a common cardiovascular disease with significant disability and mortality rates.
• Timely and effective AF recognition is crucial for diagnosis and prevention.
• Current diagnostic methods may lack accessibility or real-time capabilities.

Purpose of the Study:

• To develop a novel intelligent sensing and recognition system for AF.
• To integrate Traditional Chinese Medicine (TCM) theory with flexible wearable electronics and artificial intelligence (AI).
• To establish a new strategy for precise diagnosis and remote treatment of AF.

Main Methods:

• Designed flexible pressure sensor arrays based on TCM theory to capture 3D pulse data (Cun, Guan, Chi).
• Developed a signal acquisition system to record pulse signals, labeled by cardiovascular disease doctors.
• Utilized a convolutional neural network (CNN) and a pulse database to create an AF recognition model.

Main Results:

• The system successfully captured synchronous 3D pulse data, differentiating between individuals with and without AF.
• The AI-powered recognition model achieved up to a 90% recognition rate in laboratory settings.
• Clinical validation in hospitals demonstrated an 80% recognition rate for detecting AF patients.

Conclusions:

• The proposed system offers a novel strategy for the precise diagnosis and remote treatment of AF.
• This integration of TCM, wearable electronics, and AI accelerates the development of Modern Chinese Medicine treatments.
• The intelligent sensing system shows promise for improving AF management and patient outcomes.