[Study on Non-invasive Detection of Atherosclerosis Based on Electrocardiogram and Pulse Wave Signals]

Area of Science:

• Cardiovascular Physiology and Pathology
• Biomedical Signal Processing
• Artificial Intelligence in Medicine

Background:

• Artery stiffness is a primary contributor to cardiovascular diseases.
• Early detection of arteriosclerosis is crucial for preventing cardiovascular events.
• Non-invasive methods for assessing vascular stiffness are highly significant.

Purpose of the Study:

• To develop a non-invasive method for assessing artery stiffness and arteriosclerosis.
• To utilize electrocardiogram (ECG) and pulse wave signals for early detection of vascular changes.
• To build an assessment model for arteriosclerosis risk prediction.

Main Methods:

• Characteristic parameters from ECG (RR interval, QRS width, T amplitude) and pulse wave signals were analyzed.
• A dataset of forty ECG and pulse wave signal samples was collected.
• An Adaptive Network-based Fuzzy Interference System (ANFIS) model was constructed for arteriosclerosis assessment.

Main Results:

• The ANFIS model successfully assessed arteriosclerosis non-invasively.
• The developed method demonstrated self-learning diagnosis capabilities based on expert experience.
• The technique showed potential for evaluating cardiovascular disease risk.

Conclusions:

• The non-invasive detection method using ECG and pulse wave signals is effective for arteriosclerosis assessment.
• This technique can be further developed into a valuable tool for cardiovascular disease risk evaluation.
• Effective and prompt medical intervention facilitated by this technique can reduce cardiovascular morbidity and mortality.