Artificial Intelligence Based Blood Pressure Estimation From Auscultatory and Oscillometric Waveforms: A...

Area of Science:

• Biomedical Engineering
• Artificial Intelligence
• Cardiovascular Health

Background:

• Cardiovascular disease is the leading global cause of death, with elevated blood pressure (BP) as the primary risk factor.
• Automated non-invasive blood pressure (NIBP) devices are increasingly used for home monitoring, with cuff-based methods dominating.
• Machine learning and AI, particularly deep learning, offer novel approaches for NIBP estimation using large datasets.

Purpose of the Study:

• To review Artificial Intelligence (AI)-based blood pressure estimation methods.
• To focus on recent advances in deep learning approaches for NIBP estimation.
• To discuss the strengths, weaknesses, and future directions of AI in BP monitoring.

Main Methods:

• Literature review of AI-based NIBP estimation techniques.
• Analysis of various deep learning architectures and methodologies.
• Discussion of challenges and potential solutions for AI adoption in BP estimation.

Main Results:

• Deep learning techniques demonstrate significant potential for enhancing NIBP estimation accuracy.
• Various AI architectures offer different strengths in data classification and feature extraction for BP monitoring.
• Identified limitations exist that may impede the widespread clinical adoption of deep learning in this field.

Conclusions:

• Deep learning presents plausible benefits for the advancement of non-invasive blood pressure estimation.
• Addressing identified limitations is crucial for the successful integration of deep learning into routine cardiovascular health monitoring.
• Future frameworks are suggested to overcome challenges and promote the adoption of AI-driven NIBP solutions.