Induced bioresistance via BNP detection for machine learning-based risk assessment

Area of Science:

• Biomedical Engineering
• Artificial Intelligence in Healthcare
• Cardiovascular Diagnostics

Background:

• Cardiovascular disease (CVD) is a leading global cause of mortality.
• Early diagnosis and real-time monitoring are crucial for effective CVD management.
• Current diagnostic methods can involve significant analysis delays.

Purpose of the Study:

• To develop a rapid diagnostic tool for cardiovascular disease (CVD) using machine learning (ML).
• To integrate an ML algorithm with a flexible B-type natriuretic peptide (BNP) sensor for quick data analysis.
• To improve the speed and efficiency of CVD diagnosis and patient monitoring.

Main Methods:

• Fabrication of a flexible BNP sensor as an ion-selective field-effect transistor (ISFET).
• Characterization of the sensor using linear sweep voltammetry with artificial samples.
• Development of a machine learning algorithm (Quadratic Discriminant Analysis - QDA) in MATLAB.
• Testing the algorithm with human blood serum samples from 30 patients.

Main Results:

• The developed flexible BNP sensor enabled rapid electrical data acquisition.
• The ML algorithm, trained on sensor data, demonstrated high diagnostic power.
• Achieved 95% accuracy in classifying patient samples, indicating effective sorting power.
• Ultra-fast data collection and determination were achieved.

Conclusions:

• The combination of a flexible BNP sensor and ML algorithm offers a promising approach for rapid CVD diagnosis.
• This technology can significantly reduce diagnostic delays, facilitating timely intervention.
• The developed tool has the potential to enhance real-time patient monitoring strategies for cardiovascular health.