Toward a grey box approach for cardiovascular physiome.

Area of Science:

• Cardiovascular Medicine
• Computational Biology
• Biomedical Engineering

Background:

• Physiomic approaches are integral to diagnosing cardiovascular diseases.
• Traditional methods rely on mathematical models, while machine learning (ML) offers alternative computational speed and less reliance on known mechanisms.
• ML is rapidly advancing in cardiovascular medicine, presenting new diagnostic and simulation possibilities.

Purpose of the Study:

• To review the complementary roles of traditional mathematical modeling and machine learning (ML) in cardiovascular physiomics.
• To highlight the potential benefits of integrating ML with mathematical models for improved cardiovascular disease diagnosis and simulation.
• To introduce examples of combined approaches in cardiovascular physiome research.

Main Methods:

• Review of existing literature on mathematical modeling and ML in cardiovascular physiomics.
• Analysis of the strengths and limitations of both traditional mathematical models and ML algorithms.
• Identification of scenarios where combining mathematical and ML approaches is advantageous.

Main Results:

• Machine learning offers computational efficiency and can function without complete knowledge of system mechanisms.
• Traditional mathematical models provide accuracy based on established scientific laws.
• Integrating ML with mathematical models can enhance both accuracy and efficiency in cardiovascular simulations.

Conclusions:

• Combining mathematical models and ML in cardiovascular physiomics offers a synergistic approach.
• This integration promises to improve the accuracy and efficiency of cardiovascular disease diagnosis and simulation.
• Further research into combined methodologies is warranted for advancing cardiovascular medicine.