A multiphysics-based artificial neural networks model for atherosclerosis.

Area of Science:

• Biomedical Engineering
• Computational Biology
• Medical Physics

Background:

• Atherosclerosis, characterized by arterial wall hardening, poses significant health risks.
• Predictive mathematical multi-physics models exist but are computationally intensive.
• Improving the efficiency of these models is crucial for broader application.

Purpose of the Study:

• To enhance the computational efficiency of atherosclerosis development prediction models.
• To leverage machine learning, specifically artificial neural networks (ANN), for this purpose.
• To develop a faster and accurate predictive tool for atherosclerosis.

Main Methods:

• Creation of a comprehensive database using multi-physics Finite Element Method (FEM) simulations.
• Training and validation of an artificial neural network (ANN) model using the FEM simulation data.
• Comparison of ANN model performance against traditional FEM simulations.

Main Results:

• The developed ANN model demonstrates high accuracy in predicting atherosclerosis progression.
• A substantial computational efficiency gain was achieved using the ANN model.
• The ANN model provides rapid predictions, overcoming the limitations of traditional FEM simulations.

Conclusions:

• Artificial neural networks offer a computationally efficient and accurate method for predicting atherosclerosis.
• This machine learning approach can accelerate research and clinical applications in cardiovascular disease.
• The study highlights the potential of AI in enhancing complex biological system modeling.