Bio inspired heuristic computing scheme for the human liver nonlinear model

Area of Science:

• Computational biology
• Bio-inspired computing
• Biomedical modeling

Background:

• The human liver exhibits complex nonlinear behavior crucial for physiological understanding.
• Existing models may lack the precision to capture these intricate dynamics.

Purpose of the Study:

• To develop a bio-inspired heuristic computing approach for solving the nonlinear human liver model.
• To validate the accuracy and reliability of the proposed computational method.

Main Methods:

• Utilizing stochastic computation based on artificial neural networks (ANN).
• Employing a hybrid optimization scheme combining genetic algorithm (GA) and interior-point (IP) methods (GAIP).
• Designing a fitness function from the differential form of the liver model and optimizing it with GAIP.

Main Results:

• The proposed GAIP solver demonstrated high accuracy, evidenced by close agreement with the reference Adams scheme solutions.
• Low absolute error values confirm the efficacy and worth of the developed computational solver.
• Statistical analyses further validated the reliability of the approach for nonlinear liver modeling.

Conclusions:

• The bio-inspired GAIP approach provides an accurate and reliable method for solving nonlinear human liver models.
• This computational strategy, utilizing ANNs and hybrid optimization, holds promise for advancing biomedical simulations.