Cellular automaton simulation of tumour growth -- equivocal relationships between simulation parameters and...

Area of Science:

• Computational biology
• Cancer modeling
• Image analysis

Background:

• Tumor growth simulations require accurate parameter estimation.
• Morphometric features of simulated patterns offer potential for parameter inference.

Purpose of the Study:

• To develop a method for estimating simulation parameters from morphometric features of simulated tumor patterns.
• To assess the reliability of morphometric features in inferring parameters such as tumor cell motility, adhesion, growth control, and stroma destruction.

Main Methods:

• A cellular automaton model was used to simulate tumor growth patterns.
• 27,800 simulations were performed with varying parameters.
• Morphometric features of resulting patterns were extracted and correlated with input simulation parameters.

Main Results:

• Highly significant correlations were found between morphometric features and individual simulation parameters (correlation coefficients 0.72-0.99).
• Estimates were reliable when only one parameter varied.
• Estimates became less effective and often equivocal when multiple parameters varied simultaneously.

Conclusions:

• While simulation parameters demonstrably influence morphologic patterns, inferring these parameters solely from morphometric features yields ambiguous results.
• The developed interpretation procedure can rule out some parameter combinations but often results in multiple plausible parameter sets.