Cancer immune control dynamics: a clinical data driven model of systemic immunity in patients with metastatic melanoma

Table 2 Comparison of current math models versus CICD model

	Current modeling approaches	CICD modeling
Modeling equations	Ordinary differential equations	Ordinary differential equation
	Delayed differential equations
	Partial differential equations
	Stochastic differential equations
	Agent based models
	Cellular automata
	Multiscale/hybrid
Equations	Multiple combinations	One large matrix equation
Parameters	Multiple, limited by availability	Unlimited
	Measured experimentally and/or estimated theoretically	All are computed
	Measured experimentally and/or estimated theoretically	All possible time varying parameters are contained within the unknown value in the Kolmogorov–Gabor Polynomial
Biomarkers	Limited, most less than 10	Expandable, currently 50
Data Measurement	Measured experimentally and/or estimated theoretically	Sequential daily blood draw
Number of Biomarker Relationships	Limited by current knowledge	Currently 28,605 including all biologically possible relationships
Biomarker Relationships Used	Only known relationships are modeled	All biologically possible relationships can be included
Results	Deterministic or Probabilistic	Deterministic
Results	Generalized and/or Individualize relative to parameter and patient data availability	Individualize to patient data measurements

A side-by-side comparison highlights the advantages of the flexible CICD approach. These advantages include one simple expandable equation, no estimated parametrization, large numbers of biomarkers and multi-dimensional relationships can be included, and an individualized model of their own immune system generated directly from their own data. As compared to the generalized, parameter dependent modeling found in the current literature

ISSN: 1471-2105