Additional Departmental Affiliation
Latin hypercube sampling and Partial Rank Correlation Coefficient procedure (LHS/PRCC) can be used in combination to perform a sensitivity analysis that assesses a model over a global parameter space. Through this analysis, the uncertainty of the parameters and therefore the variability of the model output in response to this uncertainty can be observed. Latin hypercube sampling divides the parameter space into equiprobable regions and sample without replacement, producing a global, unbiased selection of parameter values. For montonic, non-linear relationships, the correlation between the outputs and parameters can be understood by performing a Partial Rank Correlation Coefficient procedure. This sensitivity analysis is applied to ordinary differential equation models describing the interactions of various biological components in the healing of a diabetic foot ulcer. The results of the LHS/PRCC sensitivity analysis are used to assess the biological significance of the parameters in relation to each compartment of the model to further understand its biological implications.
Advisor(s) or Committee Chair
Dr. Richard Schugart
Applied Mathematics | Biology | Physical Sciences and Mathematics
Pennington, Hannah M., "Applications of Latin Hypercube Sampling Scheme and Partial Rank Correlation Coefficient Analysis to Mathematical Models on Wound Healing" (2015). Mahurin Honors College Capstone Experience/Thesis Projects. Paper 581.