Spectral Analysis Using Regularized Non-Negative Least-Squares Estimation

Abstract

The implementation of spectral analysis techniques involves solving a highly underdetermined linear system equation and is prone to the effect of measurement noise. The authors propose to use a regularized non-negative least-square estimator to stabilize the implementation of the technique. They introduce a penalty term in their formulation of the function to discourage disparities in tracer kinetics between neighboring pixels and use an iterative method to impose positivity constraints. The authors show results from analysis of FDG thorax images of patients suspected to have cancers and summarize their findings.