PNNL

Abstract

We present a novel application of machine learning techniques to optimize the design of a radiation detection system. A decision tree-based algorithm is described which greedily optimizes detection energy partitioning based on a minimum detectable activity metric – appropriate for radiation measurement. Applying this method to optimizing sensitivity to radioxenon decays in the presence of radon-progeny backgrounds, we find that in general high resolution readout and high spatial segmentation yield little additional sensitivity compared to simpler detector designs.