Application of computer analysis of mammography phantom images (CAMPI) methodology to the comparison of two digital biopsy machines

摘要

Abstract: The objective of this research was to compare a FischerMammoVision/MammoTest and a LoRad DSM digital biopsymachine using the Computer Analysis of MammographyPhantom Images (CAMPI) methodology. This study reportson analysis of the 4 largest microcalcification groups(M1, M2, M3 and M4) and the largest nodule (N1) in amammography accreditation phantom on images acquired at26 kVp and different mAs values on the two machines.Both machines were linear in response but the MammoTestwas more sensitive (i.e., it yielded a larger gray-scale value for a given x-ray technique). However, evenafter correcting for this difference, the CAMPI noisemeasure was substantially smaller for the LoRad thanthe MammoTest over the range of mAS values studied.Similarly, the CAMPI signal- to-noise-ratio andcorrelation measures were higher for the LoRad than theMammoTest over the same range of mAs, especially forthe larger objects (M1/M2 and N1). For the smallerspecks in M3/M4 somewhat closer performance wasobserved. The overall differences are attributed tobetter contrastoise performance of the LoRad whichappear to outweigh its lesser resolution capability.Our results are in agreement with earlier physical andpsychophysical measurements using differentmethodologies. This work also describes betterpredictive models (i.e., fits) to describe thevariation of all CAMPI measures with mAs at constantkVp. For example, the noise measure was fitted to afunction that included physically reasonable sources ofnoise e.g., dark noise and detector gain fluctuations,in addition to the usual quantum noise. These fits canbe used to summarize machine performance and to predictdependencies on other variables (e.g., exposure ordose) that are related to the mAs. !13