ABSTRACT
To objectively and quantitatively study the effect of JPEG2000 lossy compression of nuclear medicine image, line source images of various noise levels, actual quadrant-bar phantom images and simulated quadrant-bar phantom images of various noise levels were compressed using a JPEG2000 plug-in at 10 : 1 to 50 : 1 with a 10 : 1 step. Modulation transfer function (MTF), full width at half-maximum (FWHM) and the peak to total count ratio (PTCR) of line spread function of the simulated line source images were analyzed quantitatively. Image quality of the quadrant-bar phantom images was visually evaluated according to quality control procedures and the result was used to verify the quantitative analysis. It was observed that PTCR decreased as compression ratio increased and the higher the noise level, the faster PTCR decreased, which was in accordance with the increase of the width of the smallest bars that can be visually resolved in the actual quadrant-bar phantom images and simulated quadrant-bar phantom images. On the other hand, no significant change of MTF and FWHM was observed as compression ratio increased. It was concluded that PTCR may serve as an objectively quantitative metrics of image quality assessment.
Subject(s)
Humans , Algorithms , Data Compression , Methods , Reference Standards , Diagnostic Imaging , Reference Standards , Image Processing, Computer-Assisted , Nuclear Medicine , MethodsABSTRACT
To study the usefulness of JPEG2000 compression for nuclear medicine (NM) image, normal and abnormal static images were compressed using a JPEG2000 plug-in. For lossless algorithm, the compressing ratio (CR) was calculated. For lossy algorithm, images were visually analyzed by NM physicians and ROC curves were generated. The area under ROC curve (AUC) was used as the index of image diagnostic quality. Paired sample t tests were performed to compare the AUCs. It was found that the lossless CR was (1.34+/-0.05) : 1. For lossy algorithm, the AUC decreased with the increase of CR. Comparison between the original and the compressed images revealed no significant difference for 10:1 CR but significant difference for bigger CRs. It was concluded that lossless compression has little usefulness for NM image because of very low CR. While lossy compression isused, the diagnostic quality of static NM images is preserved at CRs 50 : 1,40 : 1, 30 : 1,20 : 1 up to 10 : 1. For other types of NM image, CR should be increased or decreased according to their characteristics, especially the level of intrinsic statistical noise of NM image.