Automatic segmentation of dynamic neuroreceptor single-photon emission tomography images using fuzzy clustering

The segmentation of medical images is one of the most important steps in th e analysis and quantification of imaging data. However, partial volume arte facts make accurate tissue boundary definition difficult, particularly for images with lower resolution commonly used in nuclear medicine. In single-p hoton emission tomography (SPET) neuroreceptor studies, areas of specific b inding are usually delineated by manually: drawing regions of interest (ROI s), a time-consuming and subjective process. This paper applies the techniq ue off fuzzy c-means clustering (FCM) to automatically seg ment dynamic neu roreceptor SPET images. Fuzzy clustering was tested using a realistic, comp uter-generated, dynamic SPET phantom derived from segmenting an MR image of an anthropomorphic brain phantom. Also, the utility of applying FCM to rea l clinical data was assessed by comparison against conventional ROI;analysi s of iodine-123 iodobenzamide (IBZM) binding to dopamine D-2/D-3, receptors in the brains of humans.:in addition, a further test of the methodology wa s assessed: by applying FCM segmentation to [I-123]IDAM images (5-iodo-2-[[ 2-2-[ (dimethyl amino)methyl] phenyl]thio] benzyl alcohol) of serotonin tra nsporters in non-human primates. In the simulated dynamic SPET phantom, ove r a wide range of counts and ratios of specific binding to background, FCM correlated very strongly with the true counts (correlation coefficient r(2) >0.99, P<0.0001). Similarly, FCM gave segmentation of the [I-123]IBZM data comparable with manual ROI analysis, with the binding ratios derived from b oth methods significantly correlated (r(2)=0.83, P<0.0001). Fuzzy clusterin g is a powerful tool for the automatic, unsupervised segmentation of dynami c neuroreceptor SPET images. Where other automated techniques fail complete ly, and manual ROI definition would be highly subjective, FCM is capable of segmenting noisy images in a robust and repeatable manner.