DaTscan quantitative analysis

Objectives To develop and validate a semi-automated quantitative analysis program for DaTscan images to be used as an adjunct to visual assessment for differentiating between Parkinson's disease (PD) and essential tremor (ET).

Methods A semi-automated quantitative analysis program was developed that places manually drawn regions of interest (ROI) over the left and right putamen, left and right caudate, the entire striatum (putamen and caudate), as well as background on a composite transverse slice of attenuation-corrected SPECT images. The program calculates the binding uptake index for the striatum (SBI), putamen (PBI), and caudate (CBI), as well as putamen to caudate ratio (P/C) and striatum asymmetry index (SAI). The program was applied to 13 DaTscan patient SPECT/CT studies that were also independently assessed visually by two trained radiologists who ranked the images based on the level of uptake: 0-normal, 1-reduced uptake in one putamen, 2-reduced uptake bilaterally in the putamina, and 3-reduced uptake in the caudate. The intra- and inter-operator variability were also examined.

Results There was good linear correlation (overall r=0.7) between the concordant classification of the distribution and uptake by visual assessment and that calculated by the program indices SBI, PBI, and CBI as well as P/C ratio. SAI was useful in determining uni- vs. bi-lateral reduction in uptake. Mean values and standard deviations were established for the various stages of uptake, though the patient sample was relatively small. Inter-operator variation was 2% and intra-operator 5% respectively, both of which were not found to be statistically significant based on student paired t-test (P>0.05).

Conclusions A semi-automated quantitative analysis program for DaTscan images was developed that provides calculated indices reflecting the relative uptake in the striatum. This program can be a useful adjunct tool to visual assessment in classifying the levels of relative uptake in DaTscan images and in differentiating between PD and ET.