An evaluation of SPECT imaging for quantitative assessment of parkinson's disease

SPECT brain imaging of dopamine transporter with ^99mTc TRODAT-1 is useful for diagnosis and evaluation of Parkinson's disease, and improvements to its quantification are expected to be able to increase its diagnostic power. In previous studies, we have employed fan-beam collimation to achieve enhanced spatial resolution and counting statistics. Further improvements can be made by use of reconstruction methods that can accurately account for the physics of SPECT imaging. Many reconstruction algorithms are available for this purpose. These algorithms have their respective strengths and weaknesses, and it remains unclear which method is more suitable for clinical use. To a large extent, this situation is due to the unavailability of gold standards when working with real data, thereby making it extremely difficult to obtain an objective performance comparison of reconstruction algorithms. Recently, Hoppin et al. has proposed a promising technique that may mitigate this difficulty. In this work, we examine the impact of various reconstruction algorithms on quantification of SPECT ^99mTc TRODAT-1 brain imaging, and investigate whether Hoppin's method can achieve an objective performance comparison for these algorithms. Results obtained in our studies are quite encouraging.