Operator-less processing of myocardial perfusion SPECT studies

We have developed a completely automated algorithm to generate reoriented tomographic images from projections in myocardial perfusion SPECT. Methods: The algorithm consists of three software modules. The first module determines reconstruction limits for the projection dataset using two-dimensional feature extraction techniques. The second module reconstructs the projection images into transaxial images using standard filtered backprojection. The third module reorients the transaxial images into short-axis images. Results: The algorithm was validated on 350 rest ²⁰¹Tl and 350 stress ^99mTc- sestamibi studies acquired on a single-detector (178 studies), a 90° dual- detector (230 studies) or a triple-detector camera (292 studies). The complete processing sequence was successful in 93.6% of the studies (166/178 + 216/230 + 273/292). As for the individual modules, myocardial boundaries were correctly determined in 96.3% of the studies (171/178 + 222/230 + 281/292), while reorientation was successful in 97.2% of the studies (166/171 + 216/222 + 273/281). No significant difference in success rates for ²⁰¹Tl versus ^99mTc-sestamibi images was found. Conclusion: Our automated approach to myocardial perfusion SPECT processing is highly successful, intrinsically reproducible and can produce time and cost savings while improving accuracy in a clinical or research environment.