OBJECTIVE: In cardiac I-123 (I) imaging downscatter from high energy emissions degrades the image and introduces distortion of semi-quantitative analysis when using a low energy collimator. The effect of a triple energy window (TEW) scatter correction technique, using windows immediately above and below the principal window centered on 159keV, was examined.MATERIALS AND METHODS: A hemispherical cardiac phantom was inserted into a cylindrical phantom and both were filled with radioactive I water solutions. Phantoms were submitted to planar and tomographic scintigraphy under various acquisition and processing conditions, including the use of medium energy (ME) and low energy (LE) collimation.RESULTS: In planar imaging, there was a distance dependent count loss with the LEHR collimator which was partly restored with TEW correction. There was minimal dependence of count rate with distance in using ME collimation. Conversely, the heart to background (H/B) ratio increased with increasing distance with the LEHR collimator, but in applying the TEW correction that ratio paralleled the minimally affected values obtained with the ME collimation. In tomographic imaging the acquired H/B ratio was lower with LE collimation alone, in comparison to the ME collimator, but it was raised significantly when applying the TEW scatter correction. Quantitative measurements also depended on the background method and the reconstruction algorithm applied.CONCLUSION: In cardiac I imaging with a LE collimator the use of TEW scatter correction provides a semi-quantitative assessment comparable to that attained with ME collimation and may moderate inter-institutional inconsistencies.