METHODS: Single-photon-emission computed tomography studies were carried out for plastic bottles of various sizes as target objects containing Technetium-99m placed in a cylindrical water phantom. The phantoms were created to have varying ratios of target activity concentration to background activity concentration (T/B) ranging from 4.00 to 1.25. From the single-photon-emission computed tomography slices, the volume of the target object was estimated using an iterative thresholding method. A unique threshold that estimated the volume of the target object was determined and declared as ideal threshold for each level of background activity. Simple linear regression analysis was carried out to explore the relationship between the actual volume and estimated volume.
RESULTS: The analysis shows a statistically significant linear relationship (P < .01) between actual volume and estimated volume. For all ratios of target activity concentration to background activity concentration studied, the relative errors in estimated volumes of target objects are within ±7%. Thresholds that estimated volume of target objects vary depending on the level of background activity.
CONCLUSIONS: It is clearly demonstrated that background activity highly influenced the volume estimation. Therefore, background activity of the target object needs to be considered when selecting the ideal thresholds to obtain accurate volume estimation of target objects.