The movement and artifacts related to itself are between the most important problems in imaging procedures and nuclear medicine, both the executive part and the post-processing and interpretation of the images obtained (so that often it is obligated to repeat the exam overload of exposure to the patient as well as an additional workload for the service). The aim is to assess if, in the context of diagnostic PET of the brain is possible to obtain satisfactory results, obtained by analyzing data acquisition at different times, or if the data obtained are, albeit not overlap, otherwise significant, in such a way as not to perform a new acquisition; however the pathology and the age of patients have a strongly influence in the performance of the test and later on the “risk” of movements is higher than other diseases, especially in the final phase of the examination.
We analyzed 24 patients (3 normal and 21 pathological). Each of them, received 185 MBq of 18FDG. Cerebral 18FDG 3D-PET/CT scan imaging was performed 30 minutes after injection of the tracer, avoiding visual, acoustic or mental provocation, with eyes open, 10 min before FDG injection and during the wait. 3-D mode acquisition and CT attenuation correction was used. Discovery690 PET/CT scan was analyzed with visual inspection and automated Z-score method (3D- SSP data). Through “List mode”, we reconstructed the 15’ standard acquisition dividing it into 5, 10 and 15 minutes and processed with the algorithm GE TOF VUE POINT FX post filter 3.2 mm.
For the verification of the reliability of data collected at 5 and 10 minutes was applied to calculate the mean square error (m). The same calculation is possible to derive the value to be considered “acceptable” and that instead “out of tolerance”. Normal patients table: because of the reduced number of normal subjects championships, can’t be calculated an estimate on the correctness of the z-scores measured at 5’ and 10’. Pathological patients’ table: the data of greatest interest that emerges from the analysis is that the errors aren’t acceptable is detected more in the analysis of the first 5 minutes of acquisition, particularly in the frontal lobe (15%) and occipital (20%). According to this general analysis is to advance the hypothesis of considering not evaluable images captured in the first 5 minutes, especially those relating to the frontal lobe and the occipital but consider acceptable images obtained 10’, but considering only the information obtained by the semi quantitative analysis.
Semi-quantitative analysis of PET imaging with [18F]FDG has thus resulted in a parameter of “acceptance”: 10 minutes like the time limit for acquiring sufficient in which the distribution of the radiopharmaceutical is statistically correct in the interpretation of data. The result obtained in 10 ‘on the basis of only semi-quantitative analysis although not superimposed on the data obtained with the acquisition at 15’, has proved still acceptable, in cases where the repetition of the examination is difficult, provide at least the information.

Scarica PDF