The effect of incorporating the quantitative analysis besides visual assessments of 18F-FDG brain PET images for the localization of epileptogenic zones

Document Type : Conference Proceedings


1 Department of Medical Physics, Shahid Sadoughi University of Medical Sciences, Yazd, Iran

2 Research Center for Nuclear Medicine, Shariati Hospital, Tehran University of Medical Sciences , Tehran, Iran ,


FDG Brain PET is a valuable paraclinical tool for presurgical assessments of patients suffering from refractory epilepsy. By the widespread accessibility of PET, recognizing the functional lesions has become a current practical method especially in preoperative evaluations of the partial epilepsy disorders. The aim of our study was to assess the impact of quantitative software-aided method in localizing epileptic zones in human brain using 18F-FDG Brain PET scan beside the physician qualitative interpretations.
Materials and Methods: The subjects have been selected among patients referred to brain PET imaging who were suffering from refractory epilepsy. Seventeen patients (9 men, 8 women
; 25±11 YO) underwent PET scan. Quantitative analysis of brain functional images were applied by the Scenium software which provided statistical color-coded images of Z-scores and standard deviations (1). We considered Z-scores<-2 as an abnormal range of brain metabolism. It is worth mentioning that, we categorized the patients based on their MRI investigations to lesional and non-lesional groups. Two physicians presented visual assessments in order to localizing epileptic spots. Also, the kappa index had been calculated as an agreement coefficient via SPSS (V.19.0) software.
Results: In the case of assessing the accordance of lesions’ localization among MRI, EEG, and PET/CT evaluations, the agreement level of EEG versus quantitative and qualitative method were reported 0.67 and 0.12, respectively, which were related to non-lesional group. Also, in the lesional category, the kappa indices for quantitative and qualitative analysis against the suspicious lesions on MRI were 0.56 and 0.27, respectively
Conclusion: Epilepsy disorder has destructive effects on quality of life. Thereby, accurate diagnosis of lesion locations could have a vital impact in order to patients’ treatment, exclusively in refractory epilepsy which might treat through doing the brain’s operation (2). The foregoing discussion is a confirmation of the supplementary role of quantitative software alongside visual assessments, in order to do pre-surgical evaluations of epileptic positions in human brain via 18F-FDG Brain PET images. Hence, utilizing of quantitative software could be a complementary method together with MRI and EEG procedures to improve the confidence of surgeon about the epileptic spots(3).