|
A Comparative Study between Semi-Quantitative Analysis in [F-18]FDG Brain PET/ CT Scan using Two Different Software Packages in the Diagnosis of Alzheimer's Disease |
|---|---|
| รหัสดีโอไอ | |
| Creator | 1. Maneerat Jubjitt 2. Tanyaluck Thientunyakit 3. Chakmeedaj Sethanandha |
| Title | A Comparative Study between Semi-Quantitative Analysis in [F-18]FDG Brain PET/ CT Scan using Two Different Software Packages in the Diagnosis of Alzheimer's Disease |
| Publisher | Text and Journal Publication |
| Publication Year | 2564 |
| Journal Title | Vajira Medical Journal |
| Journal Vol. | 65 |
| Journal No. | 1 |
| Page no. | 13-26 |
| Keyword | Alzheimer, Semi-quantitative analysis, dementia, PET, PET/CT |
| URL Website | https://tci-thaijo.org/index.php/VMED |
| Website title | Vajira Medical Journal (วชิรเวชสาร) |
| ISSN | 0125-1252 |
| Abstract | Objectives: To compare the results from semi-quantitative analysis of [F-18]FDG brain PET/CT scan obtained from two different software packages (CortexID and Q.brain). In addition, to evaluate the diagnostic performance of 3D-SSP Z-score map images obtained from 2 software packages in the diagnosis of Alzheimer's disease as compared to gold standard. Methods: Retrospective study was done on pre-existing data of [F-18]FDG PET/ CT images acquired from 85 elderly Thai participants (21 cognitively normal elderly subjects, 32 patients with mild cognitive impairment and 32 patients with Alzheimer's disease). Semi-quantitative analysis of all PET images was performed using 2 software packages and Z-score results were compared. The diagnostic performance in Alzheimer's disease was also assessed using gold standard. t-test was applied for statistical analysis and p value <0.05 was considered as statistically significant. Results: There were statistically significant difference in Z-score results at bilateral medial frontal and bilateral occipital association regions using all normalized regions and left posterior cingulate using global cortex and cerebellar normalization. The sensitivity, specificity, accuracy, PPV, NPV, LR- CortexID were 79.17%, 100%, 89.13%, 100%, 81.48% and 0.21, respectively for AD diagnosis, which were better than those of Q. brain. Conclusion: The Z-score results from 2 different software packages in [F-18]FDG brain PET can be significantly different in some regions, which should be careful for interpretation. Pons normalization may reduce this difference. In this pilot study, CortexID software package shows better performance in differentiating AD from normal elderly. |
