Masters Degrees (Nuclear Medicine)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/3507

Browse

Browsing Masters Degrees (Nuclear Medicine) by Author "Munemo, Lionel Tapiwa"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Semiquantitative analysis of FDG PET brain scans using Neurostat and SPM
    (Stellenbosch : Stellenbosch University, 2021-12) Munemo, Lionel Tapiwa; Doruyter, Alex G. G.; Warwick, James; Stellenbosch University. Faculty of Medicine and Health Sciences. Dept. of Medical Imaging and Clinical Oncology. Nuclear Medicine.
    ENGLISH SUMMARY : Background: Positron emission tomography using Fluorine-18 2-Fluoro-2-Deoxy-D-Glucose (FDG-PET) has an established role in the investigation of multiple neurological conditions such as neurolupus and epilepsy. FDG PET brain scans are interpreted visually and semi-quantitatively using tools such as Neurostat or SPM. The normal databases used in Neurostat were created over 20 years ago, however since this time there has been a steady advance in camera technology and reconstruction algorithms for semiquantitative analysis. It is possible that results obtained with semiquantitative analysis using Neurostat (old PET databases) are different from results obtained with semiquantitative analysis using SPM (using a new PET database). The aim of this study was to evaluate whether there are differences in study interpretation when the same scans are read in combination with a Neurostat analysis, compared to when they are read in combination with an SPM analysis using a database built with new scan data. Methods: First, a local normal database was built with healthy control data using MATLAB R2014b and SPM12. These scans were obtained from previous research projects and prospectively enrolled participants. All scans were acquired on a Philips Gemini-TF Big bore PET/CT scanner. Second, a test dataset, comprising a mixture of clinical scans and scans of prospectively enrolled participants (other than those included in the normal database), was used to compare interpretation of results when using the SPM-based normal database (SPM) to interpretation results when using Neurostat (NS). Results: A database of normal FDG Brain PET/CT studies for the 19-35 year age group was created from a total of 26 healthy controls (13 men and 13 women – optimally matched for age and gender). The test dataset was comprised of 20 scans in the same age range: 15 clinical cases and 5 controls. Next, two expert readers read scans in the test dataset with the assistance of both the NS outputs and the SPM outputs, in separate reading sessions. There was no statistically significant difference in whether a scan was called normal or abnormal using either Neurostat or SPM. There was also no significant difference when comparing the number of lesions identified. SPM scored lesions as less severe than Neurostat (p=0.006). Conclusion: SPM-based analysis using a locally-developed normal database (with scans acquired on a modern PET-CT scanner) yielded similar results to Neurostat, justifying its continued use. Further evaluation to determine if these results are applicable to older patients with neurodegenerative conditions such as Alzheimer’s disease is planned.