الفهرس | Only 14 pages are availabe for public view |
Abstract Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL) patients typically undergo PET/CT with 18F-fluorodeoxyglucose (18F-FDG) at various stages of care, including for precise staging prior to treatment. For initial and conclusive analyses of chemotherapy response, precise quantification of tracer uptake is essential to enhance the consistency of Positron Emission Tomography (PET) in many advanced applications. Standardized Uptake Values (SUV) or their variations usually provide a semi- quantitative measurement of tracer uptake. The convergence of SUVs when iterated, Maximum Likelihood Expectation Maximization (MLEM) or Ordered Subset Expectation Maximization (OSEM) algorithms induce noise in images due to a problem of ill- conditioning, i.e., the results have a significant dependence on small changes in the primary data. Consequently, to avoid an ideal SUV convergence, which is the result of an iterative process that is the closest to the real SUV, the iteration process must be ended before it reaches that point, which is another aspect to consider when criticizing SUV quantification. The ill-conditioning issue might be resolved by including an edge-preserving penalty term in the reconstruction phase. Block sequential regularized expectation maximization (BSREM) is a reconstruction technique used by the Q. Clear algorithm (GE Healthcare, Milwaukee) in line with this strategy; it incorporates noise reduction utilizing a penalty term and point-spread function (PSF) modeling. The penalty term results in smoother cold backgrounds and an enhanced signal-to-noise ratio for hot lesions by imposing more smoothing in lower-activity regions and less in higher-activity areas or parts of high-intensity edges. In addition, applying a penalty function permits efficient SUV convergence, resulting in more accurate data. This study was conducted at the Nuclear medicine Department in Ismailia tumor educational Hospital and included PET/CT scan Images of 50 patients with 145 focal pathologically positive lesions for Lymphoma (Nodal & Extra nodal ). All sample patients are pathologically proven for Lymphoma disease under management & PET CT was done for follow-up studies. This study revealed results which, evaluating the role of the Q.clear algorithm on PET CT imaging in lymphoma patients to compare their Quantitative and Qualitative accuracy. The study result represents changing the Deauville Score (DS), which ultimately resulted in an upgrade to the PET group that was positive, a tremendous significant difference when Q.clear compared to MAC-OSEM in Measuring SUV for lesion of lymphoma, Deauville Score was statistically significantly higher in Q. clear than MAC, statistically substantial variation between Mac and Q. clear regarding measuring Mediastinum uptake for calculating DS and no significant difference between Extra and Nodular lymphoma in MAC regarding Deauville Score The study concluded Q. clear reconstruction algorithm could significantly enhance clinical image quality and the accuracy of the diagnosis of lymphoma lesions. However, Q.Clear increases the SUV of the hypermetabolic outcomes while maintaining the baseline values (which results in a higher signal-to-noise ratio). Because this reconstruction approach may overestimate the total tumour load, these interpretation criteria may no longer be applicable. |