Object
Title: Optimization of the method of reconstruction of SPECT images performed with use [99mTc-EDDA/HYNIC]octreotate in patients with neuroendocrine tumors
Abstract:
Introduction: Somatostatin receptor scintigraphy (SRS) with use of labeled somatostatin analogs has important role in diagnosis/therapy of neuroendocrine tumors (GEP-NET). Because of its sensitivity and specificity SRS is the method of choice comparing to other USG,CT,MRI techniques. In SRS, receptor expression imaging is directly correlated with intensity of the radiopharmaceutical uptake. SRS is considered useful for clinical indications: localization of primary tumor, evaluation of diseases extension. Evaluation of the uptake level in lesions is the significance indicator in therapy stratification with use of somatostatin analog labeled with 90Y or/and 177Lu. The SPECT images reconstruction is crucial in interpretation of SRS. The choice of appropriate sets of OSEM Flash 3D™ algorithm (subsets and iterations number for defined subsets number) significantly influence on quality of reconstructed images and clinical assessment of target/non-target ratio (TCS/TCB). There is still limited number of the publications which concern the optimization of iterative algorithms applied in reconstruction of SPECT images of GEP-NET. Practically this subject was presented in the several experiments and the clinical trials referred to other diseases, mainly in aspect of comparison of FBP versus MLEM or/and OSEM. The [99mTc-EDDA/HYNIC]octreotate improved the diagnostic eff ; icacy of GEP-NET in localization of primary tumors and their metastases. Currently SRS is important study in taking a decision about use of PRRT. This therapy model is considered as the most effective treatment in inoperable and diffuse GEP-NET. Optimization of the method of reconstruction of the SPECT images is necessary to increase sensitivity and specificity of the SRS. Purpose: Optimization of the iterative OSEM method of reconstruction of SPECT images in patients with GEP-NET and comparison of two reconstruction techniques: FBP versus the optimized OSEM Flash 3D™ algorithm. Materials and method: The results of 76 random patients (42F and 34M) with confirmed GEP-NET were analyzed. GEP-NET was confirmed in histopathological examination and the tumors were localized with use of the CT, MRI or endoscopy studies. 197 lesions (108-localized in the liver and 89-with out-liver localization) were analyzed for the crucial sets of reconstruction parameters. SPECT/CT fusion was used to verify each lesion described in SPECT images. The lesions were divided into two groups because of localization: group I– lesions localized in the liver and group II–with out-liver localization (in abdomen). The SPECT images were analyzed in qualitative and quantitative assessment. The qualitative estimate was the primary coefficient and influence on the consecutive steps in optimi ; zation of the reconstruction parameters. The quality of images was estimated by two independent observers, according to the two-gradual scale: 0-poor visible lesion, 1-very well visible lesion. The qualitative lesions assessment as 0 or 1 was determined by intensity tracer uptake. Special attention was paid to the lesions location near the large gatherings of activity (spleen, kidneys). All lesions in group I and II were divided into four subgroups in dependence on VOI size: subgroup 1-VOI<1,6cm3, subgroup 2 VOI:1,6–3,2cm3, subgroup 3-VOI:3,2–5cm3, subgroup 4–VOI >5cm3. The target/non-target coefficient (TCS/TCB) was analyzed for the quantitative assessment. The VOI of the lesion and background had the same value and was stable for each the reconstruction parameters. The scheme of SPECT images optimization with use of OSEM technique included 34 settings of the reconstruction parameters. The analysis was performed in the consecutive phases: phase 1–the analysis of variable subsets number (s=8,16,32) and variable iterations number (i=2-7), phase 2-the analysis of stable subsets number (s=8) and variable iterations number (i=6-30) with step equal 4 iterations, phase 3–the assessment of influence of the scatter correction on OSEM(8,10) setting, phase 4–the Gauss filter assess (FWHM=n·p[mm]; n = 1,2,3,...,7; p=3,9≈4[mm]) for OSEM(8,10) with used the scatter corre ; ction, phase 5–the comparison of two reconstructions techniques: defined OSEM versus FBP (Butterworth, fc=0,86). All results of the qualitative analysis were compared with the phantom studies. Additionally, the correlation between results of SPECT images analysis for OSEM(8,10) with scatter correction versus receptor overexpression for analog somatostatin was investigated. Results: Phase 1–The qualitative assessment–the lesions localized in the liver (group I): the highest number of lesions was assessed as very well visible (1) for OSEM(8,i) (subgroup 1: 54-63%,p<0,05; subgroup 2: 59-86%, p<0,05; subgroup 3: 88%,p>0,05). The lesions with outliver localization (group II): the highest number of lesions was assessed as 1 for OSEM(8,i) in subgroup 1 and 2 (6-38% and 54-77%, respectively;p<0,05) and OSEM(16,i) in subgroup 3 (93%,p>0,05). The quantitative assessment-for [99mTc-EDDA/HYNIC]octreotate focus uptake in the lesions the mean value of TCS/TCB ratio was:group I–subgroup 1: 1,49±0,43(OSEM(8,i)); 1,55±0,38(OSEM(16,i) and 1,54±0,33(OSEM(32,i); subgroup 2 :2,10±0,71(OSEM(8,i)); 2,14±0,70(OSEM(16,i)) and 2,15±0,63(OSEM(32,i)); subgroup 3:3,66±0,79(OSEM(8,i)); 3,51±0,67(OSEM(16,i)) and 3,51±0,65(OSEM(32,i)). Similar increase of the target/non-target ratio with increasing subsets number was observed in group II: subgroup 1:0,49±0,21; 0,61±0,24 and 0,70±0,25; sub ; group 2:1,98±0,55; 2,02±0,56 and 2,26±0,60; subgroup 3:3,56±0,64; 3,59±0,69 and 3,64±0,67 for OSEM(8,i); OSEM(16,i) and OSEM(32,i), respectively. Phase 2–The qualitative assessment–group I: the highest number of lesions was assessed as 1 for OSEM(8,10) for both subgroups-1 and 2 (44% and 82%, respectively;p<0,05). The same percentage of lesions was assessed as 1 for OSEM(8,14) and OSEM(8,18). The similar trend was observed for the lesions with out-liver localization, p<0,05: subgroup 1–35% (OSEM(8,10)), subgroup 2–89% (OSEM(8,10) and OSEM(8,14)). The quantitative assessment-Group I:the mean value of target/non-target increased with increasing iterations number– subgroup 1:2,44-2,88; subgroup 2:2,66-3,01. The differences in average values of the TCS/TCB ratio are statistically significant for the number of iterations equal:subgroup 1–i=10,14,18 (p<0,05); subgroup 2–i=6,10,14 (p<0,05). The similar increasing TCS/TCB ratio was observed in group II–subgroup 1:1,99-2,54; subgroup 2:2,54-2,96. Differences in average values of the TCS/TCB ratios are statistically significant for the iterations number i=6,10 (p<0,05). Phase 3–The qualitative assessment–scatter correction (SC) applied to the SPECT images reconstruction did not influence on the qualitative assessment of the lesions in group I. The qualitative assessment differences were observed for the lesions in gro ; up II: scatter correction versus without scatter correction (SC vs wSC) – subgroup 1:24%vs15%; subgroup 2:71%vs61%. The observed differences were not statistically significant for each investigated subgroups in group II. The quantitative assessment-Group I:the mean value of SC/wSC ratio in subgroups 1 and 2 was:1,25±0,57 and 1,21±0,59, respectively. Group II:1,32±0,98 and 1,27±0,79, respectively. Phase 4–The qualitative assessment–the most effectively Gauss filter setting was FWHM=4 and 8[mm] (p<0,05). The percentage of the lesions assessed as 1 for group I and II was:22% and 17% (subgroup 1), respectively; 71% and 79% (subgroup 2), respectively. The quantitative assessment-for the FWHM=4[mm] was the highest value of TCS/TCB ratio:group I:2,39±0,46 (subgroup 1) and 2,71±0,64 (subgroup 2) and group II:2,17±0,60 and 2,84±0,84, respectively. Phase 5–The qualitative assessment–the results of the lesions visibility assessment in comparison of two techniques of reconstruction OSEM versus FBP in both groups I and II were–subgroup 1:31%vs8% and 15%vs10%, respectively and subgroup 2:54%vs11% and 82%vs42%, respectively. Differences in quality assessment of SPECT images in subgroup 1 (group II) were not statistically significant. The quantitative assessment–the mean value of TCS/TCB ratio was 1,5 times higher for optimized setting OSEM(8,10) with applied SC versus FBP ; for both groups I and II. The results of analysis of SPECT images were confirmed in the phantom study. Furthermore, correlation between the mean value of target/non-target and overexpressed type of receptor of GEP-NET was investigated (TCS/TCB versus type of SST receptor). The highest correlation was obtained for TCS/TCB ratio vs 2A and 2B type of receptor. The lowest correlation was obtained for the SST receptor type 1. Conclusions: In the analysis of SPECT images of the receptor scintigraphy with the use of labeled somatostatin analogs in patient with GEP-NET it is important to define the particular parameters of the methods of reconstruction to achieve optimized assessment of the SRS study. The choice of reconstructions parameters significantly influence on the quality of SPECT images and the quantitative assessment of target/non-target ratio. OSEM(8,i) is the most effectively setting of reconstructions parameters of OSEM Flash 3D™ algorithm. OSEM(8,10) setting gives the highest results in the qualitative assessment of SRS study. Application of scatter correction (SC) slightly improves quality of image in the visible assessment of the lesions and increases the value of target/non-target ratio. Using Gauss post-filtering slightly degrade contrast with still high quality of images. But this result is possible if Gauss filter’s coefficient is not exceed d ; ouble value of the pixel size. OSEM algorithm gives better outcomes for both the qualitative and quantitative assessment in comparison to FBP method. The choice of reconstruction parameters is the most important in assessment of small and middle-size lesions. The choice of reconstruction parameters does not have influence on diagnostic assessment of large lesions. There is still limited number of publications which concern the optimization of iterative algorithms applied in reconstruction of SPECT images of GEP-NET. There were most often phantom studies. However, the phantom studies cannot be used completely as the indicator in clinical practice. They might only determine the choice of reconstruction parameters. The results from phantom studies always require verification in clinical trials. The key to obtain answer if choice of reconstruction technique and its parameters is appropriate and most effective is always empirical study.
Place of publishing:
Level of degree:
Degree discipline:
Degree grantor:
Promoter:
Date issued:
Identifier:
Call number:
Language:
Access rights:
Object collections:
Last modified:
Mar 15, 2023
In our library since:
Jun 24, 2022
Number of object content hits:
0
Number of object content views in PDF format
0
All available object's versions:
http://dl.cm-uj.krakow.pl:8080/publication/4758
Show description in RDF format:
Show description in OAI-PMH format:
| Edition name | Date |
|---|---|
| ZB-114714 | Mar 15, 2023 |
Objects
Similar
Lenda-Tracz, Wioletta
Holcman, Katarzyna Rita
