EJNMMI Physics
Latest Publications


TOTAL DOCUMENTS

522
(FIVE YEARS 192)

H-INDEX

19
(FIVE YEARS 6)

Published By Springer (Biomed Central Ltd.)

2197-7364, 2197-7364

2022 ◽  
Vol 9 (1) ◽  
Author(s):  
Roberto Fedrigo ◽  
Dan J. Kadrmas ◽  
Patricia E. Edem ◽  
Lauren Fougner ◽  
Ivan S. Klyuzhin ◽  
...  

Abstract Background Positron emission tomography (PET) with prostate specific membrane antigen (PSMA) have shown superior performance in detecting metastatic prostate cancers. Relative to [18F]fluorodeoxyglucose ([18F]FDG) PET images, PSMA PET images tend to visualize significantly higher-contrast focal lesions. We aim to evaluate segmentation and reconstruction algorithms in this emerging context. Specifically, Bayesian or maximum a posteriori (MAP) image reconstruction, compared to standard ordered subsets expectation maximization (OSEM) reconstruction, has received significant interest for its potential to reach convergence with minimal noise amplifications. However, few phantom studies have evaluated the quantitative accuracy of such reconstructions for high contrast, small lesions (sub-10 mm) that are typically observed in PSMA images. In this study, we cast 3 mm–16-mm spheres using epoxy resin infused with a long half-life positron emitter (sodium-22; 22Na) to simulate prostate cancer metastasis. The anthropomorphic Probe-IQ phantom, which features a liver, bladder, lungs, and ureters, was used to model relevant anatomy. Dynamic PET acquisitions were acquired and images were reconstructed with OSEM (varying subsets and iterations) and BSREM (varying β parameters), and the effects on lesion quantitation were evaluated. Results The 22Na lesions were scanned against an aqueous solution containing fluorine-18 (18F) as the background. Regions-of-interest were drawn with MIM Software using 40% fixed threshold (40% FT) and a gradient segmentation algorithm (MIM’s PET Edge+). Recovery coefficients (RCs) (max, mean, peak, and newly defined “apex”), metabolic tumour volume (MTV), and total tumour uptake (TTU) were calculated for each sphere. SUVpeak and SUVapex had the most consistent RCs for different lesion-to-background ratios and reconstruction parameters. The gradient-based segmentation algorithm was more accurate than 40% FT for determining MTV and TTU, particularly for lesions $$\le$$ ≤  6 mm in diameter (R2 = 0.979–0.996 vs. R2 = 0.115–0.527, respectively). Conclusion An anthropomorphic phantom was used to evaluate quantitation for PSMA PET imaging of metastatic prostate cancer lesions. BSREM with β = 200–400 and OSEM with 2–5 iterations resulted in the most accurate and robust measurements of SUVmean, MTV, and TTU for imaging conditions in 18F-PSMA PET/CT images. SUVapex, a hybrid metric of SUVmax and SUVpeak, was proposed for robust, accurate, and segmentation-free quantitation of lesions for PSMA PET.


2022 ◽  
Vol 9 (1) ◽  
Author(s):  
Defeng Tian ◽  
Hongwei Yang ◽  
Yan Li ◽  
Bixiao Cui ◽  
Jie Lu

Abstract Background Q.Clear is a block sequential regularized expectation maximization penalized-likelihood reconstruction algorithm for Positron Emission Tomography (PET). It has shown high potential in improving image reconstruction quality and quantification accuracy in PET/CT system. However, the evaluation of Q.Clear in PET/MR system, especially for clinical applications, is still rare. This study aimed to evaluate the impact of Q.Clear on the 18F-fluorodeoxyglucose (FDG) PET/MR system and to determine the optimal penalization factor β for clinical use. Methods A PET National Electrical Manufacturers Association/ International Electrotechnical Commission (NEMA/IEC) phantom was scanned on GE SIGNA PET/MR, based on NEMA NU 2-2012 standard. Metrics including contrast recovery (CR), background variability (BV), signal-to-noise ratio (SNR) and spatial resolution were evaluated for phantom data. For clinical data, lesion SNR, signal to background ratio (SBR), noise level and visual scores were evaluated. PET images reconstructed from OSEM + TOF and Q.Clear were visually compared and statistically analyzed, where OSEM + TOF adopted point spread function as default procedure, and Q.Clear used different β values of 100, 200, 300, 400, 500, 800, 1100 and 1400. Results For phantom data, as β value increased, CR and BV of all sizes of spheres decreased in general; images reconstructed from Q.Clear reached the peak SNR with β value of 400 and generally had better resolution than those from OSEM + TOF. For clinical data, compared with OSEM + TOF, Q.Clear with β value of 400 achieved 138% increment in median SNR (from 58.8 to 166.0), 59% increment in median SBR (from 4.2 to 6.8) and 38% decrement in median noise level (from 0.14 to 0.09). Based on visual assessment from two physicians, Q.Clear with β values ranging from 200 to 400 consistently achieved higher scores than OSEM + TOF, where β value of 400 was considered optimal. Conclusions The present study indicated that, on 18F-FDG PET/MR, Q.Clear reconstruction improved the image quality compared to OSEM + TOF. β value of 400 was optimal for Q.Clear reconstruction.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Milan Decuyper ◽  
Jens Maebe ◽  
Roel Van Holen ◽  
Stefaan Vandenberghe

AbstractThe use of deep learning in medical imaging has increased rapidly over the past few years, finding applications throughout the entire radiology pipeline, from improved scanner performance to automatic disease detection and diagnosis. These advancements have resulted in a wide variety of deep learning approaches being developed, solving unique challenges for various imaging modalities. This paper provides a review on these developments from a technical point of view, categorizing the different methodologies and summarizing their implementation. We provide an introduction to the design of neural networks and their training procedure, after which we take an extended look at their uses in medical imaging. We cover the different sections of the radiology pipeline, highlighting some influential works and discussing the merits and limitations of deep learning approaches compared to other traditional methods. As such, this review is intended to provide a broad yet concise overview for the interested reader, facilitating adoption and interdisciplinary research of deep learning in the field of medical imaging.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Tetsuya Sakashita ◽  
Shojiro Matsumoto ◽  
Shigeki Watanabe ◽  
Hirofumi Hanaoka ◽  
Yasuhiro Ohshima ◽  
...  

Abstract Background We recently reported a new absorbed dose conversion method, RAP (RAtio of Pharmacokinetics), for 211At-meta-astatobenzylguanidine (211At-MABG) using a single biodistribution measurement, the percent injected dose/g. However, there were some mathematical ambiguities in determining the optimal timing of a single measurement of the percent injected dose/g. Thus, we aimed to mathematically reconstruct the RAP method and to examine the optimal timing of a single measurement. Methods We derived a new formalism of the RAP dose conversion method at time t. In addition, we acquired a formula to determine the optimal timing of a single measurement of the percent injected dose/g, assuming the one-compartment model for biological clearance. Results We investigated the new formalism’s performance using a representative RAP coefficient with radioactive decay weighting. Dose conversions by representative RAP coefficients predicted the true [211At]MABG absorbed doses with an error of 10% or less. The inverses of the representative RAP coefficients plotted at 4 h post-injection, which was the optimal timing reported in the previous work, were very close to the new inverses of the RAP coefficients 4 h post-injection. Next, the behavior of the optimal timing was analyzed by radiolabeled compounds with physical half-lives of 7.2 h and 10 d on various biological clearance half-lives. Behavior maps of optimal timing showed a tendency to converge to a constant value as the biological clearance half-life of a target increased. The areas of optimal timing for both compounds within a 5% or 10% prediction error were distributed around the optimal timing when the biological clearance half-life of a target was equal to that of the reference. Finally, an example of RAP dose conversion was demonstrated for [211At]MABG. Conclusions The RAP dose conversion method renovated by the new formalism was able to estimate the [211At]MABG absorbed dose using a similar pharmacokinetics, such as [131I]MIBG. The present formalism revealed optimizing imaging time points on absorbed dose conversion between two radiopharmaceuticals. Further analysis and clinical data will be needed to elucidate the validity of a behavior map of the optimal timing of a single measurement for targeted alpha-nuclide therapy.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Deni Hardiansyah ◽  
Ade Riana ◽  
Peter Kletting ◽  
Nouran R. R. Zaid ◽  
Matthias Eiber ◽  
...  

Abstract Background The calculation of time-integrated activities (TIAs) for tumours and organs is required for dosimetry in molecular radiotherapy. The accuracy of the calculated TIAs is highly dependent on the chosen fit function. Selection of an adequate function is therefore of high importance. However, model (i.e. function) selection works more accurately when more biokinetic data are available than are usually obtained in a single patient. In this retrospective analysis, we therefore developed a method for population-based model selection that can be used for the determination of individual time-integrated activities (TIAs). The method is demonstrated at an example of [177Lu]Lu-PSMA-I&T kidneys biokinetics. It is based on population fitting and is specifically advantageous for cases with a low number of available biokinetic data per patient. Methods Renal biokinetics of [177Lu]Lu-PSMA-I&T from thirteen patients with metastatic castration-resistant prostate cancer acquired by planar imaging were used. Twenty exponential functions were derived from various parameterizations of mono- and bi-exponential functions. The parameters of the functions were fitted (with different combinations of shared and individual parameters) to the biokinetic data of all patients. The goodness of fits were assumed as acceptable based on visual inspection of the fitted curves and coefficients of variation CVs < 50%. The Akaike weight (based on the corrected Akaike Information Criterion) was used to select the fit function most supported by the data from the set of functions with acceptable goodness of fit. Results The function $$A_{1} { }\beta { }e^{{ - \left( {\lambda_{1} + \lambda_{{{\text{phys}}}} } \right)t}} + A_{1} { }\left( {1 - \beta } \right){ }e^{{ - \left( {\lambda_{{{\text{phys}}}} } \right)t}}$$ A 1 β e - λ 1 + λ phys t + A 1 1 - β e - λ phys t with shared parameter $$\beta$$ β was selected as the function most supported by the data with an Akaike weight of 97%. Parameters $$A_{1}$$ A 1 and $$\lambda_{1}$$ λ 1 were fitted individually for every patient while parameter $$\beta { }$$ β was fitted as a shared parameter in the population yielding a value of 0.9632 ± 0.0037. Conclusions The presented population-based model selection allows for a higher number of parameters of investigated fit functions which leads to better fits. It also reduces the uncertainty of the obtained Akaike weights and the selected best fit function based on them. The use of the population-determined shared parameter for future patients allows the fitting of more appropriate functions also for patients for whom only a low number of individual data are available.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Wonseok Whi ◽  
Hongyoon Choi ◽  
Jin Chul Paeng ◽  
Gi Jeong Cheon ◽  
Keon Wook Kang ◽  
...  

Abstract Background The whole brain is often covered in [18F]Fluorodeoxyglucose positron emission tomography ([18F]FDG-PET) in oncology patients, but the covered brain abnormality is typically screened by visual interpretation without quantitative analysis in clinical practice. In this study, we aimed to develop a fully automated quantitative interpretation pipeline of brain volume from an oncology PET image. Method We retrospectively collected 500 oncologic [18F]FDG-PET scans for training and validation of the automated brain extractor. We trained the model for extracting brain volume with two manually drawn bounding boxes on maximal intensity projection images. ResNet-50, a 2-D convolutional neural network (CNN), was used for the model training. The brain volume was automatically extracted using the CNN model and spatially normalized. For validation of the trained model and an application of this automated analytic method, we enrolled 24 subjects with small cell lung cancer (SCLC) and performed voxel-wise two-sample T test for automatic detection of metastatic lesions. Result The deep learning-based brain extractor successfully identified the existence of whole-brain volume, with an accuracy of 98% for the validation set. The performance of extracting the brain measured by the intersection-over-union of 3-D bounding boxes was 72.9 ± 12.5% for the validation set. As an example of the application to automatically identify brain abnormality, this approach successfully identified the metastatic lesions in three of the four cases of SCLC patients with brain metastasis. Conclusion Based on the deep learning-based model, extraction of the brain volume from whole-body PET was successfully performed. We suggest this fully automated approach could be used for the quantitative analysis of brain metabolic patterns to identify abnormalities during clinical interpretation of oncologic PET studies.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Peter Frøhlich Staanum ◽  
Anders Floor Frellsen ◽  
Marie Louise Olesen ◽  
Peter Iversen ◽  
Anne Kirstine Arveschoug

Abstract Background Kidney dosimetry after peptide receptor radionuclide therapy using 177Lu-labelled somatostatin analogues is a procedure with multiple steps. We present the SPECT/CT-based implementation at Aarhus University Hospital and evaluate the uncertainty of the various steps in order to estimate the total uncertainty and to identify the major sources of uncertainty. Absorbed dose data from 115 treatment fractions are reported. Results The total absorbed dose with uncertainty is presented for 59 treatments with [177Lu]Lu-DOTATOC and 56 treatments with [177Lu]Lu-DOTATATE. For [177Lu]Lu-DOTATOC the mean and median specific absorbed dose (dose per injected activity) is 0.37 Gy/GBq and 0.38 Gy/GBq, respectively, while for [177Lu]Lu-DOTATATE the median and mean are 0.47 Gy/GBq and 0.46 Gy/GBq, respectively. The uncertainty of the procedure is estimated to be about 13% for a single treatment fraction, where the absorbed dose calculation is based on three SPECT/CT scans 1, 4 and 7 days post-injection, while it increases to about 19% if only a single SPECT/CT scan is performed 1 day post-injection. Conclusions The specific absorbed dose values obtained with the described procedure are comparable to those from other treatment sites for both [177Lu]Lu-DOTATOC and [177Lu]Lu-DOTATATE, but towards the lower end of the range of reported values. The estimated uncertainty is also comparable to that from other reports and judged acceptable for clinical and research use, thus proving the kidney dosimetry procedure a useful tool. The greatest reduction in uncertainty can be obtained by improved activity determination, partial volume correction and additional SPECT/CT scans.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Carlo Chiesa ◽  
Katarina Sjogreen-Gleisner ◽  
Stephan Walrand ◽  
Lidia Strigari ◽  
Glenn Flux ◽  
...  

AbstractThe aim of this standard operational procedure is to standardize the methodology employed for the evaluation of pre- and post-treatment absorbed dose calculations in 90Y microsphere liver radioembolization. Basic assumptions include the permanent trapping of microspheres, the local energy deposition method for voxel dosimetry, and the patient–relative calibration method for activity quantification.The identity of 99mTc albumin macro-aggregates (MAA) and 90Y microsphere biodistribution is also assumed. The large observed discrepancies in some patients between 99mTc-MAA predictions and actual 90Y microsphere distributions for lesions is discussed. Absorbed dose predictions to whole non-tumoural liver are considered more reliable and the basic predictors of toxicity. Treatment planning based on mean absorbed dose delivered to the whole non-tumoural liver is advised, except in super-selective treatments.Given the potential mismatch between MAA simulation and actual therapy, absorbed doses should be calculated both pre- and post-therapy. Distinct evaluation between target tumours and non-tumoural tissue, including lungs in cases of lung shunt, are vital for proper optimization of therapy. Dosimetry should be performed first according to a mean absorbed dose approach, with an optional, but important, voxel level evaluation. Fully corrected 99mTc-MAA Single Photon Emission Computed Tomography (SPECT)/computed tomography (CT) and 90Y TOF PET/CT are regarded as optimal acquisition methodologies, but, for institutes where SPECT/CT is not available, non-attenuation corrected 99mTc-MAA SPECT may be used. This offers better planning quality than non dosimetric methods such as Body Surface Area (BSA) or mono-compartmental dosimetry. Quantitative 90Y bremsstrahlung SPECT can be used if dedicated correction methods are available.The proposed methodology is feasible with standard camera software and a spreadsheet. Available commercial or free software can help facilitate the process and improve calculation time.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Yukito Maeda ◽  
Yuka Yamamoto ◽  
Takashi Norikane ◽  
Katsuya Mitamura ◽  
Tetsuhiro Hatakeyama ◽  
...  

Abstract Background The present study tested the possible utility of fractal analysis from l-[methyl-11C]-methionine (MET) uptake in patients with newly diagnosed gliomas for differentiating glioma, especially in relation to isocitrate dehydrogenase 1 (IDH1) mutation status, and as compared with the conventional standardized uptake value (SUV) parameters. Methods Investigations of MET PET/CT were performed retrospectively in 47 patients with newly diagnosed glioma. Tumors were divided into three groups: lower grade glioma (IDH1-mutant diffuse astrocytoma and IDH1-mutant anaplastic astrocytoma), higher grade glioma (IDH1-wildtype diffuse astrocytoma and IDH1-wildtype anaplastic astrocytoma), and glioblastoma. The fractal dimension for tumor, maximum SUV (SUVmax) for tumor (T) and mean SUV for normal contralateral hemisphere (N) were calculated, and the tumor-to-normal (T/N) ratio was determined. Metabolic tumor volume (MTV) and total lesion MET uptake (TLMU) were also measured. Results There were significant differences in SUVmax (p = 0.006) and T/N ratio (p = 0.02) between lower grade glioma and glioblastoma. There were no significant differences among any of the three groups in MTV or TLMU. Significant differences were obtained in the fractal dimension between lower grade glioma and higher grade glioma (p = 0.006) and glioblastoma (p < 0.001). Conclusions The results of this preliminary study in a small patient population suggest that the fractal dimension using MET PET in patients with newly diagnosed gliomas is useful for differentiating glioma, especially in relation to IDH1 mutation status, which has not been possible with SUV parameters.


2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Richard Laforest ◽  
Mehdi Khalighi ◽  
Yutaka Natsuaki ◽  
Abhejit Rajagopal ◽  
Dharshan Chandramohan ◽  
...  

Abstract Objective Simultaneous PET/MRIs vary in their quantitative PET performance due to inherent differences in the physical systems and differences in the image reconstruction implementation. This variability in quantitative accuracy confounds the ability to meaningfully combine and compare data across scanners. In this work, we define image reconstruction parameters that lead to comparable contrast recovery curves across simultaneous PET/MRI systems. Method The NEMA NU-2 image quality phantom was imaged on one GE Signa and on one Siemens mMR PET/MRI scanner. The phantom was imaged at 9.7:1 contrast with standard spheres (diameter 10, 13, 17, 22, 28, 37 mm) and with custom spheres (diameter: 8.5, 11.5, 15, 25, 32.5, 44 mm) using a standardized methodology. Analysis was performed on a 30 min listmode data acquisition and on 6 realizations of 5 min from the listmode data. Images were reconstructed with the manufacturer provided iterative image reconstruction algorithms with and without point spread function (PSF) modeling. For both scanners, a post-reconstruction Gaussian filter of 3–7 mm in steps of 1 mm was applied. Attenuation correction was provided from a scaled computed tomography (CT) image of the phantom registered to the MR-based attenuation images and verified to align on the non-attenuation corrected PET images. For each of these image reconstruction parameter sets, contrast recovery coefficients (CRCs) were determined for the SUVmean, SUVmax and SUVpeak for each sphere. A hybrid metric combining the root-mean-squared discrepancy (RMSD) and the absolute CRC values was used to simultaneously optimize for best match in CRC between the two scanners while simultaneously weighting toward higher resolution reconstructions. The image reconstruction parameter set was identified as the best candidate reconstruction for each vendor for harmonized PET image reconstruction. Results The range of clinically relevant image reconstruction parameters demonstrated widely different quantitative performance across cameras. The best match of CRC curves was obtained at the lowest RMSD values with: for CRCmean, 2 iterations-7 mm filter on the GE Signa and 4 iterations-6 mm filter on the Siemens mMR, for CRCmax, 4 iterations-6 mm filter on the GE Signa, 4 iterations-5 mm filter on the Siemens mMR and for CRCpeak, 4 iterations-7 mm filter with PSF on the GE Signa and 4 iterations-7 mm filter on the Siemens mMR. Over all reconstructions, the RMSD between CRCs was 1.8%, 3.6% and 2.9% for CRC mean, max and peak, respectively. The solution of 2 iterations-3 mm on the GE Signa and 4 iterations-3 mm on Siemens mMR, both with PSF, led to simultaneous harmonization and with high CRC and low RMSD for CRC mean, max and peak with RMSD values of 2.8%, 5.8% and 3.2%, respectively. Conclusions For two commercially available PET/MRI scanners, user-selectable parameters that control iterative updates, image smoothing and PSF modeling provide a range of contrast recovery curves that allow harmonization in harmonization strategies of optimal match in CRC or high CRC values. This work demonstrates that nearly identical CRC curves can be obtained on different commercially available scanners by selecting appropriate image reconstruction parameters.


Sign in / Sign up

Export Citation Format

Share Document