scholarly journals Deep learning-assisted ultra-fast/low-dose whole-body PET/CT imaging

2021 ◽  
Author(s):  
Amirhossein Sanaat ◽  
Isaac Shiri ◽  
Hossein Arabi ◽  
Ismini Mainta ◽  
René Nkoulou ◽  
...  
Keyword(s):  
Deep Learning ◽  
Low Dose ◽  
Standardized Uptake Value ◽  
Patient Comfort ◽  
Whole Body ◽  
Acquisition Time ◽  
Lesion Detectability ◽  
Pet Ct ◽  
Malignant Lesions

Abstract Purpose Tendency is to moderate the injected activity and/or reduce acquisition time in PET examinations to minimize potential radiation hazards and increase patient comfort. This work aims to assess the performance of regular full-dose (FD) synthesis from fast/low-dose (LD) whole-body (WB) PET images using deep learning techniques. Methods Instead of using synthetic LD scans, two separate clinical WB 18F-Fluorodeoxyglucose (18F-FDG) PET/CT studies of 100 patients were acquired: one regular FD (~ 27 min) and one fast or LD (~ 3 min) consisting of 1/8th of the standard acquisition time. A modified cycle-consistent generative adversarial network (CycleGAN) and residual neural network (ResNET) models, denoted as CGAN and RNET, respectively, were implemented to predict FD PET images. The quality of the predicted PET images was assessed by two nuclear medicine physicians. Moreover, the diagnostic quality of the predicted PET images was evaluated using a pass/fail scheme for lesion detectability task. Quantitative analysis using established metrics including standardized uptake value (SUV) bias was performed for the liver, left/right lung, brain, and 400 malignant lesions from the test and evaluation datasets. Results CGAN scored 4.92 and 3.88 (out of 5) (adequate to good) for brain and neck + trunk, respectively. The average SUV bias calculated over normal tissues was 3.39 ± 0.71% and − 3.83 ± 1.25% for CGAN and RNET, respectively. Bland-Altman analysis reported the lowest SUV bias (0.01%) and 95% confidence interval of − 0.36, + 0.47 for CGAN compared with the reference FD images for malignant lesions. Conclusion CycleGAN is able to synthesize clinical FD WB PET images from LD images with 1/8th of standard injected activity or acquisition time. The predicted FD images present almost similar performance in terms of lesion detectability, qualitative scores, and quantification bias and variance.

scholarly journals The Impact of Varying Number of OSEM Iterations on Standardized Uptake Value and Image Quality of Discovery STE PET/CT Scanner

2018 ◽  
Vol 4 (Supplement 2) ◽  
pp. 68s-68s
Author(s):  
B. Ali ◽  
A. Afshan ◽  
M.B. Kakakhel
Keyword(s):  
Image Quality ◽  
Standardized Uptake Value ◽  
Acquisition Time ◽  
List Mode ◽  
Ct Scanner ◽  
Matrix Size ◽  
Cold Lesion ◽  
Pet Ct ◽  
The Impact

Background: In PET imaging both quantitative and qualitative interpretations are used. Qualitative and quantitative interpretations depend upon PET/CT image quality that along with many biologic factors strongly depends upon image reconstruction parameters. Aim: The objective of this experimental work was to study the impact of one of the key reconstruction parameter, i.e., number of reconstruction iteration, on standardized uptake value and image quality of PET/CT scan. Methods: Images of NEMA IEC Image Quality Phantom were acquired in list mode for 10 mins on Discovery STE PET/CT scanner, using tumor to background ratio of 4:1 and 18F-FDG as radiotracer. List mode data were further transformed into data sets of varying acquisition time (0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0 mins) per bed position. Transformed data set of 5.0 mins were used to study the impact of varying number of iterations (2, 3, 4, 5, 7, 10, 15, 20) using OSEM approach of iterative reconstruction. Standardized uptake value (SUV) and underestimation in SUV were calculated as quantitative measures, while hot lesion contrast, cold lesion contrast and background variability were calculated as qualitative measures. Results: Standardized uptake value, hot and cold lesion contrast, image spatial resolution and background variability showed increasing trend with increase in reconstruction iterations. Maximum increase of 20.25%, 16.33%, 9.79% and 6.88% was observed in SUV for 10 mm, 13 mm, 17 mm and 22 mm lesions as number of iteration change from 2 to 3. Smallest and the largest diameter lesions showed maximum underestimations of 54.67% and 8.20% at 2 iterations respectively. Percentage hot lesion contrast showed rapid increase as the number of iteration change from 2 to 7 and increased slowly afterward. Background variability range from 4.4% to 6.4%, 4.1%–5.7%, 3.6%–4.6%, 3%–3.8%, 2.7%–3.2%, 2.4%–2.7% for 10.0 mm, 13.0 mm, 17.0 mm, 22.0 mm, 28.0 mm and 37.0 mm sphere respectively. Conclusion: Optimized reconstruction parameters for routine clinical studies 3 iterations with image matrix size of 128 × 128 with filters FWHM of 6 mm and for high resolution studies 3 iterations image matrix size of 256×256 with filters FWHM of 6 mm.

scholarly journals A Prospective Comparison of [18F]FAPI-42 and [68Ga]Ga-FAPI-04 PET/CT in Patients With Various Cancers

2021 ◽  
Author(s):  
Kongzhen Hu ◽  
Lijuan Wang ◽  
Hubing Wu ◽  
Shun Huang ◽  
Ying Tian ◽  
...  
Keyword(s):  
Standardized Uptake Value ◽  
Tumor Detection ◽  
Whole Body ◽  
Acquisition Time ◽  
Bone Lesions ◽  
Tumor Uptake ◽  
Detection Ability ◽  
Pet Ct ◽  
High Level

Abstract Purpose: [18F]FAPI-42 is a new fibroblast activation protein (FAP) specific tracer used for cancer imaging. Here, we describe the in vivo evaluation of [18F]FAPI-42 and compared intra-individual biodistribution, tumor uptake, and detection ability to [68Ga]Ga-FAPI-04.Methods: A total of 22 patients with various types of cancer received [18F]FAPI-42 whole-body positron emission tomography/computed tomography (PET/CT). Among them, 4 patients underwent PET/CT scans, including an early dynamic 20-min, static 1-hour and static 2-hours. The in vivo biodistribution in normal organs and tumor uptake were semi-quantitatively evaluated using the standardized uptake value (SUV) and tumor-to-background ratio (TBR). Furthermore, both [18F]FAPI-42 and [68Ga]Ga-FAPI-04 PET/CT were performed in 12 patients to compare biodistribution, tumor uptake, and tumor detection ability.Results: [18F]FAPI-42 uptake in the tumors was rapid and reached a high level with an average SUVmax of 15.8 at 18 minutes, which stayed at a similarly high level to 2 hours. The optimal image acquisition time for [18F]FAPI-42 was determined to be 1 hour post injection. Compared to [68Ga]Ga-FAPI-04, [18F]FAPI-42 had a higher uptake in the parotid, salivary gland, thyroid, and pancreas (P < 0.05). For tumor detection, [18F]FAPI-42 had a high uptake and could be clearly visualized in the lesions. [18F]FAPI-42 and [68Ga]Ga-FAPI-04 showed the same detectability for 144 positive lesions. In addition, [18F]FAPI-42 had a higher SUVmax in liver and bone lesions (P < 0.05) and higher TBRs in liver, bone, lymph node, pleura and peritonea lesions (all P < 0.05).Conclusion: The present study demonstrates that [18F]FAPI-42 is a good tracer for imaging malignant tumors and exhibited comparable lesion detectability to [68Ga]Ga-FAPI-04. The 1-hour scan is an appropriate time for tumor detection and is superior to the early 10-min scan for the detection of small lesions.Trial registration Chinese Clinical Trial Registry (ChiCTR2100045757)

Restaging of patients with lymphoma

2008 ◽  
Vol 47 (01) ◽  
pp. 37-42 ◽  
Author(s):  
T. Pfluger ◽  
V. Schneider ◽  
M. Hacker ◽  
N. Bröckel ◽  
D. Morhard ◽  
...  
Keyword(s):  
Low Dose ◽  
Non Hodgkin Lymphoma ◽  
Contrast Enhanced ◽  
Pet Ct ◽  
Significant Difference ◽  
18F Fdg ◽  
Sensitivity Specificity ◽  
Fdg Pet Ct

SummaryAim: Assessment of the clinical benefit of i.v. contrast enhanced diagnostic CT (CE-CT) compared to low dose CT with 20 mAs (LD-CT) without contrast medium in combined [18F]-FDG PET/CT examinations in restaging of patients with lymphoma. Patients, methods: 45 patients with non-Hodgkin lymphoma (n = 35) and Hodgkin's disease (n = 10) were included into this study. PET, LD-CT and CECT were analyzed separately as well as side-by-side. Lymphoma involvement was evaluated separately for seven regions. Indeterminate diagnoses were accepted whenever there was a discrepancy between PET and CT findings. Results for combined reading were calculated by rating indeterminate diagnoses according the suggestions of either CT or PET. Each patient had a clinical follow-up evaluation for >6 months. Results: Region-based evaluation suggested a sensitivity/specificity of 66/93% for LD-CT, 87%/91% for CE-CT, 95%/96% for PET, 94%/99% for PET/LD-CT and 96%/99% for PET/CE-CT. The data for PET/CT were obtained by rating indeterminate results according to the suggestions of PET, which turned out to be superior to CT. Lymphoma staging was changed in two patients using PET/ CE-CT as compared to PET/LD-CT. Conclusion: Overall, there was no significant difference between PET/LD-CT and PET/CE-CT. However, PET/CE-CT yielded a more precise lesion delineation than PET/LD-CT. This was due to the improved image quality of CE-CT and might lead to a more accurate investigation of lymphoma.

Low-dose whole-body CT using deep learning image reconstruction: image quality and lesion detection

2021 ◽  
Vol 94 (1121) ◽  
pp. 20201329
Author(s):  
Yoshifumi Noda ◽  
Tetsuro Kaga ◽  
Nobuyuki Kawai ◽  
Toshiharu Miyoshi ◽  
Hiroshi Kawada ◽  
...  
Keyword(s):  
Deep Learning ◽  
Image Reconstruction ◽  
Image Quality ◽  
Low Dose ◽  
Standard Dose ◽  
Lesion Detection ◽  
Whole Body ◽  
Hybrid Iterative Reconstruction ◽  
Whole Body Ct ◽  
Attenuation Values

Objectives: To evaluate image quality and lesion detection capabilities of low-dose (LD) portal venous phase whole-body computed tomography (CT) using deep learning image reconstruction (DLIR). Methods: The study cohort of 59 consecutive patients (mean age, 67.2 years) who underwent whole-body LD CT and a prior standard-dose (SD) CT reconstructed with hybrid iterative reconstruction (SD-IR) within one year for surveillance of malignancy were assessed. The LD CT images were reconstructed with hybrid iterative reconstruction of 40% (LD-IR) and DLIR (LD-DLIR). The radiologists independently evaluated image quality (5-point scale) and lesion detection. Attenuation values in Hounsfield units (HU) of the liver, pancreas, spleen, abdominal aorta, and portal vein; the background noise and signal-to-noise ratio (SNR) of the liver, pancreas, and spleen were calculated. Qualitative and quantitative parameters were compared between the SD-IR, LD-IR, and LD-DLIR images. The CT dose-index volumes (CTDIvol) and dose-length product (DLP) were compared between SD and LD scans. Results: The image quality and lesion detection rate of the LD-DLIR was comparable to the SD-IR. The image quality was significantly better in SD-IR than in LD-IR (p < 0.017). The attenuation values of all anatomical structures were comparable between the SD-IR and LD-DLIR (p = 0.28–0.96). However, background noise was significantly lower in the LD-DLIR (p < 0.001) and resulted in improved SNRs (p < 0.001) compared to the SD-IR and LD-IR images. The mean CTDIvol and DLP were significantly lower in the LD (2.9 mGy and 216.2 mGy•cm) than in the SD (13.5 mGy and 1011.6 mGy•cm) (p < 0.0001). Conclusion: LD CT images reconstructed with DLIR enable radiation dose reduction of >75% while maintaining image quality and lesion detection rate and superior SNR in comparison to SD-IR. Advances in knowledge: Deep learning image reconstruction algorithm enables around 80% reduction in radiation dose while maintaining the image quality and lesion detection compared to standard-dose whole-body CT.

SUVmax of FDG PET-CT Can Be Used As An Identifier of Lesion for AI To Interpret Radiology Reports

2020 ◽  
Author(s):  
Kenji Hirata ◽  
Osamu Manabe ◽  
Keiichi Magota ◽  
Sho Furuya ◽  
Tohru Shiga ◽  
...  
Keyword(s):  
Fdg Pet ◽  
Standardized Uptake Value ◽  
Local Maximum ◽  
Ct Images ◽  
Whole Body ◽  
Training Dataset ◽  
Imaging Biomarkers ◽  
Radiology Reports ◽  
Pet Ct ◽  
Fdg Pet Ct

Abstract Background Radiology reports contribute not only to the particular patient, but also to constructing massive training dataset in the era of artificial intelligence (AI). The maximum standardized uptake value (SUVmax) is often described in daily radiology reports of FDG PET-CT. If SUVmax can be used as an identifier of lesion, that would greatly help AI interpret radiology reports. We aimed to clarify whether the lesion can be localized using SUVmax written in radiology reports.Methods The institutional review board approved this retrospective study. We investigated a total of 112 lesions from 30 FDG PET-CT images acquired with 3 different scanners. SUVmax was calculated from DICOM files based on the latest Quantitative Imaging Biomarkers Alliance (QIBA) publication. The voxels showing the given SUVmax were exhaustively searched in the whole-body images and counted. SUVmax was provided with 5 different degrees of precision: integer (e.g., 3), 1st decimal places (DP) (3.1), 2nd DP (3.14), 3rd DP (3.142), and 4th DP (3.1416). For instance, when SUVmax=3.14 was given, the voxels with 3.135≤SUVmax<3.145 were extracted. We also evaluated whether local maximum restriction could improve the identifying performance, where only the voxels showing the highest intensity within some neighborhood were considered. We defined that “identical detection” was achieved when only single voxel satisfied the criterion.Results A total of 112 lesions from 30 FDG PET-CT images were investigated. SUVmax ranged from 1.3 to 49.1 (median = 5.6, IQR = 5.2). Generally, when larger and more precise SUVmax values were given, fewer voxels satisfied the criterion. The local maximum restriction was very effective. When SUVmax was determined to 4 decimal places (e.g., 3.1416) and the local maximum restriction was applied, identical detection was achieved in 33.3% (lesions with SUVmax<2), 79.5% (2≤SUVmax<5), and 97.8% (5≤SUVmax) of lesions.Conclusions SUVmax of FDG PET-CT can be used as an identifier to localize the lesion if precise SUVmax is provided and local maximum restriction was applied, although the lesions showing SUVmax<2 were difficult to identify. The proposed method may have potential to make use of radiology reports retrospectively for constructing training datasets for AI.

Imaging myeloma and related monoclonal plasma cell disorders using MRI, low-dose whole-body CT and FDG PET/CT

2015 ◽  
Vol 3 (2) ◽  
pp. 95-109
Author(s):  
N. Withofs ◽  
C. Nanni ◽  
P. Simoni ◽  
S. Fanti ◽  
Y. Beguin ◽  
...  
Keyword(s):  
Plasma Cell ◽  
Fdg Pet ◽  
Low Dose ◽  
Whole Body ◽  
Whole Body Ct ◽  
Pet Ct ◽  
Plasma Cell Disorders ◽  
Fdg Pet Ct ◽  
Body Ct

scholarly journals Body region localization in whole-body low-dose CT images of PET/CT scans using virtual landmarks

2019 ◽  
Vol 46 (3) ◽  
pp. 1286-1299 ◽  
Author(s):  
Peirui Bai ◽  
Jayaram K. Udupa ◽  
Yubing Tong ◽  
ShiPeng Xie ◽  
Drew A. Torigian
Keyword(s):  
Low Dose ◽  
Ct Images ◽  
Whole Body ◽  
Ct Scans ◽  
Body Region ◽  
Low Dose Ct ◽  
Pet Ct

The benefit of whole-body low-dose unenhanced multidetector computer tomography (WBLD-MDCT) for follow up and therapy response monitoring in patients with multiple myeloma: Correlation with hematologic parameters

2006 ◽  
Vol 24 (18_suppl) ◽  
pp. 7607-7607
Author(s):  
M. S. Horger ◽  
C. Driessen ◽  
C. Brodoefel ◽  
C. Faul ◽  
P. Pereira ◽  
...  
Keyword(s):  
Multiple Myeloma ◽  
Computer Tomography ◽  
Low Dose ◽  
Laboratory Data ◽  
Whole Body ◽  
Acquisition Time ◽  
Bone Lesions ◽  
Appendicular Skeleton ◽  
Multidetector Computer Tomography

7607 Background: To assessthe value of whole-body low-dose multidetector computer tomography (WBLD-MDCT) as diagnostic and survey modality in multiple myeloma (MM), and as a one-stop alternative (Horger et al. EJR 2005;54:289–297) to established imaging techniques (e.g. x-ray and MRI). Methods: Between 7/2001 and 2/2005, WBLD-MDCT scans were obtained in 90 consecutive patients with histologically proven stage II-III MM, all patients having 2 or more scans (mean = 3,8; range = 2–6). CT-scans were performed using a standardized low-dose protocol and the number, size and density of focal or diffuse medullary (in the appendicular skeleton and pelvis) and extra-medullary lesions as well as osteolysis were analysed for each examination and at follow up. Results were correlated with current standard MM laboratory data and at follow up in order to assess correct temporal recognition of significant myeloma changes by both methods. Results: Detection and follow up of medullary and extra-medullary MM lesions and osteolysis by WBLD-MDCT resulted in a sensitivity of 92%, a specificity of 93%, a NPV of 95%, a PPV of 85% and a likelihood ratio for patients with CT-abnormalities to present changes in the course of their disease of 12. Results of radiologic and hematologic analysis showed high agreement at follow up (median, 3 mo). However, agreement of both techniques at the time of investigation was only moderate (κ = 0.629), with CT being correct in 60% of mismatching cases. Thus, CT enabled earlier detection of MM changes. WBLD-MDCT assessed correctly the course of disease in all 4 patients with nonsecretory MM. Evaluation of stability was optimal in all patients. Conclusions: WBLD-MD represents a reliable, widespread, quick (75s acquisition time), and cost-effective imaging technique in MM, allowing detection of bone marrow involvement, extra-medullary tumors and lytic bone lesions in different clinical settings (staging, follow up, therapy monitoring, evaluation of stability). WBLD-MDCT repeatedly allowed detection of changes in the course of the disease prior to laboratory data, especially in extramedullary MM relapse and nonsecretory MM. No significant financial relationships to disclose.

Sign in / Sign up

Export Citation Format

Share Document