Purpose
The performance of
68
Ga-PSMA PET/CT-MR has been evaluated in prostate cancer (PCa), showing significant results. However, even a technically accurate imaging procedure requires a high ...interobserver agreement in its interpretation to implement in patients’ management. This study aims to perform a systematic review and meta-analysis on the interobserver variability in
68
Ga-PSMA PET/CT-MR imaging in PCa patients.
Methods
We conducted a systematic review and meta-analysis on the interobserver variability, including studies: (1) providing Kappa (K) as the inter-observer agreement test or the essential data to calculate it, (2) providing the K confidence interval or the essential crude data to calculate it, (3) measuring K statistic based on the appropriate use criteria for the inter-observer agreement.
Results
Twelve studies, providing 1585
68
Ga-PSMA PET/CT-MR studies reviewed by 62 independent readers, were included. In general, the pooled inter-observer agreement was interpreted as substantial for all analyzed groups, including tumoral lesions in the prostate bed, lymphadenopathies, bone metastasis, and soft-tissue metastasis (all between 0.6 and 0.8). The regional lymphadenopathy group (0.74) obtained the highest agreement, while the lowest was for soft tissue metastasis (0.65).
Conclusion
This study showed a substantial interobserver agreement in the overall interpretation and detecting locoregional and distant involvement with
68
Ga-PSMA PET/CT-MR in PCa patients.
Deep learning for whole-body medical image generation Schaefferkoetter, Joshua; Yan, Jianhua; Moon, Sangkyu ...
European journal of nuclear medicine and molecular imaging,
11/2021, Volume:
48, Issue:
12
Journal Article
Peer reviewed
Background
Artificial intelligence (AI) algorithms based on deep convolutional networks have demonstrated remarkable success for image transformation tasks. State-of-the-art results have been ...achieved by generative adversarial networks (GANs) and training approaches which do not require paired data. Recently, these techniques have been applied in the medical field for cross-domain image translation.
Purpose
This study investigated deep learning transformation in medical imaging. It was motivated to identify generalizable methods which would satisfy the simultaneous requirements of quality and anatomical accuracy across the entire human body. Specifically, whole-body MR patient data acquired on a PET/MR system were used to generate synthetic CT image volumes. The capacity of these synthetic CT data for use in PET attenuation correction (AC) was evaluated and compared to current MR-based attenuation correction (MR-AC) methods, which typically use multiphase Dixon sequences to segment various tissue types.
Materials and methods
This work aimed to investigate the technical performance of a GAN system for general MR-to-CT volumetric transformation and to evaluate the performance of the generated images for PET AC. A dataset comprising matched, same-day PET/MR and PET/CT patient scans was used for validation.
Results
A combination of training techniques was used to produce synthetic images which were of high-quality and anatomically accurate. Higher correlation was found between the values of mu maps calculated directly from CT data and those derived from the synthetic CT images than those from the default segmented Dixon approach. Over the entire body, the total amounts of reconstructed PET activities were similar between the two MR-AC methods, but the synthetic CT method yielded higher accuracy for quantifying the tracer uptake in specific regions.
Conclusion
The findings reported here demonstrate the feasibility of this technique and its potential to improve certain aspects of attenuation correction for PET/MR systems. Moreover, this work may have larger implications for establishing generalized methods for inter-modality, whole-body transformation in medical imaging. Unsupervised deep learning techniques can produce high-quality synthetic images, but additional constraints may be needed to maintain medical integrity in the generated data.
Graphical abstract
•This study examined the stability of radiomic features from T2-weighted MRI of cervical cancer.•Three tests were applied: 1. test–retest; 2. diagnostic vs simulation MRI; and 3. inter-observer ...delineation.•The inter-observer cohort had the greatest number of reproducible features.•The diagnostic–simulation cohort had the fewest reproducible features.•Shape features were the most stable features in all three cohorts.
The aims of this study are to evaluate the stability of radiomic features from T2-weighted MRI of cervical cancer in three ways: (1) repeatability via test–retest; (2) reproducibility between diagnostic MRI and simulation MRI; (3) reproducibility in inter-observer setting.
This retrospective cohort study included FIGO stage IB-IVA cervical cancer patients treated with chemoradiation between 2005 and 2014. There were three cohorts of women corresponding to each aim of the study: (1) 8 women who underwent test–retest MRI; (2) 20 women who underwent MRI on different scanners (diagnostic and simulation MRI); (3) 34 women whose diagnostic MRIs were contoured by three observers. Radiomic features based on first-order statistics, shape features and texture features were extracted from the original, Laplacian of Gaussian (LoG)-filtered and wavelet-filtered images, for a total of 1761 features. Stability of radiomic features was assessed using intraclass correlation coefficient (ICC).
The inter-observer cohort had the most reproducible features (95.2% with ICC ≥0.75) whereas the diagnostic–simulation cohort had the fewest (14.1% with ICC ≥0.75). Overall, 229 features had ICC ≥0.75 in all three tests. Shape features emerged as the most stable features in all cohorts.
The diagnostic–simulation test resulted in the fewest reproducible features. Further research in MRI-based radiomics is required to validate the use of reproducible features in prognostic models.
Objective
The aim of this study was to systematically review the literature to evaluate the clinical performance of integrated
18
F-FDG PET/MR as compared with
18
F-FDG PET/CT in oncologic imaging.
...Methods
The literature was searched using MEDLINE and EMBASE via OVID. Studies comparing the diagnostic accuracy of integrated
18
F-FDG PET/MR and
18
F-FDG PET/CT in the diagnosis, staging/restaging, assessment of treatment response, or evaluation of metastasis in patients with suspected or diagnosed cancers were deemed eligible for inclusion. Risk of bias and applicability concerns were assessed using the QUADAS-2 tool.
Results
Twenty studies met the inclusion criteria. The overall quality of the studies was rated favorably with bias or applicability concerns in a few studies. Our review suggests that
18
F-FDG PET/MR performs comparably to
18
F-FDG PET/CT in the detection of local lymph node and distant metastases and superiorly in determining the local extent of tumor. SUV obtained from
18
F-FDG PET/MR correlated highly with those obtained from
18
F-FDG PET/CT.
Conclusions
Based on early evidence,
18
F-FDG PET/MR is comparable to
18
F-FDG PET/CT in the clinical scenarios examined in this review. The potential for interchangeability of
18
F-FDG PET/MR with
18
F-FDG PET/CT will vary by indication and the body site that is being imaged, with PET scanners integrated with MRI predicted to provide greater detail in the evaluation of local tumor extent, where
18
F-FDG PET/CT can be limited.
Management of Wolffian adnexal tumors Kim, Soyoun Rachel; Heredia, Fernando; Pakbaz, Sara ...
International journal of gynecological cancer,
06/2021, Volume:
31, Issue:
6
Journal Article
Goal
PET is a relatively noisy process compared to other imaging modalities, and sparsity of acquisition data leads to noise in the images. Recent work has focused on machine learning techniques to ...improve PET images, and this study investigates a deep learning approach to improve the quality of reconstructed image volumes through denoising by a 3D convolution neural network. Potential improvements were evaluated within a clinical context by physician performance in a reading task.
Methods
A wide range of controlled noise levels was emulated from a set of chest PET data in patients with lung cancer, and a convolutional neural network was trained to denoise the reconstructed images using the full-count reconstructions as the ground truth. The benefits, over conventional Gaussian smoothing, were quantified across all noise levels by observer performance in an image ranking and lesion detection task.
Results
The CNN-denoised images were generally ranked by the physicians equal to or better than the Gaussian-smoothed images for all count levels, with the largest effects observed in the lowest-count image sets. For the CNN-denoised images, overall lesion contrast recovery was 60% and 90% at the 1 and 20 million count levels, respectively. Notwithstanding the reduced lesion contrast recovery in noisy data, the CNN-denoised images also yielded better lesion detectability in low count levels. For example, at 1 million true counts, the average true positive detection rate was around 40% for the CNN-denoised images and 30% for the smoothed images.
Conclusion
Significant improvements were found for CNN-denoising for very noisy images, and to some degree for all noise levels. The technique presented here offered however limited benefit for detection performance for images at the count levels routinely encountered in the clinic.
Purpose
To evaluate the diagnostic and prognostic significance of combined cardiac
18
F-fluorodeoxyglucose (FDG) PET/MRI with T1/T2 mapping in the evaluation of suspected cardiac sarcoidosis.
Methods
...Patients with suspected cardiac sarcoidosis were prospectively enrolled for cardiac
18
F-FDG PET/MRI, including late gadolinium enhancement (LGE) and T1/T2 mapping with calculation of extracellular volume (ECV). The final diagnosis of cardiac sarcoidosis was established using modified JMHW guidelines. Major adverse cardiac events (MACE) were assessed as a composite of cardiovascular death, ventricular tachyarrhythmia, bradyarrhythmia, cardiac transplantation or heart failure. Statistical analysis included Cox proportional hazard models.
Results
Forty-two patients (53 ± 13 years, 67% male) were evaluated, 13 (31%) with a final diagnosis of cardiac sarcoidosis. Among patients with cardiac sarcoidosis, 100% of patients had at least one abnormality on PET/MRI: FDG uptake in 69%, LGE in 100%, elevated T1 and ECV in 100%, and elevated T2 in 46%. FDG uptake co-localized with LGE in 69% of patients with cardiac sarcoidosis compared to 24% of those without,
p
= 0.014. Diagnostic specificity for cardiac sarcoidosis was highest for FDG uptake (69%), elevated T2 (79%), and FDG uptake co-localizing with LGE (76%). Diagnostic sensitivity was highest for LGE, elevated T1 and ECV (100%). After median follow-up duration of 634 days, 13 patients experienced MACE. All patients who experienced MACE had LGE, elevated T1 and elevated ECV. FDG uptake (HR 14.7,
p
= 0.002), elevated T2 (HR 9.0,
p
= 0.002) and native T1 (HR 1.1 per 10 ms increase,
p
= 0.044) were significant predictors of MACE even after adjusting for left ventricular ejection fraction and immune suppression treatment. The presence of FDG uptake co-localizing with LGE had the highest diagnostic performance overall (AUC 0.73) and was the best predictor of MACE based on model goodness of fit (HR 14.9,
p
= 0.001).
Conclusions
Combined cardiac FDG-PET/MRI with T1/T2 mapping provides complementary diagnostic information and predicts MACE in patients with suspected cardiac sarcoidosis.
Background
In early oral cavity cancer, elective neck dissection (END) for the clinically node‐negative (cN0) neck improves survival compared with observation. This paradigm has been challenged ...recently by the use of positron emission tomography–computed tomography (PET‐CT) imaging in the cN0 neck. To inform this debate, we performed an economic evaluation comparing PET‐CT–guided therapy with routine END in the cN0 neck.
Methods
Patients with T1‐2N0 lateralized oral tongue cancer were analyzed. A Markov model over a 40‐year time horizon simulated treatment, disease recurrence, and survival from a US health care payer perspective. Model parameters were derived from a review of the literature.
Results
The END strategy was dominant, with a cost savings of $1576.30 USD, an increase of 0.055 quality‐adjusted life years (QALYs), a net monetary benefit of $4303 USD, and a 0.22 life‐year advantage. END was sensitive to variation in cost and utilities in deterministic and probabilistic sensitivity analyses. PET‐CT became the preferred strategy when decreasing occult nodal disease to 18% and increasing the negative predictive value (NPV) of PET‐CT to 89% in 1‐way sensitivity analyses. In probabilistic sensitivity analysis, assuming a cost effectiveness threshold of $50,000 USD/QALY, END was dominant in 64% of simulations and cost effective in 69.8%.
Conclusion
END is a cost‐effective strategy compared with PET‐CT in patients who have node‐negative oral cancer. Although lower PET standardized uptake value thresholds would result in fewer false negatives and improved NPV, it is still uncertain that PET‐CT would be cost effective, as this would likely result in more false positive tests.
Positron emission tomography–computed tomography (PET‐CT) has been investigated as a management strategy for clinically node‐negative early oral cavity cancer to avoid unnecessary morbidity associated with elective neck dissection in patients who are free of occult disease. However, this study demonstrates that elective neck dissection remains cost effective over PET‐CT–guided management of the clinically node‐negative neck in patients with early oral cavity cancer.