Dataset Information

A deep learning framework for ¹⁸F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy.

ABSTRACT:

Purpose

Epilepsy is one of the most disabling neurological disorders, which affects all age groups and often results in severe consequences. Since misdiagnoses are common, many pediatric patients fail to receive the correct treatment. Recently, ¹⁸F-fluorodeoxyglucose positron emission tomography (¹⁸F-FDG PET) imaging has been used for the evaluation of pediatric epilepsy. However, the epileptic focus is very difficult to be identified by visual assessment since it may present either hypo- or hyper-metabolic abnormality with unclear boundary. This study aimed to develop a novel symmetricity-driven deep learning framework of PET imaging for the identification of epileptic foci in pediatric patients with temporal lobe epilepsy (TLE).

Methods

We retrospectively included 201 pediatric patients with TLE and 24 age-matched controls who underwent ¹⁸F-FDG PET-CT studies. ¹⁸F-FDG PET images were quantitatively investigated using 386 symmetricity features, and a pair-of-cube (PoC)-based Siamese convolutional neural network (CNN) was proposed for precise localization of epileptic focus, and then metabolic abnormality level of the predicted focus was calculated automatically by asymmetric index (AI). Performances of the proposed framework were compared with visual assessment, statistical parametric mapping (SPM) software, and Jensen-Shannon divergence-based logistic regression (JS-LR) analysis.

Results

The proposed deep learning framework could detect the epileptic foci accurately with the dice coefficient of 0.51, which was significantly higher than that of SPM (0.24, P < 0.01) and significantly (or marginally) higher than that of visual assessment (0.31-0.44, P = 0.005-0.27). The area under the curve (AUC) of the PoC classification was higher than that of the JS-LR (0.93 vs. 0.72). The metabolic level detection accuracy of the proposed method was significantly higher than that of visual assessment blinded or unblinded to clinical information (90% vs. 56% or 68%, P < 0.01).

Conclusion

The proposed deep learning framework for ¹⁸F-FDG PET imaging could identify epileptic foci accurately and efficiently, which might be applied as a computer-assisted approach for the future diagnosis of epilepsy patients.

Trial registration

NCT04169581. Registered November 13, 2019 Public site: https://clinicaltrials.gov/ct2/show/NCT04169581.

SUBMITTER: Zhang Q

PROVIDER: S-EPMC8241642 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesTo investigate whether quantitative T2 mapping is complementary to [18F]FDG PET in epileptogenic zone detection, thus improving the lateralization accuracy for drug-resistant mesial temporal lobe epilepsy (MTLE) using hybrid PET/MR.MethodsWe acquired routine structural MRI, T2-weighted FLAIR, whole brain T2 mapping, and [18F]FDG PET in 46 MTLE patients and healthy controls on a hybrid PET/MR scanner, followed with computing voxel-based z-score maps of patients in reference to healthy controls. Asymmetry indexes of the hippocampus were calculated for each imaging modality, which then enter logistic regression models as univariate or multivariate for lateralization. Stereoelectroencephalography (SEEG) recordings and clinical decisions were collected as gold standard.ResultsRoutine structural MRI and T2w-FLAIR lateralized 47.8% (22/46) of MTLE patients, and FDG PET lateralized 84.8% (39/46). T2 mapping combined with [18F]FDG PET improved the lateralization accuracy by correctly lateralizing 95.6% (44/46) of MTLE patients. The asymmetry indexes of hippocampal T2 relaxometry and PET exhibit complementary tendency in detecting individual laterality, especially for MR-negative patients. In the quantitative analysis of z-score maps, the ipsilateral hippocampus had significantly lower SUVR (LTLE, p < 0.001; RTLE, p < 0.001) and higher T2 value (LTLE, p < 0.001; RTLE, p = 0.001) compared to the contralateral hippocampus. In logistic regression models, PET/T2 combination resulted in the highest AUC of 0.943 in predicting lateralization for MR-negative patients, followed by PET (AUC = 0.857) and T2 (AUC = 0.843).ConclusionsThe combination of quantitative T2 mapping and [18F]FDG PET could improve lateralization for temporal lobe epilepsy.Key points• Quantitative T2 mapping and18F-FDG PET are complementary in the characterization of hippocampal alterations of MR-negative temporal lobe epilepsy patients. • The combination of quantitative T2 and18F-FDG PET obtained from hybrid PET/MR could improve lateralization for temporal lobe epilepsy.

Project description:Knowledge of the within-subject variability of 18F-FDG PET/MRI measurements is necessary for proper interpretation of quantitative PET or MRI metrics in the context of therapeutic efficacy assessments with integrated PET/MRI scanners. The goal of this study was to determine the test-retest repeatability of these metrics on PET/MRI, with comparison to similar metrics acquired by PET/CT. Methods: This prospective study enrolled subjects with pathology-proven pelvic malignancies. Baseline imaging consisted of PET/CT immediately followed by PET/MRI, using a single 370-MBq 18F-FDG dose. Repeat imaging was performed within 7 d using an identical imaging protocol, with no oncologic therapy between sessions. PET imaging on both scanners consisted of a list-mode acquisition at a single pelvic station. The MRI consisted of 2-point Dixon imaging for attenuation correction, standard sequences for anatomic correlation, and diffusion-weighted imaging. PET data were statically reconstructed using various frame durations and minimizing uptake time differences between sessions. SUV metrics were extracted for both PET/CT and PET/MRI in each imaging session. Apparent diffusion coefficient (ADC) metrics were extracted for both PET/MRI sessions. Results: The study cohort consisted of 14 subjects (13 female, 1 male) with various pelvic cancers (11 cervical, 2 rectal, 1 endometrial). For SUVmax, the within-subject coefficient of variation (wCV) appeared higher for PET/CT (8.5%-12.8%) than PET/MRI (6.6%-8.7%) across all PET reconstructions, though with no significant repeatability differences (all P values ≥ 0.08) between modalities. For lean body mass-adjusted SUVpeak, the wCVs appeared similar for PET/CT (9.9%-11.5%) and PET/MRI (9.2%-11.3%) across all PET reconstructions, again with no significant repeatability differences (all P values ≥ 0.14) between modalities. For PET/MRI, the wCV for ADCmedian of 3.5% appeared lower than the wCVs for SUVmax (6.6%-8.7%) and SULpeak (9.2%-11.3%), though without significant repeatability differences (all P values ≥ 0.23). Conclusion: For solid tumors of the pelvis, the repeatability of the evaluated SUV and ADC metrics on 18F-FDG PET/MRI is both acceptably high and similar to previously published values for 18F-FDG PET/CT and MRI, supporting the use of 18F-FDG PET/MRI for quantitative oncologic treatment response assessments.

Project description:BackgroundConsidering the limited accessibility of amyloid position emission tomography (PET) in patients with dementia, we proposed a deep learning (DL)-based amyloid PET positivity classification model from PET images with 2-deoxy-2-[fluorine-18]fluoro-D-glucose (2-[18F]FDG).MethodsWe used 2-[18F]FDG PET datasets from the Alzheimer's Disease Neuroimaging Initiative and Korean Brain Aging Study for the Early diagnosis and prediction of Alzheimer's disease for model development. Moreover, we used an independent dataset from another hospital. A 2.5-D deep learning architecture was constructed using 291 submodules and three axes images as the input. We conducted the voxel-wise analysis to assess the regions with substantial differences in glucose metabolism between the amyloid PET-positive and PET-negative participants. This facilitated an understanding of the deep model classification. In addition, we compared these regions with the classification probability from the submodules.ResultsThere were 686 out of 1433 (47.9%) and 50 out of 100 (50%) amyloid PET-positive participants in the training and internal validation datasets and the external validation datasets, respectively. With 50 times iterations of model training and validation, the model achieved an AUC of 0.811 (95% confidence interval (CI) of 0.803-0.819) and 0.798 (95% CI, 0.789-0.807) on the internal and external validation datasets, respectively. The area under the curve (AUC) was 0.860 when tested with the model with the highest value (0.864) on the external validation dataset. Moreover, it had 75.0% accuracy, 76.0% sensitivity, 74.0% specificity, and 75.0% F1-score. We found an overlap between the regions within the default mode network, thus generating high classification values.ConclusionThe proposed model based on the 2-[18F]FDG PET imaging data and a DL framework might successfully classify amyloid PET positivity in clinical practice, without performing amyloid PET, which have limited accessibility.

Dataset Information

A deep learning framework for ¹⁸F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy.

Purpose

Methods

Results

Conclusion

Trial registration

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

A deep learning framework for 18F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy.

Purpose

Methods

Results

Conclusion

Trial registration

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

A deep learning framework for ¹⁸F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy.