Related Experiment Video
Updated: Nov 22, 2025
Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET
Published on: October 22, 2019
A deep learning framework for 18F-FDG PET imaging diagnosis in pediatric patients with temporal lobe epilepsy
Qinming Zhang1,2, Yi Liao1,3, Xiawan Wang1,3
1Department of Nuclear Medicine and PET-CT Center, The Second Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.
Insights
A novel deep learning framework accurately identifies epileptic foci in pediatric temporal lobe epilepsy using 18F-FDG PET imaging. This computer-assisted approach improves diagnostic accuracy over traditional methods.
Area of Science:
- Neurology
- Medical Imaging
- Artificial Intelligence
Background:
- Epilepsy is a disabling neurological disorder with common misdiagnoses, particularly in pediatric cases.
- Accurate identification of epileptic foci is crucial for effective treatment, but visual assessment of 18F-FDG PET scans can be challenging due to unclear boundaries of metabolic abnormalities.
- Temporal lobe epilepsy (TLE) is a common form of epilepsy requiring precise localization of the seizure onset zone.
Purpose of the Study:
- To develop and evaluate a novel symmetricity-driven deep learning framework for identifying epileptic foci in pediatric patients with TLE using 18F-FDG PET imaging.
- To compare the performance of the deep learning framework against visual assessment, statistical parametric mapping (SPM), and Jensen-Shannon divergence-based logistic regression (JS-LR).
Main Methods:
- A retrospective study included 201 pediatric TLE patients and 24 age-matched controls who underwent 18F-FDG PET-CT.
- A pair-of-cube (PoC)-based Siamese convolutional neural network (CNN) was developed to analyze 386 symmetricity features from PET images for precise focus localization.
- The asymmetric index (AI) was used to automatically calculate the metabolic abnormality level of the predicted focus.
Main Results:
- The deep learning framework achieved a significantly higher Dice coefficient (0.51) for epileptic foci detection compared to SPM (0.24) and visual assessment (0.31-0.44).
- The PoC classification demonstrated a higher area under the curve (AUC) (0.93) than JS-LR (0.72).
- Metabolic level detection accuracy was significantly higher with the proposed method (90%) compared to blinded or unblinded visual assessment (56%-68%).
Conclusions:
- The proposed deep learning framework accurately and efficiently identifies epileptic foci in pediatric TLE patients using 18F-FDG PET imaging.
- This framework shows potential as a computer-assisted diagnostic tool for epilepsy.
- Further application of this AI-driven approach could enhance the diagnostic process for epilepsy patients.
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.
