Related Experiment Video
Updated: Oct 6, 2025
Radiotracer Administration for High Temporal Resolution Positron Emission Tomography of the Human Brain: Application to FDG-fPET
Published on: October 22, 2019
Individual [18F]FDG PET and functional MRI based on simultaneous PET/MRI may predict seizure recurrence after
Jingjuan Wang1, Kun Guo1, Bixiao Cui1
1Department of Radiology and Nuclear Medicine, Xuanwu Hospital Capital Medical University, Beijing, 100053, China.
Objectives:
To investigate the individual measures of brain glucose metabolism, neural activity obtained from simultaneous 18[F]FDG PET/MRI, and their association with surgical outcomes in medial temporal lobe epilepsy due to hippocampal sclerosis (mTLE-HS).
Methods:
Thirty-nine unilateral mTLE-HS patients who underwent anterior temporal lobectomy were classified as having completely seizure-free (Engel class IA; n = 22) or non-seizure-free (Engel class IB-IV; n = 17) outcomes at 1 year after surgery. Preoperative [18F]FDG PET and functional MRI (fMRI) were obtained from a simultaneous PET/MRI scanner, and individual glucose metabolism and fractional amplitude of low-frequency fluctuation (fALFF) were evaluated by standardizing these with respect to healthy controls. These abnormality measures and clinical data from each patient were incorporated into a machine learning framework (gradient boosting decision tree and logistic regression analysis) to estimate seizure recurrence. The predictive values of features were evaluated by the receiver operating characteristic (ROC) curve in the training and test cohorts.
Results:
The machine learning classification model showed [18F]FDG PET and fMRI variations in contralateral hippocampal network and age of onset identify unfavorable surgical outcomes effectively. In the validation dataset, the logistic regression model with [18F]FDG PET and fALFF obtained from simultaneous [18F]FDG PET/MRI gained the maximum area under the ROC curve of 0.905 for seizure recurrence, higher than 0.762 with 18[F]-FDG PET, and 0.810 with fALFF alone.
Conclusion:
Machine learning model suggests individual [18F]FDG PET and fMRI variations in contralateral hippocampal network based on 18[F]-FDG PET/MRI could serve as a potential biomarker of unfavorable surgical outcomes.
Key Points:
• Individual [18F]FDG PET and fMRI obtained from preoperative [18F]FDG PET/MR were investigated. • Individual differences were further assessed based on a seizure propagation network. • Machine learning can classify surgical outcomes with 90.5% accuracy.
Insights
Preoperative brain imaging using [18F]FDG PET/MRI can predict surgical outcomes in epilepsy. Machine learning models accurately identify patients likely to experience continued seizures after surgery, aiding treatment decisions.
Area of Science:
- Neuroimaging
- Epilepsy Research
- Machine Learning in Medicine
Background:
- Medial temporal lobe epilepsy due to hippocampal sclerosis (mTLE-HS) is a common cause of drug-resistant epilepsy.
- Predicting surgical outcomes is crucial for patient management.
- Simultaneous [18F]FDG PET/MRI offers insights into both glucose metabolism and neural activity.
Purpose of the Study:
- To investigate brain glucose metabolism and neural activity using simultaneous [18F]FDG PET/MRI.
- To assess the association of these measures with surgical outcomes in mTLE-HS patients.
- To develop a machine learning model for predicting seizure recurrence post-surgery.
Main Methods:
- Thirty-nine mTLE-HS patients undergoing anterior temporal lobectomy were studied.
- Preoperative simultaneous [18F]FDG PET/MRI was used to assess individual glucose metabolism and fractional amplitude of low-frequency fluctuation (fALFF).
- A machine learning framework (gradient boosting and logistic regression) incorporated imaging data and clinical factors to predict seizure outcomes.
Main Results:
- Machine learning models effectively identified unfavorable surgical outcomes based on [18F]FDG PET and fMRI variations in the contralateral hippocampal network.
- A logistic regression model using [18F]FDG PET and fALFF achieved an area under the ROC curve of 0.905 for predicting seizure recurrence.
- This combined imaging approach outperformed predictions based on [18F]FDG PET or fALFF alone.
Conclusions:
- Individual [18F]FDG PET and fMRI variations, particularly within the contralateral hippocampal network, show potential as biomarkers for predicting unfavorable surgical outcomes in mTLE-HS.
- Simultaneous [18F]FDG PET/MRI combined with machine learning offers a powerful tool for outcome prediction.
- This approach can aid in patient selection and surgical planning for epilepsy surgery.
More Related Videos
12:09Network Analysis of the Default Mode Network Using Functional Connectivity MRI in Temporal Lobe Epilepsy
Published on: August 5, 2014
09:57Author Spotlight: Advancing Pediatric Epilepsy Surgery in Children Through Novel Biomarkers and Enhanced Localization
Published on: September 20, 2024
