Dataset Information

Deep Convolutional Neural Network Based Interictal-Preictal Electroencephalography Prediction: Application to Focal Cortical Dysplasia Type-II.

ABSTRACT: We aimed to differentiate between the interictal and preictal states in epilepsy patients with focal cortical dysplasia (FCD) type-II using deep learning-based classifiers based on intracranial electroencephalography (EEG). We also investigated the practical conditions for high interictal-preictal discriminability in terms of spatiotemporal EEG characteristics and data size efficiency. Intracranial EEG recordings of nine epilepsy patients with FCD type-II (four female, five male; mean age: 10.7 years) were analyzed. Seizure onset and channel ranking were annotated by two epileptologists. We performed three consecutive interictal-preictal classification steps by varying the preictal length, number of electrodes, and sampling frequency with convolutional neural networks (CNN) using 30 s time-frequency data matrices. Classification performances were evaluated based on accuracy, F1 score, precision, and recall with respect to the above-mentioned three parameters. We found that (1) a 5 min preictal length provided the best classification performance, showing a remarkable enhancement of >13% on average compared to that with the 120 min preictal length; (2) four electrodes provided considerably high classification performance with a decrease of only approximately 1% on average compared to that with all channels; and (3) there was minimal performance change when quadrupling the sampling frequency from 128 Hz. Patient-specific performance variations were noticeable with respect to the preictal length, and three patients showed above-average performance enhancements of >28%. However, performance enhancements were low with respect to both the number of electrodes and sampling frequencies, and some patients showed at most 1-2% performance change. CNN-based classifiers from intracranial EEG recordings using a small number of electrodes and efficient sampling frequency are feasible for predicting the interictal-preictal state transition preceding seizures in epilepsy patients with FCD type-II. Preictal lengths affect the predictability in a patient-specific manner; therefore, pre-examinations for optimal preictal length will be helpful in seizure prediction.

SUBMITTER: Chung YG

PROVIDER: S-EPMC7674929 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep Convolutional Neural Network Based Interictal-Preictal Electroencephalography Prediction: Application to Focal Cortical Dysplasia Type-II.

Chung Yoon Gi YG Jeon Yonghoon Y Choi Sun Ah SA Cho Anna A Kim Hunmin H Hwang Hee H Kim Ki Joong KJ

Frontiers in neurology 20201105

We aimed to differentiate between the interictal and preictal states in epilepsy patients with focal cortical dysplasia (FCD) type-II using deep learning-based classifiers based on intracranial electroencephalography (EEG). We also investigated the practical conditions for high interictal-preictal discriminability in terms of spatiotemporal EEG characteristics and data size efficiency. Intracranial EEG recordings of nine epilepsy patients with FCD type-II (four female, five male; mean age: 10.7 ...[more]

PMID: 33250854

Similar Datasets

Project description:PURPOSE:Focal cortical dysplasia (FCD) is a common cause of epilepsy; the only treatment is surgery. Therefore, detecting FCD using noninvasive imaging technology can help doctors determine whether surgical intervention is required. Since FCD lesions are small and not obvious, diagnosing FCD through visual evaluations of magnetic resonance imaging (MRI) scans is difficult. The purpose of this study is to detect and segment histologically confirmed FCD lesions in images of normal fluid-attenuated inversion recovery (FLAIR)-negative lesions using convolutional neural network (CNN) technology. METHODS:The technique involves training a six-layer CNN named Net-Pos, which consists of two convolutional layers (CLs); two pooling layers (PLs); and two fully connected (FC) layers, including 60 943 learning parameters. We employed activation maximization (AM) to optimize a series of pattern image blocks (PIBs) that were most similar to a lesion image block by using the trained Net-Pos. We developed an AM and convolutional localization (AMCL) algorithm that employs the mean PIBs combined with convolution to locate and segment FCD lesions in FLAIR-negative patients. Five evaluation indices, namely, recall, specificity, accuracy, precision, and the Dice coefficient, were applied to evaluate the localization and segmentation performance of the algorithm. RESULTS:The PIBs most similar to an FCD lesion image block were identified by the trained Net-Pos as image blocks with brighter central areas and darker surrounding image blocks. The technique was evaluated using 18 FLAIR-negative lesion images from 12 patients. The subject-wise recall of the AMCL algorithm was 83.33% (15/18). The Dice coefficient for the segmentation performance was 52.68. CONCLUSION:We developed a novel algorithm referred to as the AMCL algorithm with mean PIBs to effectively and automatically detect and segment FLAIR-negative FCD lesions. This work is the first study to apply a CNN-based model to detect and segment FCD lesions in images of FLAIR-negative lesions.

Project description:Magnetoencephalography (MEG) is a useful non-invasive technique for presurgical evaluation of focal cortical dysplasia patients. We aimed at clarifying the precise spatial relationship between the spiking volume determined with MEG, the seizure onset zone and the lesional volume in patients with focal cortical dysplasia. We studied the spatial relationships between the MEG spiking volume determined with a recent analysis pipeline, the seizure-onset zone location determined with a quantitative index calculated from intracranial EEG signals ('Epileptogenicity Index') and the lesional volume delineated on brain MRI in 11 patients with Focal Cortical Dysplasia explored with Stereo-electroencephalography (SEEG). A significant correlation between the MEG spiking activity and the Epileptogenicity Index was found in 8/11 patients. 7/8 patients were operated upon and had good surgical outcome. For three patients, no correlation between Epileptogenicity Index and spiking activity was observed; only one of those three patients had good surgical outcome. The lesion was at least partially overlapping with the seizure-onset zone in 8/9 patients with a lesion clearly identifiable by MRI. However, 57% of the SEEG epileptogenic contacts were located outside of the lesional volume. Lastly 44% of the highly epileptogenic SEEG contacts were located within the spiking volume and 22% of them were located exclusively in the spiking volume and not in the lesion. For 7/9 patients with a lesion, < 50% of epileptogenic SEEG contacts were included within the lesion: for 5/7 patients MEG provided an added value for targeting the epileptogenic region through intracranial electrodes, while for two of seven patients MEG detected only a few extralesional epileptogenic contacts. Our study suggests that modeling of the spiking volume with MEG is a promising tool to localize non-invasively the seizure-onset zone in patients with focal cortical dysplasia. Combined with brain MRI, MEG modeling of the spiking volume contributes to delineate the spatial extent of the seizure-onset zone.

Project description:Background and objectiveTo test the hypothesis that a multicenter-validated computer deep learning algorithm detects MRI-negative focal cortical dysplasia (FCD).MethodsWe used clinically acquired 3-dimensional (3D) T1-weighted and 3D fluid-attenuated inversion recovery MRI of 148 patients (median age 23 years [range 2-55 years]; 47% female) with histologically verified FCD at 9 centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed MRI-negative in 51% of patients, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 cases with FCD (13 ± 10 years). Applying the algorithm to 42 healthy controls and 89 controls with temporal lobe epilepsy disease tested specificity.ResultsOverall sensitivity was 93% (137 of 148 FCD detected) using a leave-one-site-out cross-validation, with an average of 6 false positives per patient. Sensitivity in MRI-negative FCD was 85%. In 73% of patients, the FCD was among the clusters with the highest confidence; in half, it ranked the highest. Sensitivity in the independent cohort was 83% (19 of 23; average of 5 false positives per patient). Specificity was 89% in healthy and disease controls.DiscussionThis first multicenter-validated deep learning detection algorithm yields the highest sensitivity to date in MRI-negative FCD. By pairing predictions with risk stratification, this classifier may assist clinicians in adjusting hypotheses relative to other tests, increasing diagnostic confidence. Moreover, generalizability across age and MRI hardware makes this approach ideal for presurgical evaluation of MRI-negative epilepsy.Classification of evidenceThis study provides Class III evidence that deep learning on multimodal MRI accurately identifies FCD in patients with epilepsy initially diagnosed as MRI negative.

Project description:Focal cortical dysplasias are a common subtype of malformation of cortical development, which frequently presents with a spectrum of cognitive and behavioural abnormalities as well as pharmacoresistant epilepsy. Focal cortical dysplasia type II is typically caused by somatic mutations resulting in mammalian target of rapamycin (mTOR) hyperactivity, and is the commonest pathology found in children undergoing epilepsy surgery. However, surgical resection does not always result in seizure freedom, and is often precluded by proximity to eloquent brain regions. Gene therapy is a promising potential alternative treatment and may be appropriate in cases that represent an unacceptable surgical risk. Here, we evaluated a gene therapy based on overexpression of the Kv1.1 potassium channel in a mouse model of frontal lobe focal cortical dysplasia. An engineered potassium channel (EKC) transgene was placed under control of a human promoter that biases expression towards principal neurons (CAMK2A) and packaged in an adeno-associated viral vector (AAV9). We used an established focal cortical dysplasia model generated by in utero electroporation of frontal lobe neural progenitors with a constitutively active human Ras homolog enriched in brain (RHEB) plasmid, an activator of mTOR complex 1. We characterized the model by quantifying electrocorticographic and behavioural abnormalities, both in mice developing spontaneous generalized seizures and in mice only exhibiting interictal discharges. Injection of AAV9-CAMK2A-EKC in the dysplastic region resulted in a robust decrease (∼64%) in the frequency of seizures. Despite the robust anti-epileptic effect of the treatment, there was neither an improvement nor a worsening of performance in behavioural tests sensitive to frontal lobe function. AAV9-CAMK2A-EKC had no effect on interictal discharges or behaviour in mice without generalized seizures. AAV9-CAMK2A-EKC gene therapy is a promising therapy with translational potential to treat the epileptic phenotype of mTOR-related malformations of cortical development. Cognitive and behavioural co-morbidities may, however, resist an intervention aimed at reducing circuit excitability.

Dataset Information

Deep Convolutional Neural Network Based Interictal-Preictal Electroencephalography Prediction: Application to Focal Cortical Dysplasia Type-II.

Publications

Deep Convolutional Neural Network Based Interictal-Preictal Electroencephalography Prediction: Application to Focal Cortical Dysplasia Type-II.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets