Dataset Information

Automated detection of cortical dysplasia type II in MRI-negative epilepsy.

ABSTRACT:

Objective

To detect automatically focal cortical dysplasia (FCD) type II in patients with extratemporal epilepsy initially diagnosed as MRI-negative on routine inspection of 1.5 and 3.0T scans.

Methods

We implemented an automated classifier relying on surface-based features of FCD morphology and intensity, taking advantage of their covariance. The method was tested on 19 patients (15 with histologically confirmed FCD) scanned at 3.0T, and cross-validated using a leave-one-out strategy. We assessed specificity in 24 healthy controls and 11 disease controls with temporal lobe epilepsy. Cross-dataset classification performance was evaluated in 20 healthy controls and 14 patients with histologically verified FCD examined at 1.5T.

Results

Sensitivity was 74%, with 100% specificity (i.e., no lesions detected in healthy or disease controls). In 50% of cases, a single cluster colocalized with the FCD lesion, while in the remaining cases a median of 1 extralesional cluster was found. Applying the classifier (trained on 3.0T data) to the 1.5T dataset yielded comparable performance (sensitivity 71%, specificity 95%).

Conclusion

In patients initially diagnosed as MRI-negative, our fully automated multivariate approach offered a substantial gain in sensitivity over standard radiologic assessment. The proposed method showed generalizability across cohorts, scanners, and field strengths. Machine learning may assist presurgical decision-making by facilitating hypothesis formulation about the epileptogenic zone.

Classification of evidence

This study provides Class II evidence that automated machine learning of MRI patterns accurately identifies FCD among patients with extratemporal epilepsy initially diagnosed as MRI-negative.

SUBMITTER: Hong SJ

PROVIDER: S-EPMC4114179 | biostudies-literature | 2014 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated detection of cortical dysplasia type II in MRI-negative epilepsy.

Hong Seok-Jun SJ Kim Hosung H Schrader Dewi D Bernasconi Neda N Bernhardt Boris C BC Bernasconi Andrea A

Neurology 20140604 1

<h4>Objective</h4>To detect automatically focal cortical dysplasia (FCD) type II in patients with extratemporal epilepsy initially diagnosed as MRI-negative on routine inspection of 1.5 and 3.0T scans.<h4>Methods</h4>We implemented an automated classifier relying on surface-based features of FCD morphology and intensity, taking advantage of their covariance. The method was tested on 19 patients (15 with histologically confirmed FCD) scanned at 3.0T, and cross-validated using a leave-one-out stra ...[more]

PMID: 24898923

Similar Datasets

Project description:Purpose: The superior frontal sulcus (SFS), located in the prefrontal and premotor cortex, is considered as one of the common locations of focal cortical dysplasia (FCD). However, the characteristics of seizures arising from this area are incompletely known. The primary purpose of this study was to investigate the clinical features and the epileptic networks of seizures originating from the SFS. Methods: We included seventeen patients with type II FCD within the SFS. SFS was identified both visually and automatically. Semiological features were evaluated and grouped. Interictal 18FDG-PET imaging in all patients was compared to controls using statistical parametric mapping (SPM-PET). In those subjects with stereoelectroencephalography (SEEG), two different quantitative intracranial electroencephalography analyses were applied. Finally, the locations of the SFS-related hypometabolic regions and epileptogenic zones (EZs) were transformed into standard space for group analysis. Results: We identified two semiological groups. Group 1 (9/17) showed elementary motor signs (head version and tonic posturing), while group 2 (8/17) exhibited complex motor behavior (fear, hypermotor, and ictal pouting). Based on SPM-PET, an SFS-supplementary motor area (SMA) epileptic propagation network was found in group 1, and an SFS-middle cingulate cortex (MCC)-pregenual anterior cingulate cortex (pACC) propagation network was discovered in group 2. Intracranial EEG analysis suggested similar affected structures with high epileptogenicity. The SFS-related hypometabolic regions and EZs in these groups showed a posterior-anterior spatial relationship. Conclusions: Even though originating from the spatially restricted cortex, SFS seizures can be divided into two groups based on semiological features. The SFS-SMA and SFS-MCC-pACC epileptic propagation networks may play pivotal roles in the generation of different semiologies. The posterior-anterior spatial relationship of both hypometabolic regions and EZs provides potentially useful information for distinguishing different types of SFS seizures and surgical evaluation.

Project description:ObjectiveFocal cortical dysplasia (FCD) represents a heterogeneous group of disorders of the cortical formation and is one of the most common causes of epilepsy. Magnetic resonance imaging (MRI) is the modality of choice for detecting structural lesions, and the surgical prognosis in patients with MR lesions is favorable. However, the surgical prognosis of patients with MR-negative FCD is unknown. We aimed to evaluate the long-term surgical outcomes and prognostic factors in MR-negative FCD patients through comprehensive presurgical data.MethodsWe retrospectively reviewed data from 719 drug-resistant epilepsy patients who underwent resective surgery and selected cases in which surgical specimens were pathologically confirmed as FCD Type I or II. If the epileptogenic focus and surgical specimens were obtained from brain areas with a normal MRI appearance, they were classified as MR-negative FCD. Surgical outcomes were evaluated at 2 and 5 years, and clinical, neurophysiological, and neuroimaging data of MR-negative FCD were compared to those of MR-positive FCD.ResultsFinally, 47 MR-negative and 34 MR-positive FCD patients were enrolled in the study. The seizure-free rate after surgery (Engel classification I) at postoperative 2 year was 59.5% and 64.7% in the MR-negative and positive FCD groups, respectively (p = 0.81). This rate decreased to 57.5% and 44.4% in the MR-negative and positive FCD groups (p = 0.43) at postoperative 5 years. MR-negative FCD showed a higher proportion of FCD type I (87.2% vs. 50.0%, p = 0.001) than MR-positive FCD. Unilobar cerebral perfusion distribution (odds ratio, OR 5.41) and concordance of interictal epileptiform discharges (OR 5.10) were significantly associated with good surgical outcomes in MR-negative FCD.ConclusionIn this study, MR-negative and positive FCD patients had a comparable surgical prognosis, suggesting that comprehensive presurgical evaluations, including multimodal neuroimaging studies, are crucial for obtaining excellent surgical outcomes even in epilepsy patients with MR-negative FCD.

Dataset Information

Automated detection of cortical dysplasia type II in MRI-negative epilepsy.

Objective

Methods

Results

Conclusion

Classification of evidence

Publications

Automated detection of cortical dysplasia type II in MRI-negative epilepsy.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure