Neuroimaging of focal malformations of cortical development

Development and prospective clinical validation of a convolutional neural network for automated detection and segmentation of focal cortical dysplasias

PURPOSE

Focal cortical dysplasias (FCDs) are a leading cause of drug-resistant epilepsy. Early detection and resection of FCDs have favorable prognostic implications for postoperative seizure freedom. Despite advancements in imaging methods, FCD detection remains challenging. House et al. (2021) introduced a convolutional neural network (CNN) for automated FCD detection and segmentation, achieving a sensitivity of 77.8%. However, its clinical applicability was limited due to a low specificity of 5.5%. The objective of this study was to improve the CNN's performance through data-driven training and algorithm optimization, followed by a prospective validation on daily-routine MRIs.

MATERIAL AND METHODS

A dataset of 300 3 T MRIs from daily clinical practice, including 3D T1 and FLAIR sequences, was prospectively compiled. The MRIs were visually evaluated by two neuroradiologists and underwent morphometric assessment by two epileptologists. The dataset included 30 FCD cases (11 female, mean age: 28.1 ± 10.1 years) and a control group of 150 normal cases (97 female, mean age: 32.8 ± 14.9 years), along with 120 non-FCD pathological cases (64 female, mean age: 38.4 ± 18.4 years). The dataset was divided into three subsets, each analyzed by the CNN. Subsequently, the CNN underwent a two-phase-training process, incorporating subset MRIs and expert-labeled FCD maps. This training employed both classical and continual learning techniques. The CNN's performance was validated by comparing the baseline model with the trained models at two training levels.

RESULTS

In prospective validation, the best model trained using continual learning achieved a sensitivity of 90.0%, specificity of 70.0%, and accuracy of 72.0%, with an average of 0.41 false positive clusters detected per MRI. For FCD segmentation, an average Dice coefficient of 0.56 was attained. The model's performance improved in each training phase while maintaining a high level of sensitivity. Continual learning outperformed classical learning in this regard.

CONCLUSIONS

Our study presents a promising CNN for FCD detection and segmentation, exhibiting both high sensitivity and specificity. Furthermore, the model demonstrates continuous improvement with the inclusion of more clinical MRI data. We consider our CNN a valuable tool for automated, examiner-independent FCD detection in daily clinical practice, potentially addressing the underutilization of epilepsy surgery in drug-resistant focal epilepsy and thereby improving patient outcomes.

Detecting Cortical Thickness Changes in Epileptogenic Lesions Using Machine Learning

Epilepsy is a neurological disorder characterized by abnormal brain activity. Epileptic patients suffer from unpredictable seizures, which may cause a loss of awareness. Seizures are considered drug resistant if treatment does not affect success. This leads practitioners to calculate the cortical thickness to measure the distance between the brain’s white and grey matter surfaces at various locations to perform a surgical intervention. In this study, we introduce using machine learning as an approach to classify extracted measurements from T1-weighted magnetic resonance imaging. Data were collected from the epilepsy unit at King Abdulaziz University Hospital. We applied two trials to classify the extracted measurements from T1-weighted MRI for drug-resistant epilepsy and healthy control subjects. The preprocessing sequence on T1-weighted MRI images was performed using C++ through BrainSuite’s pipeline. The first trial was performed on seven different combinations of four commonly selected measurements. The best performance was achieved in Exp6 and Exp7, with 80.00% accuracy, 83.00% recall score, and 83.88% precision. It is noticeable that grey matter volume and white matter volume measurements are more significant than the cortical thickness measurement. The second trial applied four different machine learning classifiers after applying 10-fold cross-validation and principal component analysis on all extracted measurements as in the first trial based on the mentioned previous works. The K-nearest neighbours model outperformed the other machine learning classifiers with 97.11% accuracy, 75.00% recall score, and 75.00% precision.

Automated detection and segmentation of focal cortical dysplasias (FCDs) with artificial intelligence: Presentation of a novel convolutional neural network and its prospective clinical validation

PURPOSE

Focal cortical dysplasias (FCDs) represent one of the most frequent causes of pharmaco-resistant focal epilepsies. Despite improved clinical imaging methods over the past years, FCD detection remains challenging, as FCDs vary in location, size, and shape and commonly blend into surrounding tissues without clear definable boundaries. We developed a novel convolutional neural network for FCD detection and segmentation and validated it prospectively on daily-routine MRIs.

MATERIAL AND METHODS

The neural network was trained on 201 T1 and FLAIR 3 T MRI volume sequences of 158 patients with mainly FCDs, regardless of type, and 7 focal PMG. Non-FCD/PMG MRIs, drawn from 100 normal MRIs and 50 MRIs with non-FCD/PMG pathologies, were added to the training. We applied the algorithm prospectively on 100 consecutive MRIs of patients with focal epilepsy from daily clinical practice. The results were compared with corresponding neuroradiological reports and morphometric MRI analyses evaluated by an experienced epileptologist.

RESULTS

Best training results reached a sensitivity (recall) of 70.1 % and a precision of 54.3 % for detecting FCDs. Applied on the daily-routine MRIs, 7 out of 9 FCDs were detected and segmented correctly with a sensitivity of 77.8 % and a specificity of 5.5 %. The results of conventional visual analyses were 33.3 % and 94.5 %, respectively (3/9 FCDs detected); the results of morphometric analyses with overall epileptologic evaluation were both 100 % (9/9 FCDs detected) and thus served as reference.

CONCLUSION

We developed a 3D convolutional neural network with autoencoder regularization for FCD detection and segmentation. Our algorithm employs the largest FCD training dataset to date with various types of FCDs and some focal PMG. It provided a higher sensitivity in detecting FCDs than conventional visual analyses. Despite its low specificity, the number of false positively predicted lesions per MRI was lower than with morphometric analysis. We consider our algorithm already useful for FCD pre-screening in everyday clinical practice.

Normal and Disordered Formation of the Cerebral Cortex : Normal Embryology, Related Molecules, Types of Migration, Migration Disorders

The expansion and folding of the cerebral cortex occur during brain development and are critical factors that influence cognitive ability and sensorimotor skills. The disruption of cortical growth and folding may cause neurological disorders, resulting in severe intellectual disability and intractable epilepsy in humans. Therefore, understanding the mechanism that regulates cortical growth and folding will be crucial in deciphering the key steps of brain development and finding new therapeutic targets for the congenital anomalies of the cerebral cortex. This review will start with a brief introduction describing the anatomy of the brain cortex, followed by a description of our understanding of the proliferation, differentiation, and migration of neural progenitors and important genes and molecules that are involved in these processes. Finally, various types of disorders that develop due to malformation of the cerebral cortex will be discussed.

Detection of structural abnormalities of cortical and subcortical gray matter in patients with MRI-negative refractory epilepsy using neurite orientation dispersion and density imaging

PURPOSE

NODDI (Neurite Orientation Dispersion and Density Imaging) and DTI (Diffusion tensor imaging) may be useful in identifying abnormal regions in patients with MRI-negative refractory epilepsy. The aim of this study was to determine whether NODDI and DTI maps including neurite density (ND), orientation dispersion index (ODI), mean diffusivity (MD) and fractional anisotropy (FA) can detect structural abnormalities in cortical and subcortical gray matter (GM) in these patients. The correlation between these parameters and clinical characteristics of the disease was also investigated.

METHODS

NODDI and DTI maps of 17 patients were obtained and checked visually. Region of interest (ROI) was drawn on suspected areas and contralateral regions in cortex. Contrast-to-noise ratio (CNR) was determined for each region. Furthermore volumetric data and mean values of ND, ODI, FA and MD of subcortical GM structures were calculated in both of the patients and controls. Finally, the correlations of these parameters in the subcortical with age of onset and duration of epilepsy were investigated.

RESULTS

Cortical abnormalities on ODI images were observed in eight patients qualitatively. CNR of ODI was significantly greater than FA and MD. The subcortical changes including decrease of FA and ND and increase of ODI in left nucleus accumbens and increase of the volume in right amygdala were detected in the patients.

CONCLUSIONS

The results revealed that NODDI can improve detection of microstructural changes in cortical and subcortical GM in patients with MRI negative epilepsy.

Whole Transcriptome Screening Reveals Myelination Deficits in Dysplastic Human Temporal Neocortex

Focal cortical dysplasias (FCDs) are local malformations of the human neocortex with strong epileptogenic potential. To investigate the underlying pathomechanisms, we performed a whole human transcriptome screening to compare the gene expression pattern of dysplastic versus nondysplastic temporal neocortex. Tissue obtained from FCD IIIa cases (mean age 20.5 years) who had undergone surgical treatment, due to intractable epilepsy, was compared with nondysplastic specimens (mean age 19.9 years) by means of Affymetrix arrays covering 28 869 genes. We found 211 differentially expressed genes (DEX) among which mainly genes important for oligodendrocyte differentiation and myelination were downregulated in FCD IIIa. These findings were confirmed as functionally important by Database for Annotation, Visualization, and Integrated Discovery (DAVID) analysis. The reduced expression of myelin-associated transcripts was confirmed for FCD Ia, IIa, and IIIa by real-time RT-qPCR. In addition, we found that the density of myelin basic protein mRNA-expressing oligodendrocytes and of 2',3'-cyclic nucleotide 3'-phosphodiesterase-positive myelin fibers was significantly reduced in dysplastic cortex. Moreover, high-resolution confocal imaging and 3D reconstruction revealed that the myelin fiber network was severely disorganized in dysplastic neocortex, indicating a disturbance of myelin sheath formation and maintenance in FCD.

The use of hippocampal volumetric measurements to improve diagnostic accuracy in pediatric patients with mesial temporal sclerosis

OBJECTIVE Many patients with medically intractable epilepsy have mesial temporal sclerosis (MTS), which significantly affects their quality of life. The surgical excision of MTS lesions can result in marked improvement or even complete resolution of the epileptic episodes. Reliable radiological diagnosis of MTS is a clinical challenge. The purpose of this study was to evaluate the utility of volumetric mapping of the hippocampi for the identification of MTS in a case-controlled series of pediatric patients who underwent resection for medically refractory epilepsy, using pathology as a gold standard. METHODS A cohort of 57 pediatric patients who underwent resection for medically intractable epilepsy between 2005 and 2015 was evaluated. On pathological investigation, this group included 24 patients with MTS and 33 patients with non-MTS findings. Retrospective quantitative volumetric measurements of the hippocampi were acquired for 37 of these 57 patients. Two neuroradiologists with more than 10 years of experience who were blinded to the patients' MTS status performed the retrospective review of MR images. To produce the volumetric data, MR scans were parcellated and segmented using the FreeSurfer software suite. Hippocampal regions of interest were compared against an age-weighted local regression curve generated with data from the pediatric normal cohort. Standard deviations and percentiles of specific subjects were calculated. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were determined for the original clinical read and the expert readers. Receiver operating characteristic curves were generated for the methods of classification to compare results from the readers with the authors' results, and an optimal threshold was determined. From that threshold the sensitivity, specificity, PPV, and NPV were calculated for the volumetric analysis. RESULTS With the use of quantitative volumetry, a sensitivity of 72%, a specificity of 95%, a PPV of 93%, an NPV of 78%, and an area under the curve of 0.84 were obtained using a percentage difference of normalized hippocampal volume. The resulting specificity (95%) and PPV (93%) are superior to the original clinical read and to Reader A and Reader B's findings (range for specificity 74%-86% and for PPV 64%-71%). The sensitivity (72%) and NPV (78%) are comparable to Reader A's findings (73% and 81%, respectively) and are better than those of the original clinical read and of Reader B (sensitivity 45% and 63% and NPV 71% and 70%, respectively). CONCLUSIONS Volumetric measurement of the hippocampi outperforms expert readers in specificity and PPV, and it demonstrates comparable to superior sensitivity and NPV. Volumetric measurements can complement anatomical imaging for the identification of MTS, much like a computer-aided detection tool would. The implementation of this approach in the daily clinical workflow could significantly improve diagnostic accuracy.

Detection of Lesions Underlying Intractable Epilepsy on T1-Weighted MRI as an Outlier Detection Problem

Pattern recognition methods, such as computer aided diagnosis (CAD) systems, can help clinicians in their diagnosis by marking abnormal regions in an image. We propose a machine learning system based on a one-class support vector machine (OC-SVM) classifier for the detection of abnormalities in magnetic resonance images (MRI) applied to patients with intractable epilepsy. The system learns the features associated with healthy control subjects, allowing a voxelwise assessment of the deviation of a test subject pattern from the learned patterns. While any number of various features can be chosen and learned, here we focus on two texture parameters capturing image patterns associated with epileptogenic lesions on T1-weighted brain MRI e.g. heterotopia and blurred junction between the grey and white matter. The CAD output consists of patient specific 3D maps locating clusters of suspicious voxels ranked by size and degree of deviation from control patterns. System performance was evaluated using realistic simulations of challenging detection tasks as well as clinical data of 77 healthy control subjects and of eleven patients (13 lesions). It was compared to that of a mass univariate statistical parametric mapping (SPM) single subject analysis based on the same set of features. For all simulations, OC-SVM yielded significantly higher values of the area under the ROC curve (AUC) and higher sensitivity at low false positive rate. For the clinical data, both OC-SVM and SPM successfully detected 100% of the lesions in the MRI positive cases (3/13). For the MRI negative cases (10/13), OC-SVM detected 7/10 lesions and SPM analysis detected 5/10 lesions. In all experiments, OC-SVM produced fewer false positive detections than SPM. OC-SVM may be a versatile system for unbiased lesion detection.

Non-invasive treatment options for focal cortical dysplasia

Focal cortical dysplasia (FCD) presents a strong clinical challenge especially for the treatment of the associated epilepsy. Epilepsy in FCD is often treatment-resistant and constitutes 50% of treatment-resistant cases. Antiepileptic drugs (AEDs) have been widely used in the treatment of FCD. However, evidence to suggest their specific effect on the treatment of FCD remains to be established. In view of this resistance, several alternative treatments have been suggested. Although treatment currently involves surgical management, non-invasive treatments have been identified. The aim of the present review, was to assess non-invasive management strategies including, i) mammalian target of rapamycin (mTOR) inhibitors, ii) ketogenic diet (KD), and iii) vagus nerve stimulation (VNS). In addition, we discussed the literature available regarding the use of AEDs in FCD. Experiments conducted with mammals detailing rapamycin gene mutations in FCD have produced vital information for exploring treatment options using mTOR inhibitors. Of note is the importance of KD in children with FCD. This diet has been shown to modify disease progression by attenuating chromatin modification, a master regulator for gene expression and functional adaptation of the cell. FCD has also been studied widely with neurostimulation techniques. The outcomes of these techniques have been found to be variable. For widespread dysplasias, VNS has been shown to produce responder rates of >50%. Nevertheless, non-invasive cranial nerve stimulation techniques such as transcutaneous VNS and non-invasive VNS are gaining better patient compatibility, albeit their efficacy remains to be established.

Diffusion Tensor Imaging Abnormalities in Focal Cortical Dysplasia

Purpose:

Focal cortical dysplasia (FCD) is one of the most common underlying pathologic substrates in patients with medically intractable epilepsy. While magnetic resonance imaging (MRI) evidence of FCD is an important predictor of good surgical outcome, conventional MRI is not sensitive enough to detect all lesions. Previous reports of diffusion tensor imaging (DTI) abnormalities in FCD suggest the potential of DTI in the detection of FCD. The purpose of this study was to study subcortical white matter underlying small lesions of FCD using DTI.

Methods:

Five patients with medically intractable epilepsy and FCD were investigated. Diffusion tensor imaging images were acquired (20 contiguous 3mm thick axial slices) with maps of fractional anisotropy (FA), trace apparent diffusion coefficient (trace/3 ADC), and principal eigenvalues (ADC parallel and ADC perpendicular to white matter tracts) being calculated for each slice. Region of interest analysis was used to compare subcortical white matter ipsilateral and contralateral to the lesion.

Results:

Three subjects with FCD associated with underlying white matter hyperintensities on T2 weighted MRI were observed to have increased trace/3 ADC, reduced fractional anisotropy and increased perpendicular water diffusivity which was greater than the relative increase in the parallel diffusivity. No DTI abnormalities were identified in two patients with FCD without white matter hyperintensities on conventional T2-weighted MRI.

Conclusions:

While DTI abnormalities in FCD with obvious white matter involvement are consistent with micro-structural degradation of the underlying subcortical white matter, DTI changes were not identified in FCD lesions with normal appearing white matter.

Emerging surgical strategies of intractable frontal lobe epilepsy with cortical dysplasia in terms of extent of resection

Objective

Cortical dysplasia (CD) is one of the common causes of epilepsy surgery. However, surgical outcome still remains poor, especially with frontal lobe epilepsy (FLE), despite the advancement of neuroimaging techniques and expansion of surgical indications. The aim of this study was to focus on surgical strategies in terms of extent of resection to improve surgical outcome in the cases of FLE with CD.

Methods

A total of 11 patients of FLE were selected among 67 patients who were proven pathologically as CD, out of a total of 726 epilepsy surgery series since 1992. This study categorized surgical groups into three according to the extent of resection : 1) focal corticectomy, 2) regional corticectomy, and 3) partial functional lobectomy, based on the preoperative evaluation, in particular, ictal scalp EEG onset and/or intracranial recordings, and the lesions in high-resolution MRI. Surgical outcome was assessed following Engel's classification system.

Results

Focal corticectomy was performed in 5 patients and regional corticectomy in another set of 5 patients. Only 1 patient underwent partial functional lobectomy. Types I and II CD were detected with the same frequency (45.45% each) and postoperative outcome was fully satisfactory (91%).

Conclusion

The strategy of epilepsy surgery is to focus on the different characteristics of each individual, considering the extent of real resection, which is based on the focal ictal onset consistent with neuroimaging, especially in the practical point of view of neurosurgery.

Advanced diffusion imaging sequences could aid assessing patients with focal cortical dysplasia and epilepsy

•

Malformations of cortical development are a common cause of refractory epilepsy.

•

They are often invisible on structural imaging and only detected following surgery.

•

We assess a novel diffusion imaging technique (NODDI) in patients with dysplasia.

•

This shows more conspicuous changes than other clinical or diffusion scans.

•

This technique may assist the identification of FCD in patients with epilepsy.

Malformations of cortical development (MCD), particularly focal cortical dysplasia (FCD), are a common cause of refractory epilepsy but are often invisible on structural imaging. NODDI (neurite orientation dispersion and density imaging) is an advanced diffusion imaging technique that provides additional information on tissue microstructure, including intracellular volume fraction (ICVF), a marker of neurite density.

We applied this technique in 5 patients with suspected dysplasia to show that the additional parameters are compatible with the underlying disrupted tissue microstructure and could assist in the identification of the affected area.

The consistent finding was reduced ICVF in the area of dysplasia. In one patient, an area of reduced ICVF and increased fibre dispersion was identified that was not originally seen on the structural imaging. The focal reduction in ICVF on imaging is compatible with previous iontophoretic data in surgical specimens, was more conspicuous than on other clinical or diffusion images (supported by an increased contrast-to-noise ratio) and more localised than on previous DTI studies.

NODDI may therefore assist the clinical identification and localisation of FCD in patients with epilepsy. Future studies will assess this technique in a larger cohort including MRI negative patients.

Focal cortical dysplasia - review

Focal cortical dysplasia is a malformation of cortical development, which is the most common cause of medically refractory epilepsy in the pediatric population and the second/third most common etiology of medically intractable seizures in adults.Both genetic and acquired factors are involved in the pathogenesis of cortical dysplasia. Numerous classifications of the complex structural abnormalities of focal cortical dysplasia have been proposed - from Taylor et al. in 1971 to the last modification of Palmini classification made by Blumcke in 2011. In general, three types of cortical dysplasia are recognized.Type I focal cortical dysplasia with mild symptomatic expression and late onset, is more often seen in adults, with changes present in the temporal lobe.Clinical symptoms are more severe in type II of cortical dysplasia usually seen in children. In this type, more extensive changes occur outside the temporal lobe with predilection for the frontal lobes.New type III is one of the above dysplasias with associated another principal lesion as hippocampal sclerosis, tumor, vascular malformation or acquired pathology during early life.Brain MRI imaging shows abnormalities in the majority of type II dysplasias and in only some of type I cortical dysplasias.THE MOST COMMON FINDINGS ON MRI IMAGING INCLUDE: focal cortical thickening or thinning, areas of focal brain atrophy, blurring of the gray-white junction, increased signal on T2- and FLAIR-weighted images in the gray and subcortical white matter often tapering toward the ventricle. On the basis of the MRI findings, it is possible to differentiate between type I and type II cortical dysplasia. A complete resection of the epileptogenic zone is required for seizure-free life. MRI imaging is very helpful to identify those patients who are likely to benefit from surgical treatment in a group of patients with drug-resistant epilepsy.However, in type I cortical dysplasia, MR imaging is often normal, and also in both types the lesion seen on MRI may be smaller than the seizure-generating region seen in the EEG. The abnormalities may also involve vital for life brain parts, where curative surgery will not be an option. Therefore, other diagnostic imaging techniques such as FDG PET, MEG, DTI and intra-cranial EEG are widely used to establish the diagnosis and to decide on management.With advances in both genetics and neuroimaging, we may develop a better understanding of patients with drug-resistant epilepsy, which will help us to provide more successful pharmacological and/or surgical treatment in the future.

Malformations of cortical development and epilepsy in adult patients

OBJECTIVE

To describe clinical features of epilepsy secondary to Malformation of Cortical Development (MCD) in a series of adult patients.

MATERIALS AND METHODS

We searched our database for all cases with confirmed epilepsy and MCD and included in the study only those with complete data. Mean age, sex, age at seizure onset (ASO), seizure types, abnormal neurological exam (ANE), mental retardation, family history, gestational or perinatal insults (G-PI), interictal EEG and response to treatment were analyzed. Cases were classified into the 3 main groups (G) according to the Barkovich classification (BC) and then compared: (G1) "malformations due to abnormal cell proliferation", (G2) "malformations due to abnormal migration" and (G3) "malformations due to abnormal cortical organization".

RESULTS

We identified 152 (5.06%) patients with MCD from a total of 3000 with epilepsy. In total, 138 patients with complete medical data were included in this study. The mean age of patients was 36.2 years, 52.2% were female, the mean ASO was 12.3 years, 5.1% of cases had a positive family history and 21% had G-PI. An ANE was observed in 21% and mental retardation in 31.9%. Most of the patients (84.8%) had refractory epilepsy. The distribution of cases according to the BC was: 51.4% in G1, 28.9% in G2 and 19.6% in G3. Comparing the 3 groups, we found that an ANE was statistically more frequent in G3 and was present in 70.4% of cases.

CONCLUSION

Our series of adult patients with epilepsy and MCD suggests that MCD are identified as commonly in a developing country as in previous "first world" series. Neurological deficits were more common in the subgroup of patients with polymicrogyria and schizencephaly (BC Group 3).

Noninvasive approach to focal cortical dysplasias: clinical, EEG, and neuroimaging features

Purpose. The main purpose is to define more accurately the epileptogenic zone (EZ) with noninvasive methods in those patients with MRI diagnosis of focal cortical dysplasia (FCD) and epilepsy who are candidates of epilepsy surgery. Methods. Twenty patients were evaluated prospectively between 2007 and 2010 with comprehensive clinical evaluation, video-electroencephalography, diffusion tensor imaging (DTI), and high-resolution EEG to localize the equivalent current dipole (ECD). Key Findings. In 11 cases with white matter asymmetries in DTI the ECDs were located next to lesion on MRI with mean distance of 14.63 millimeters with topographical correlation with the EZ. Significance. We could establish a hypothesis of EZ based on Video-EEG, high-resolution EEG, ECD method, MRI, and DTI. These results are consistent with the hypothesis that the EZ in the FCD is complex and is often larger than visible lesion in MRI.

Detection of epileptogenic cortical malformations with surface-based MRI morphometry

Magnetic resonance imaging has revolutionized the detection of structural abnormalities in patients with epilepsy. However, many focal abnormalities remain undetected in routine visual inspection. Here we use an automated, surface-based method for quantifying morphometric features related to epileptogenic cortical malformations to detect abnormal cortical thickness and blurred gray-white matter boundaries. Using MRI morphometry at 3T with surface-based spherical averaging techniques that precisely align anatomical structures between individual brains, we compared single patients with known lesions to a large normal control group to detect clusters of abnormal cortical thickness, gray-white matter contrast, local gyrification, sulcal depth, jacobian distance and curvature. To assess the effects of threshold and smoothing on detection sensitivity and specificity, we systematically varied these parameters with different thresholds and smoothing levels. To test the effectiveness of the technique to detect lesions of epileptogenic character, we compared the detected structural abnormalities to expert-tracings, intracranial EEG, pathology and surgical outcome in a homogeneous patient sample. With optimal parameters and by combining thickness and GWC, the surface-based detection method identified 92% of cortical lesions (sensitivity) with few false positives (96% specificity), successfully discriminating patients from controls 94% of the time. The detected structural abnormalities were related to the seizure onset zones, abnormal histology and positive outcome in all surgical patients. However, the method failed to adequately describe lesion extent in most cases. Automated surface-based MRI morphometry, if used with optimized parameters, may be a valuable additional clinical tool to improve the detection of subtle or previously occult malformations and therefore could improve identification of patients with intractable focal epilepsy who may benefit from surgery.

Multimodal neuroimaging in presurgical evaluation of childhood epilepsy

In pre-surgical evaluation of pediatric epilepsy, the combined use of multiple imaging modalities for precise localization of the epileptogenic focus is a worthwhile endeavor. Advanced neuroimaging by high field Magnetic resonance imaging (MRI), diffusion tensor images, and MR spectroscopy have the potential to identify subtle lesions. ¹⁸F-FDG positron emission tomography and single photon emission tomography provide visualization of metabolic alterations of the brain in the ictal and interictal states. These techniques may have localizing value for patients which exhibit normal MRI scans. Functional MRI is helpful for non-invasively identifying areas of eloquent cortex. These advances are improving our ability to noninvasively detect epileptogenic foci which have gone undetected in the past and whose accurate localization is crucial for a favorable outcome following surgical resection.

Imaging findings of structural causes of epilepsy in children: a guide for the radiologist in the emergency room

Evaluation of the child with epilepsy is a relatively common indication for imaging in the emergency room setting. This room outlines some of the more important imaging features of causes of epilepsy in children.

Texture analysis: a review of neurologic MR imaging applications

Texture analysis describes a variety of image-analysis techniques that quantify the variation in surface intensity or patterns, including some that are imperceptible to the human visual system. Texture analysis may be particularly well-suited for lesion segmentation and characterization and for the longitudinal monitoring of disease or recovery. We begin this review by outlining the general procedure for performing texture analysis, identifying some potential pitfalls and strategies for avoiding them. We then provide an overview of some intriguing neuro-MR imaging applications of texture analysis, particularly in the characterization of brain tumors, prediction of seizures in epilepsy, and a host of applications to MS.

Small focal cortical dysplasia lesions are located at the bottom of a deep sulcus

Focal cortical dysplasia (FCD) is often characterized by minor structural changes that may go unrecognized by standard radiological analysis. Visual assessment of morphological characteristics of FCD and sulci harbouring them is difficult due to the complexity of brain convolutions. Our purpose was to elucidate and quantify the spatial relationship between FCD lesions and brain sulci using automated sulcal extraction and morphometry. We studied 43 consecutive FCD patients using high-resolution MRI. Lesions were classified into small and large using qualitative (detection on initial clinical assessment of conventional MRI) and quantitative (volume) criteria. Sulci were identified and labelled automatically using an algorithm based on a congregation of neural networks. Segmented FCD lesions and sulci were then simultaneously visualized in 3D. We measured mean and maximum depth of sulci related to each FCD and of the corresponding sulci in 21 healthy controls. In addition, we calculated sulcal depth within the FCD neighbourhood. Twenty-one (21/43 = 49%) patients had small FCD lesions (volume range: 128-3093 mm(3)). Among them, 17 (81%) had been overlooked during initial radiological evaluation and were subsequently identified using image processing. Eighteen (18/21 = 86%) small FCD lesions were located at the bottom of a sulcus. Two others were related to the walls of two sulci and one was located at the crown of a gyrus. Mean and maximum depth of sulci related to the FCD was higher than that of the corresponding sulci in controls (P < 0.008). Sulcal depth within lesional neighbourhood had larger mean depth than that of the entire sulcus (P < 0.0002). Evidence that small FCD lesions are preferentially located at the bottom of an abnormally deep sulcus may be used to direct the search for developmental abnormalities, particularly in patients in whom large-scale MRI features are only mildly abnormal or absent.

Long-term outcome of the ketogenic diet for intractable childhood epilepsy with focal malformation of cortical development

OBJECTIVE

We evaluated the efficacy and long-term outcome of the ketogenic diet in patients with intractable childhood epilepsy as a result of focal malformation of cortical development.

METHODS

A retrospective analysis evaluated seizure outcomes of 47 patients who had intractable epilepsy from (and) surgically remediable focal malformation of cortical development and were first treated with the classic ketogenic diet, involving the 4:1 lipid/nonlipid ratio. The long-term prognosis of 21 patients, who became seizure-free 3 months after the ketogenic diet, was followed up with that of 22 patients who eventually underwent epilepsy surgery.

RESULTS

Three months after diet initiation, 29 (61.7%) patients showed a reduction in seizure frequency of >50%, including 21 (44.7%) who became seizure-free. Of the 21 patients with complete seizure control at 3 months, 16 (76.2%) successfully completed the diet for 2 years without relapse, and 10 (47.6%) have remained seizure-free after cessation of the diet (mean follow-up for 3 years and 10 months), including 1 patient who remained seizure-free with additional medication after a relapse. Of the 22 patients who underwent epilepsy surgery, a seizure-free outcome was obtained for 13 (59.1%).

CONCLUSIONS

The ketogenic diet should be considered to be an additional option even in patients with focal malformation of cortical development, and long-term seizure-free outcome can be expected for patients who become seizure-free 3 months after the diet.

Malformations of cortical development and epilepsy

Malformations of cortical development (MCDs) are macroscopic or microscopic abnormalities of the cerebral cortex that arise as a consequence of an interruption to the normal steps of formation of the cortical plate. The human cortex develops its basic structure during the first two trimesters of pregnancy as a series of overlapping steps, beginning with proliferation and differentiation of neurons, which then migrate before finally organizing themselves in the developing cortex. Abnormalities at any of these stages, be they environmental or genetic in origin, may cause disruption of neuronal circuitry and predispose to a variety of clinical consequences, the most common of which is epileptic seizures, A large number of MCDs have now been described, each with characteristic pathological, clinical, and imaging features. The causes of many of these MCDs have been determined through the study of affected individuals, with many MCDs now established as being secondary to mutations in cortical development genes. This review will highlight the best-known of the human cortical malformations associated with epilepsy. The pathological, clinical, imaging, and etioiogic features of each MCD will be summarized, with representative magnetic resonance imaging (MRI) images shown for each MCD, The malformations tuberous sclerosis, focal cortical dysplasia, hemimegalencephaiy, classical iissencephaly, subcortical band heterotopia, periventricular nodular heterotopia, polymicrogyria, and schizencephaly will be presented.

Aberrant midsagittal fiber tracts in patients with hemimegalencephaly

BACKGROUND AND PURPOSE

In hemimegalencephaly, MR imaging often reveals midsagittal bandlike structures between the 2 lateral ventricles. To determine whether these structures are aberrant midsagittal fibers, we retrospectively reviewed them on conventional MR imaging and prospectively examined them by diffusion tensor MR and fiber tract (FT) reconstruction imaging.

MATERIALS AND METHODS

We retrospectively reviewed conventional MR images of 26 consecutive patients with hemimegalencephaly by 2 neuroradiologists, focusing on abnormal midsagittal structures. The distance between the 2 anterior horns and widths of midsagittal bandlike structures were measured. Prospective analysis was performed in 7 consecutive patients with hemimegalencephaly examined for midsagittal aberrant fibers by diffusion tensor imaging, and cortical distribution areas of the fibers were observed.

RESULTS

The distance between the 2 anterior horns was wide (>4 mm) due to white matter-intensity structures in 20 of 26 patients (76.9%). Mid-sagittal bandlike structures were observed in 15 patients (57.7%). Asymmetry of the fornices was detected in 7 patients (26.9%), and both fornices were thickened in 7 (26.9%) patients. On FT reconstruction, images showed that 4 of 7 patients with hemimegalencephaly had aberrant midsagittal fibers connecting frontal, occipital, or parietal lobes, bilaterally (n = 3) or ipsilaterally (n = 1). All 4 patients had increased width between the 2 anterior horns, and 3 of them exhibited midsagittal bandlike structures on conventional MR imaging. On the other hand, these MR imaging findings were not noted in 3 patients who did not have aberrant midsagittal fibers on diffusion tensor imaging.

CONCLUSIONS

Aberrant midsagittal FTs running intra- or interhemispherically do not infrequently exist in patients with hemimegalencephaly.

Evaluation of reflux nephropathy, pyelonephritis and renal dysplasia

MR urography has the potential to significantly improve our understanding of the relationship between reflux nephropathy, pyelonephritis, vesicoureteric reflux and renal dysplasia. MR urography utilizes multiple parameters to assess both renal anatomy and function and provides a more complete characterization of acquired and congenital disease. Pyelonephritis and renal scarring can be distinguished by assessing the parenchymal contours and signal intensity. Characteristic imaging features of renal dysplasia include small size, subcortical cysts, disorganized architecture, decreased and patchy contrast enhancement as well as a dysmorphic pelvicalyceal system. Because of its ability to subdivide and categorize this heterogeneous group of disorders, it seems inevitable that MR urography will replace DMSA renal scintigraphy as the gold standard for assessment of pyelonephritis and renal scarring. MR urography will contribute to our understanding of renal dysplasia and its relationship to reflux nephropathy.

Intralesional recordings and epileptogenic zone in focal polymicrogyria

PURPOSE

Polymicrogyria (PMG) is recognized as an epileptogenic lesion but few data concerning organization of the epileptogenic zone (EZ) are available.

METHODS

We analyzed the distribution of the EZ according to Stereo-EEG (SEEG) with intralesional recordings in four patients evaluated for intractable partial epilepsy associated with focal unilateral PMG, involving the posterior temporal region in two, the perisylvian area in one and the temporoparietal junction in the other. All had ictal scalp EEG, high-resolution structural and functional MRI, fluorodeoxyglucose positron emission tomography (FDG-PET), and SEEG. For each patient, several depth electrodes were implanted both within the PMG and in extralesional areas.

RESULTS

In three patients, the PMG displayed high-frequency spiking activity. However, interictal and ictal recordings demonstrated a large epileptogenic network, which was more widespread than the PMG, including the mesial temporal structures in two. In another patient, interictal spiking and seizure onset site were located within the hippocampus and outside of the PMG, although it was rapidly involved during seizure spread. Overall, EZ was considered to be larger than the PMG in all patients although hypometabolic areas detected by PET were concordant with EZ. Three patients underwent extensive surgery including the PMG and are seizure free with a follow-up >2 years.

DISCUSSION

Although intralesional recordings demonstrated intrinsic epileptogenicity in PMG, our data provide evidence that unilateral focal PMG belongs to a large epileptogenic network extending beyond the MRI lesion. SEEG may be helpful for planning surgery with favorable outcome, providing large resections are feasible, even in apparently focal PMG.

Clinical and Electroencephalographic Features of Patients With Polymicrogyria

Polymicrogyria is a malformation of cortical organization. The aim of this historic cohort study was to describe clinical and EEG features of patients with polymicrogyria. Patients underwent clinical and neurologic examination and a prolonged routine EEG to allow recording during sleep. Neuroimaging data were classified as: perisylvian polymicrogyria (subdivided into holosylvian, posterior parietal, and generalized), hemispheric polymicrogyria, and frontal polymicrogyria. Forty patients were studied: 16 with holosylvian polymicrogyria, 14 with posterior parietal polymicrogyria, 4 with generalized polymicrogyria, 3 with hemispheric polymicrogyria, and 3 with frontal polymicrogyria. Patients with polymicrogyria usually did not have epilepsy and their EEGs were mostly normal (55%); the severity of the clinical and EEG features correlated with the extent of the cortical lesion. In perisylvian polymicrogyria, epileptiform abnormalities predominated in fronto-temporal regions. Dour patients had focal electrical status (FES) in awakeness and electrical status epilepticus of sleep (ESES); these four patients had right hemispheric polymicrogyria and asymmetric bilateral perisylvian polymicrogyria, mostly on the right hemisphere. The authors conclude that the EEG is usually normal in patients with polymicrogyria, despite it being associated with FES and ESES in certain patients.

PET tracer technology for monitoring focal epilepsies

Studies using positron emission tomography (PET) have advanced our pathophysiological and biochemical understanding of focal and generalized epilepsies. H(2) (15)O PET allows quantification of cerebral blood flow and (18)F-fluorodeoxyglucose-PET quantification of cerebral glucose metabolism. Neurotransmitters are directly responsible for modulating synaptic activity and newer PET tracers can provide information about synaptic activity and specific ligand-receptor relationships, which are important for epileptogenesis and the spread of epileptic activity.

Metabolic brain imaging by magnetic resonance

Novel magnetic resonance methods have been developed to noninvasively measure biochemical compounds in the human brain as guided by magnetic resonance imaging. Together, these methods are referred to as magnetic resonance spectroscopy (MRS) and can be divided into three major categories: single voxel MRS, magnetic resonance spectroscopic imaging and dynamic MRS, which is a novel adaption of the first method. The techniques and range of biochemical compounds that can be measured safely and serially are advancing rapidly, with many technical developments. MRS methods, when applied to the human brain, have an important diagnostic role, help monitor and guide therapeutic interventions and provide a tool to investigate the mechanisms of neuropsychiatric disease processes, normal brain development and neuropharmacology in vivo.

Epidemiology of hemimegalencephaly: a case series and review

Hemimegalencephaly (HME) is a congenital brain malformation characterized by unilateral enlargement of the cerebral hemisphere. Clinically, HME is typically associated with hemiparesis, psychomotor retardation, and intractable seizures usually apparent soon after birth. HME is often an isolated finding, but it has been described as an occasional feature of a large number of syndromes, many of which may not be readily identified at birth. There are a multitude of case series and reports of HME in the English literature; however, there is no comprehensive, unbiased, detailed survey characterizing the proportion of cases of HME that are associated with a syndrome. We performed a retrospective study of all cases of HME seen at our institution from 1990 to 2003. Of the 15 cases of HME identified, 53% (8/15) were non-syndromic and 47% (7/15) of the cases were associated with a known or suspected genetic syndrome. In patients with syndromic HME, many of the syndromic features were not readily discernible at birth or in early infancy. It is, therefore, imperative to continually evaluate any infant with HME for signs and symptoms of these and other syndromes. Knowing the relative differential diagnosis will lead to a more comprehensive evaluation, improvement in expectant management, and appropriate counseling of families before considering radical surgical options such as hemispherectomy.

Segmentation of focal cortical dysplasia lesions on MRI using level set evolution

Focal cortical dysplasia (FCD) is the most frequent malformation of cortical development in patients with medically intractable epilepsy. On MRI, FCD lesions are not easily differentiable from the normal cortex and defining their spatial extent is challenging. In this paper, we introduce a method to segment FCD lesions on T1-weighted MRI. It relies on two successive three-dimensional deformable models, whose evolutions are based on the level set framework. The first deformable model is driven by probability maps obtained from three MRI features: cortical thickness, relative intensity and gradient. These features correspond to the visual characteristics of FCD and allow discriminating lesions and normal tissues. In a second stage, the previous result is expanded towards the underlying and overlying cortical boundaries, throughout the whole cortical section. The method was quantitatively evaluated by comparison with manually traced labels in 18 patients with FCD. The automated segmentations achieved a strong agreement with the manuals labels, demonstrating the applicability of the method to assist the delineation of FCD lesions on MRI. This new approach may become a useful tool for the presurgical evaluation of patients with intractable epilepsy related to cortical dysplasia.

Segmentation of focal cortical dysplasia lesions using a feature-based level set

Focal cortical dysplasia (FCD), a malformation of cortical development, is an important cause of medically intractable epilepsy. FCD lesions are difficult to distinguish from non-lesional cortex and their delineation on MRI is a challenging task. This paper presents a method to segment FCD lesions on T1-weighted MRI, based on a 3D deformable model, implemented using the level set framework. The deformable model is driven by three MRI features: cortical thickness, relative intensity and gradient. These features correspond to the visual characteristics of FCD and allow to differentiate lesions from normal tissues. The proposed method was tested on 18 patients with FCD and its performance was quantitatively evaluated by comparison with the manual tracings of two trained raters. The validation showed that the similarity between the level set segmentation and the manual labels is similar to the agreement between the two human raters. This new approach may become a useful tool for the presurgical evaluation of patients with intractable epilepsy.

Genetics of the polymicrogyria syndromes

Polymicrogyria is a relatively common malformation of cortical development, characterised by multiple small gyri with abnormal cortical lamination. The different forms of polymicrogyria encompass a wide range of clinical, aetiological, and histological findings. Advances in imaging have improved the diagnosis and classification of the condition. The molecular basis of polymicrogyria is beginning to be elucidated with the identification of a gene, GPR56, for bilateral frontoparietal polymicrogyria. Functional studies of the GPR56 gene product will yield insights not only into the causes of polymicrogyria but also into the mechanisms of normal cortical development and the regional patterning of the cerebral cortex. Based on imaging studies, several other region specific patterns of polymicrogyria have been identified, and there is increasing evidence that these may also have a significant genetic component to their aetiology. This paper reviews current knowledge of the different polymicrogyria syndromes, with discussion of clinical and imaging features, patterns of inheritance, currently mapped loci, candidate genes, chromosomal abnormalities, and implications for genetic counselling.

Prenatal MR diffusion-weighted imaging in a fetus with hemimegalencephaly

We report a case of hemimegalencephaly diagnosed by prenatal MRI with an emphasis on its appearance on diffusion-weighted images. This case shows that in this condition the enlarged hemisphere may show restricted diffusion on prenatal MRI. In our opinion, this finding may result from a combination of increased cellularity and advanced myelination in the affected hemisphere. Restricted diffusion is an additional valuable indicator in the analysis of the fetal brain.

Focal cortical dysplasia: a review of pathological features, genetics, and surgical outcome

Focal cortical dysplasia (FCD) is found in approximately one-half of patients with medically refractory epilepsy. These lesions may involve only mild disorganization of the cortex, but they may also contain abnormal neuronal elements such as balloon cells. Advances in neuroimaging have allowed better identification of these lesions, and thus more patients have become surgical candidates. Molecular biology techniques have been used to explore the genetics and pathophysiological characteristics of FCD. Data from surgical series have shown that surgery often results in significant reduction or cessation of seizures, especially if the entire lesion is resected.

Clinical and electroencephalographic features of infantile spasms associated with malformations of cortical development

The aim of this study was to reveal the clinical and encephalographic (EEG) features of infantile spasms (IS) with malformations of cortical development (MCDs). The clinical features, EEG findings, neuroimaging studies and outcomes of various therapeutic modalities for 27 patients with IS and MCDs were reviewed. Background activities of EEG on the MCDs, i.e. asymmetric hypsarrhythmia, localized persistent polymorphic slowing, asymmetric slowing and diffuse fast activities, were shown in 22, 15, 9 and 2 patients, respectively. Partial epileptiform discharges such as localized paroxysmal fast activities, spindle-shaped fast activities and subclinical seizures were shown in 15, 8 and 10 patients, respectively, and the lateralized prominence of generalized paroxysmal fast activities and generalized sharp and wave discharges in 4 and 5 patients, respectively. MCDs were suspected in 5 patients, as revealed by EEG and/or functional neuroimagings without distinct magnetic resonance imaging lesions, and confirmed by pathologic findings. Of the 11 patients treated with surgical resection, 8 became seizure free. EEG features can be very useful in the identification of underlying cortical dysgenesis in patients with IS. However, in one patient who underwent epilepsy surgery in early infancy, we observed easy bleeding and difficult hemostasis from friable vascular and parenchymal tissues. Various EEG features can be very useful in screening underlying MCDs. In addition, epilepsy surgery can be an effective therapeutic modality in many patients with otherwise medically intractable IS with MCDs. However, surgical intervention in extremely young infants should be performed with caution.

Individual voxel-based analysis of gray matter in focal cortical dysplasia

High-resolution MRI of the brain has made it possible to identify focal cortical dysplasia (FCD) in an increasing number of patients. There is evidence for structural abnormalities extending beyond the visually identified FCD lesion. Voxel-based morphometry (VBM) has the potential of detecting both lesions and extra-lesional abnormalities because it performs a whole brain voxel-wise comparison. However, on T1-weighted MRI, FCD lesions are characterized by a wide spectrum of signal hyperintensity that may compromise the results of the segmentation step in VBM. Our purpose was to investigate gray matter (GM) changes in individual FCD patients using voxel-based morphometry (VBM). In addition, we sought to assess the performance of this technique for FCD detection with respect to lesion intensity using an operator designed to emphasize areas of hyperintense T1 signal. We studied 27 patients with known FCD and focal epilepsy and 39 healthy controls. We compared the GM map of each subject (controls and patients) with the average GM map of all controls and obtained a GM z-score map for each individual. The protocol being designed to achieve a maximal specificity, no differences in GM concentration were found in the control group. The z-score maps showed an increase in GM that coincided with the lesion in 21/27 (78%) patients. Five of the six remaining patients whose lesions were not detected by VBM presented with a strong lesion hyperintensity, and a significant part of their lesion was misclassified as white matter. In 16/27 (59%) patients, there were additional areas of GM increase distant from the primary lesion. Areas of GM decrease were found in 8/27 (30%) patients. In conclusion, individual voxel-based analysis was able to detect FCD in a majority of patients. Moreover, FCD was often associated with widespread GM changes extending beyond the visible lesion. In its current form, however, individual VBM may be unable to detect lesions characterized by strong signal intensity abnormalities.