Automated quantitative FLAIR analysis in hippocampal sclerosis

Voxel-based and surface-based cortical morphometric MRI applications for identifying the epileptogenic zone: A narrative review

Approximately 40% of patients with drug-resistant epilepsy referred for surgical evaluation have no epileptogenic lesion on MRI (MRI-negative). MRI-negative epilepsy is associated with poorer seizure freedom prognosis and has therefore motivated the development of structural post-processing methods to "convert" MRI-negative to MRI-positive cases. In this article, we review the principles, advances, and challenges of voxel- and surface-based cortical morphometric MRI techniques in detecting the epileptogenic zone. The ground truth for the presumed epileptogenic zone in imaging studies can be classified into lesion-based (MRI lesion mask or histopathology) or epileptogenicity-based ground truth (anatomical-electroclinical correlations or resections that lead to seizure freedom). Voxel-based techniques are reported to have a 13%-97% concordance rate, while surface-based techniques have 67%-92% compared to lesion-based ground truths. Epileptogenicity-based ground truth may be more relevant in the case of MRI-negative cases; however, the sensitivity and concordance rate (voxel-based technique 7.1%-66.7%, and surface-based technique 62%) are limited by the reliance on scalp EEG and qualitative analysis of seizure-onset pattern. The use of stereo-EEG and quantitative EEG analysis may fill this gap to evaluate the correlation between cortical morphometry results and electrophysiological epileptogenic biomarkers of the epileptogenic zone and help improve the yield of structural post-processing tools. PLAIN LANGUAGE SUMMARY: Locating the epileptogenic zone (the brain area that is responsible for seizure generation) is important to diagnose and plan epilepsy treatments. An abnormal brain imaging (MRI) result can help clinical decision-making; however, around 40% of patients have normal MRI results (MRI-negative). We are reviewing the potential of two advanced MRI methods (voxel- and surface-based cortical morphometry) to localize the epileptogenic zone in the presence or absence of visible MRI abnormalities. We also describe the current challenge of applying the above methods in daily clinical practice and propose using advanced brain recording analysis to aid this translation process.

Automated and Interpretable Detection of Hippocampal Sclerosis in Temporal Lobe Epilepsy: AID-HS

OBJECTIVE

Hippocampal sclerosis (HS), the most common pathology associated with temporal lobe epilepsy (TLE), is not always visible on magnetic resonance imaging (MRI), causing surgical delays and reduced postsurgical seizure-freedom. We developed an open-source software to characterize and localize HS to aid the presurgical evaluation of children and adults with suspected TLE.

METHODS

We included a multicenter cohort of 365 participants (154 HS; 90 disease controls; 121 healthy controls). HippUnfold was used to extract morphological surface-based features and volumes of the hippocampus from T1-weighted MRI scans. We characterized pathological hippocampi in patients by comparing them to normative growth charts and analyzing within-subject feature asymmetries. Feature asymmetry scores were used to train a logistic regression classifier to detect and lateralize HS. The classifier was validated on an independent multicenter cohort of 275 patients with HS and 161 healthy and disease controls.

RESULTS

HS was characterized by decreased volume, thickness, and gyrification alongside increased mean and intrinsic curvature. The classifier detected 90.1% of unilateral HS patients and lateralized lesions in 97.4%. In patients with MRI-negative histopathologically-confirmed HS, the classifier detected 79.2% (19/24) and lateralized 91.7% (22/24). The model achieved similar performances on the independent cohort, demonstrating its ability to generalize to new data. Individual patient reports contextualize a patient's hippocampal features in relation to normative growth trajectories, visualise feature asymmetries, and report classifier predictions.

INTERPRETATION

Automated and Interpretable Detection of Hippocampal Sclerosis (AID-HS) is an open-source pipeline for detecting and lateralizing HS and outputting clinically-relevant reports. ANN NEUROL 2024.

MRI evidence of olfactory system alterations in patients with COVID-19 and neurological symptoms

BACKGROUND AND OBJECTIVE

Despite olfactory disorders being among the most common neurological complications of coronavirus disease 2019 (COVID-19), their pathogenesis has not been fully elucidated yet. Brain MR imaging is a consolidated method for evaluating olfactory system's morphological modification, but a few quantitative studies have been published so far. The aim of the study was to provide MRI evidence of olfactory system alterations in patients with COVID-19 and neurological symptoms, including olfactory dysfunction.

METHODS

196 COVID-19 patients (median age: 53 years, 56% females) and 39 controls (median age 55 years, 49% females) were included in this cross-sectional observational study; 78 of the patients reported olfactory loss as the only neurological symptom. MRI processing was performed by ad-hoc semi-automatic processing procedures. Olfactory bulb (OB) volume was measured on T2-weighted MRI based on manual tracing and normalized to the brain volume. Olfactory tract (OT) median signal intensity was quantified on fluid attenuated inversion recovery (FLAIR) sequences, after preliminary intensity normalization.

RESULTS

COVID-19 patients showed significantly lower left, right and total OB volumes than controls (p < 0.05). Age-related OB atrophy was found in the control but not in the patient population. No significant difference was found between patients with olfactory disorders and other neurological symptoms. Several outliers with abnormally high OT FLAIR signal intensity were found in the patient group.

CONCLUSIONS

Brain MRI findings demonstrated OB damage in COVID-19 patients with neurological complications. Future longitudinal studies are needed to clarify the transient or permanent nature of OB atrophy in COVID-19 pathology.

'Hippocampal innate inflammatory gliosis only' in pharmacoresistant temporal lobe epilepsy

Drug-resistant mesial-temporal lobe epilepsy is a devastating disease with seizure onset in the hippocampal formation. A fraction of hippocampi samples from epilepsy-surgical procedures reveals a peculiar histological pattern referred to as 'gliosis only' with unresolved pathogenesis and enigmatic sequelae. Here, we hypothesize that 'gliosis only' represents a particular syndrome defined by distinct clinical and molecular characteristics. We curated an in-depth multiparameter integration of systematic clinical, neuropsychological as well as neuropathological analysis from a consecutive cohort of 627 patients, who underwent hippocampectomy for drug-resistant temporal lobe epilepsy. All patients underwent either classic anterior temporal lobectomy or selective amygdalohippocampectomy. On the basis of their neuropathological exam, patients with hippocampus sclerosis and 'gliosis only' were characterized and compared within the whole cohort and within a subset of matched pairs. Integrated transcriptional analysis was performed to address molecular differences between both groups. 'Gliosis only' revealed demographics, clinical and neuropsychological outcome fundamentally different from hippocampus sclerosis. 'Gliosis only' patients had a significantly later seizure onset (16.3 versus 12.2 years, P = 0.005) and worse neuropsychological outcome after surgery compared to patients with hippocampus sclerosis. Epilepsy was less amendable by surgery in 'gliosis only' patients, resulting in a significantly worse rate of seizure freedom after surgery in this subgroup (43% versus 68%, P = 0.0001, odds ratio = 2.8, confidence interval 1.7-4.7). This finding remained significant after multivariate and matched-pairs analysis. The 'gliosis only' group demonstrated pronounced astrogliosis and lack of significant neuronal degeneration in contrast to characteristic segmental neuron loss and fibrillary astrogliosis in hippocampus sclerosis. RNA-sequencing of gliosis only patients deciphered a distinct transcriptional programme that resembles an innate inflammatory response of reactive astrocytes. Our data indicate a new temporal lobe epilepsy syndrome for which we suggest the term 'Innate inflammatory gliosis only'. 'Innate inflammatory gliosis only' is characterized by a diffuse gliosis pattern lacking restricted hippocampal focality and is poorly controllable by surgery. Thus, 'innate inflammatory gliosis only' patients need to be clearly identified by presurgical examination paradigms of pharmacoresistant temporal lobe epilepsy patients; surgical treatment of this subgroup should be considered with great precaution. 'Innate inflammatory gliosis only' requires innovative pharmacotreatment strategies.

CERMEP-IDB-MRXFDG: a database of 37 normal adult human brain [18F]FDG PET, T1 and FLAIR MRI, and CT images available for research

We present a database of cerebral PET FDG and anatomical MRI for 37 normal adult human subjects (CERMEP-IDB-MRXFDG). Thirty-nine participants underwent static [18F]FDG PET/CT and MRI, resulting in [18F]FDG PET, T1 MPRAGE MRI, FLAIR MRI, and CT images. Two participants were excluded after visual quality control. We describe the acquisition parameters, the image processing pipeline and provide participants' individual demographics (mean age 38 ± 11.5 years, range 23-65, 20 women). Volumetric analysis of the 37 T1 MRIs showed results in line with the literature. A leave-one-out assessment of the 37 FDG images using Statistical Parametric Mapping (SPM) yielded a low number of false positives after exclusion of artefacts. The database is stored in three different formats, following the BIDS common specification: (1) DICOM (data not processed), (2) NIFTI (multimodal images coregistered to PET subject space), (3) NIFTI normalized (images normalized to MNI space). Bona fide researchers can request access to the database via a short form.

Automated detection of hippocampal sclerosis: Comparison of a composite MRI-based index with conventional MRI measures

PURPOSE

This study aims to compare the performance of an MRI-based composite index (HSI) with conventional MRI-based measures in hippocampal sclerosis (HS) detection and postoperative outcome estimation.

METHODS

Seventy-two temporal lobe epilepsy (TLE) patients with pathologically confirmed HS and fifteen TLE patients without HS were included retrospectively. The T1-weighted and FLAIR images of these patients were processed with AccuBrain to quantify the hippocampal volume (HV) and the hippocampal FLAIR signal. The HSI index that considered both HV and hippocampal FLAIR signal was also calculated. Two experienced neuropathologists rated the HS severity with the resected tissue and reached an agreement for all cases. The asymmetry indices of the MRI measures were used to lateralize the sclerotic side, and the original MRI measures were applied to detect HS vs. normal hippocampi. Operating characteristic curve (ROC) analyses were performed for these predictions. We also investigated the sensitivity of the ipsilateral MRI measures in characterizing the pathological severity of HS and the associations of the MRI measures with postoperative outcomes (Engel class categories).

RESULTS

With the optimal cutoffs, the asymmetry indices of HSI and HV both achieved excellent performance in differentiating left vs. right HS (accuracy = 100 %), and the absolute value of the asymmetry index of HSI performed best in differentiating unilateral vs. bilateral HS (accuracy = 91.7 %). Regarding the detection of HS, HSI performed better in sensitivity (94.4 % vs. 87.5 %) while HV performed better in specificity (93.6 % vs. 89.4 %) when the contralateral site of unilateral HS and both sides of non-HS patients were considered as the normal reference, and HSI performed even better than HV when only both sides of non-HS patients were considered as the normal reference (AUC: 0.956 vs. 0.934, p = 0.038). The ipsilateral HSI presented the strongest association with the pathological rating of HS severity (r = 0.405, p < 0.001). None of the ipsilateral or contralateral MRI measures was associated with the postoperative outcomes. Among the asymmetry indices, only the absolute value of the asymmetry index of HV presented a significant association with the Engel classifications for the Year 2∼3 visit (r = -0.466, p = 0.004) or the latest visit with >1 year follow-up (r = -0.374, p = 0.003) while controlling for disease duration and follow-up duration.

CONCLUSION

The HSI index and HV presented comparable good performance in HS detection, and HSI may have better sensitivity than HV in differentiating pathological HS severity. Higher magnitude of HV dissymmetry may indicate better post-surgical outcomes for HS patients.

Automatic segmentation of white matter hyperintensities from brain magnetic resonance images in the era of deep learning and big data - A systematic review

BACKGROUND

White matter hyperintensities (WMH), of presumed vascular origin, are visible and quantifiable neuroradiological markers of brain parenchymal change. These changes may range from damage secondary to inflammation and other neurological conditions, through to healthy ageing. Fully automatic WMH quantification methods are promising, but still, traditional semi-automatic methods seem to be preferred in clinical research. We systematically reviewed the literature for fully automatic methods developed in the last five years, to assess what are considered state-of-the-art techniques, as well as trends in the analysis of WMH of presumed vascular origin.

METHOD

We registered the systematic review protocol with the International Prospective Register of Systematic Reviews (PROSPERO), registration number - CRD42019132200. We conducted the search for fully automatic methods developed from 2015 to July 2020 on Medline, Science direct, IEE Explore, and Web of Science. We assessed risk of bias and applicability of the studies using QUADAS 2.

RESULTS

The search yielded 2327 papers after removing 104 duplicates. After screening titles, abstracts and full text, 37 were selected for detailed analysis. Of these, 16 proposed a supervised segmentation method, 10 proposed an unsupervised segmentation method, and 11 proposed a deep learning segmentation method. Average DSC values ranged from 0.538 to 0.91, being the highest value obtained from an unsupervised segmentation method. Only four studies validated their method in longitudinal samples, and eight performed an additional validation using clinical parameters. Only 8/37 studies made available their methods in public repositories.

CONCLUSIONS

We found no evidence that favours deep learning methods over the more established k-NN, linear regression and unsupervised methods in this task. Data and code availability, bias in study design and ground truth generation influence the wider validation and applicability of these methods in clinical research.

Improved detection of focal cortical dysplasia in normal-appearing FLAIR images using a Bayesian classifier

PURPOSE

Focal cortical dysplasia (FCD) is a malformation of cortical development that often causes pharmacologically intractable epilepsy. However, FCD lesions are frequently characterized by minor structural abnormalities that can easily go unrecognized, making diagnosis difficult. Therefore, many epileptic patients have had pathologically confirmed FCD lesions that appeared normal in pre-surgical fluid-attenuated inversion recovery (FLAIR) magnetic resonance (MR) studies. Such lesions are called "FLAIR-negative." This study aimed to improve the detection of histopathologically verified FCD in a sample of patients without visually appreciable lesions.

METHODS

The technique first extracts a series of features from a FLAIR image. Then, three naive Bayesian classifiers with probability (NBCP) are trained based on different numbers of feature maps to classify voxels as lesional or healthy voxels and assign the lesions a probability of correct classification. This method classifies the three-dimensional (3D) images of all patients using leave-one-out cross-validation (LOOCV). Finally, the 3D lesion probability map, including epileptogenic lesions, is obtained by removing false-positive voxel outliers using the morphological method. The performance of the NBCP was assessed for quantitative analysis by specificity, accuracy, recall, precision, and Dice coefficient in subject-wise, lesion-wise, and voxel-wise manners.

RESULTS

The best detection results were obtained by using four features: cortical thickness, symmetry, K-means, and modified texture energy. There were eight lesions in seven patients. The subject-wise sensitivity of the proposed method was 85.71% (6/7). Seven out of eight lesions were detected, so the lesion-wise sensitivity was 87.50% (7/8). No significant differences in effectiveness were found between automated lesion detection using four features and lesion detection using manual segmentation, as voxels were quantitatively analyzed in terms of specificity (mean ± SD = 99.64 ± 0.13), accuracy (mean ± SD = 99.62 ± 0.14), recall (mean ± SD = 73.27 ± 26.11), precision (mean ± SD = 11.93 ± 8.16), and Dice coefficient (mean ± SD = 22.82 ± 15.57).

CONCLUSION

We developed a novel automatic voxel-based method to improve the detection of FCD FLAIR-negative lesions. To the best of our knowledge, this study is the first to detect FCD lesions that appear normal in pre-surgical 3D high-resolution FLAIR images alone with a limited number of radiomics features. We optimized the algorithm and selected the best prior probability to improve the detection. For non-temporal lobe epilepsy (non-TLE) patients, lesions could be accurately located, although there were still false-positive areas.

Brain MRI post-processing with MAP07 in the preoperative evaluation of patients with pharmacoresistant epilepsy - Croatian single centre experience

OBJECTIVE

This study aimed to determine the role of brain MRI post-processing method MAP07 (Morphometric Analysis Program) in detecting epileptogenic brain lesions in patients with pharmacoresistant epilepsy (PE). MAP07 is a sophisticated diagnostic program that offers several morphometric maps and facilitates the detection and localization of hippocampal sclerosis (HS), focal cortical dysplasias (FCD), and other types of cortical malformations, which could be undetected by conventional visual MRI analysis (CVA).

METHODS

120 patients aged > 16 years with PE have been recruited. 3 T MRI was performed according to epilepsy imaging protocol followed by image postprocessing with a fully automated MATLAB script, MAP07, by applying SPM5 algorithms. Statistical analysis was performed in IBM SPSS Statistics, version 25.0.

RESULTS

Analysis in our patients showed a high sensitivity of MAP07 with low specificity and with a high proportion of false-positive patients. After MRI analysis, out of 120 patients, 32 were found to have no structural abnormalities by conventional visual analysis in whom after MAP07 in 5 patients structural lesions were found (in one HS, in one FCD, in two perinatal vascular lesions, and in one hippocampal hyperintensity). There was a quite high overall coincidence of the findings of MAP07 and MRI for the detection of FCD, HS, perinatal ischemia/chronic vascular lesions, heterotopias, and polymicrogyria (kappa coefficient above 0.700).

CONCLUSIONS

MAP07 analysis is a useful, additional, and automated method that may guide re-evaluation of MRI by highlighting suspicious cortical regions, as a complementary method to CVA, by enhancing the visualization of cortical malformations and lesions.

Voxel-Based Morphometry-from Hype to Hope. A Study on Hippocampal Atrophy in Mesial Temporal Lobe Epilepsy

BACKGROUND AND PURPOSE

Automated volumetry of the hippocampus is considered useful to assist the diagnosis of hippocampal sclerosis in temporal lobe epilepsy. However, voxel-based morphometry is rarely used for individual subjects because of high rates of false-positives. We investigated whether an approach with high dimensional warping to the template and nonparametric statistics would be useful to detect hippocampal atrophy in patients with hippocampal sclerosis.

MATERIALS AND METHODS

We performed single-subject voxel-based morphometry with nonparametric statistics within the framework of Statistical Parametric Mapping to compare MRI from 26 well-characterized patients with temporal lobe epilepsy individually against a group of 110 healthy controls. The following statistical threshold was used: P < .05 corrected for multiple comparisons with family-wise error over the region of interest right and left hippocampus.

RESULTS

The sensitivity for the detection of atrophy related to hippocampal sclerosis was 0.92 (95% CI, 0.67-0.99) for the right hippocampus and 0.60 (0.31-0.83) for the left, and the specificity for volume changes was 0.98 (0.93-0.99). All clusters of decreased hippocampal volumes were correctly lateralized to the seizure focus. Hippocampal volume decrease was in accordance with neuronal cell loss on histology reports.

CONCLUSIONS

Nonparametric voxel-based morphometry is sensitive and specific for hippocampal atrophy in patients with mesial temporal lobe epilepsy and may be useful in clinical practice.

Adult-onset temporal lobe epilepsy suspicious for autoimmune pathogenesis: Autoantibody prevalence and clinical correlates

Temporal lobe adult-onset seizures (TAOS) related to autoimmunity represent an increasingly recognized disease syndrome within the spectrum of epilepsies. In this context, certain autoantibodies (autoABs) were often associated with limbic encephalitis (LE). Here, we aimed to gain insights into (a) the distribution of ‘neurological’ autoABs (neuroABs, defined as autoABs targeting neuronal surface structures or ‘onconeuronal’ ABs or anti-glutamate acid decarboxylase 65 (GAD65) autoABs) in a large consecutive TAOS patient cohort, to characterize (b) clinical profiles of seropositive versus seronegative individuals and to find (c) potential evidence for other autoABs. Blood sera/cerebrospinal fluid (CSF) of TAOS patients (n = 800) and healthy donors (n = 27) were analyzed for neuroABs and screened for other autoABs by indirect immunofluorescence on hippocampal/cerebellar sections and immunoblots of whole brain and synaptosome lysates. Serological results were correlated with clinico-neuropsychological features. 13% of TAOS patients (n = 105) were neuroAB⁺, with anti-GAD65 and anti-N-methyl-D-aspartate receptors (NMDAR) as most frequent autoABs in this group. In our screening tests 25% of neuroAB^- patients (n = 199) were positive (screening⁺), whereas all control samples were negative (n = 27). Intriguingly, key clinico-neuropsychological characteristics including magnetic resonance imaging (MRI) findings, epileptiform electroencephalographic (EEG) activity, and inflammatory cellular infiltrates in CSF were shared to a greater extent by neuroAB⁺ with neuroAB^-/screening⁺ patients than with neuroAB^-/screening^- patients. Serological testing in a large consecutive TAOS patient series revealed seropositivity for anti-GAD65 autoABs as the most frequent neuroAB. Intriguingly, neuroAB⁺ individuals were virtually indistinguishable from neuroAB^-/screening⁺ patients in several major clinical features. In contrast, neuroAB^-/screening^- TAOS patients differed in many parameters. These data support the potential presence of so far unrecognized autoABs in patients with TAOS.

An image processing algorithm to aid diagnosis of mesial temporal sclerosis in children: a case-control study

BACKGROUND

Mesial temporal sclerosis (MTS) is an important cause of intractable epilepsy. Early and accurate diagnosis of MTS is essential to providing curative and life-changing therapy but can be challenging in children in whom the impact of diagnosis is particularly high. Magnetic resonance imaging (MRI) plays an important role in the diagnosis of MTS, and image processing of MRI is a recently studied strategy to improve its accuracy.

OBJECTIVE

In a retrospective case-control study, we assessed the performance of an image processing algorithm (Correlative Image Enhancement [CIE]) for detecting MTS-related hippocampal signal abnormality in children.

MATERIALS AND METHODS

Twenty-seven pediatric MTS cases (9 males, 18 females; mean age: 16±standard deviation [SD] 6.7 years) were identified from a pathology database of amygdylo-hippocampectomies performed in children with epilepsy. Twenty-seven children with no seizure history (9 males, 18 females; mean age: 13.8±SD 2.8 years), and with normal brain MRI, were identified for the control group. Blinded investigators processed the MRI coronal FLAIR (fluid-attenuated inversion recovery) images with CIE, saved the processed images as a separate series, and made equivalent region of interest measurements on the processed and unprocessed series to calculate contrast-to-noise ratio. Six blinded reviewers then rated the randomized series for hippocampal signal abnormality and MTS disease status.

RESULTS

CIE increased signal intensity and contrast-to-noise ratio in 26/27 hippocampi with pathologically confirmed MTS (96.3%) with the mean (SD) contrast-to-noise ratio of cases increasing from 14.9 (11.1) to 77.7 (58.7) after processing (P<0.001). Contrast-to-noise ratio increased in 1/54 normal control hippocampi (1.9%), with no significant change in the mean contrast-to-noise ratio of the control group after processing (P=0.57). Mean (SD) reader sensitivity for detecting abnormal signal intensity increased from 83.3% (14.2) to 94.8% (3.3) after processing. Mean specificity for abnormal signal intensity increased from 94.4% (7.3) to 96.3% (0). While sensitivity improved after processing for detection of MTS disease status in 4/6 readers, the mean reader sensitivity and specificity for MTS detection increased only minimally after processing, from 79.6% to 80.7% and from 95.7% to 96.3%, respectively.

CONCLUSION

The CIE image processing algorithm selectively increased the contrast-to-noise ratio of hippocampi affected by MTS, improved reader performance in detecting MTS-related hippocampal signal abnormality and could have high impact on pediatric patients undergoing work-up for seizures.

Hippocampal profiling: Localized magnetic resonance imaging volumetry and T2 relaxometry for hippocampal sclerosis

Objective

Hippocampal sclerosis (HS) is the most common cause of drug‐resistant temporal lobe epilepsy, and its accurate detection is important to guide epilepsy surgery. Radiological features of HS include hippocampal volume loss and increased T2 signal, which can both be quantified to help improve detection. In this work, we extend these quantitative methods to generate cross‐sectional area and T2 profiles along the hippocampal long axis to improve the localization of hippocampal abnormalities.

Methods

T1‐weighted and T2 relaxometry data from 69 HS patients (32 left, 32 right, 5 bilateral) and 111 healthy controls were acquired on a 3‐T magnetic resonance imaging (MRI) scanner. Automated hippocampal segmentation and T2 relaxometry were performed and used to calculate whole‐hippocampal volumes and to estimate quantitative T2 (qT2) values. By generating a group template from the controls, and aligning this so that the hippocampal long axes were along the anterior‐posterior axis, we were able to calculate hippocampal cross‐sectional area and qT2 by a slicewise method to localize any volume loss or T2 hyperintensity. Individual patient profiles were compared with normative data generated from the healthy controls.

Results

Profiling of hippocampal volumetric and qT2 data could be performed automatically and reproducibly. HS patients commonly showed widespread decreases in volume and increases in T2 along the length of the affected hippocampus, and focal changes may also be identified. Patterns of atrophy and T2 increase in the left hippocampus were similar between left, right, and bilateral HS. These profiles have potential to distinguish between sclerosis affecting volume and qT2 in the whole or parts of the hippocampus, and may aid the radiological diagnosis in uncertain cases or cases with subtle or focal abnormalities where standard whole‐hippocampal measurements yield normal values.

Significance

Hippocampal profiling of volumetry and qT2 values can help spatially localize hippocampal MRI abnormalities and work toward improved sensitivity of subtle focal lesions.

Image Processing to Improve Detection of Mesial Temporal Sclerosis in Adults

In this retrospective case-control study, we investigated whether an image-processing algorithm designed to exaggerate the intensity of diseased hippocampi on FLAIR images can improve the diagnostic accuracy and interobserver reliability of radiologists in detecting mesial temporal sclerosis-related hippocampal signal alteration. Herein, we share the results of this study that showed that the image processing improved the confidence of radiologists in detecting mesial temporal sclerosis-related signal alteration, allowing an improved sensitivity, specificity, and interobserver reliability.

Use of an Automated Quantitative Analysis of Hippocampal Volume, Signal, and Glucose Metabolism to Detect Hippocampal Sclerosis

Purpose: Magnetic resonance imaging (MRI) and positron emission tomography (PET) with ¹⁸F-fluorodeoxyglucose (¹⁸FDG) are valuable tools for evaluating hippocampal sclerosis (HS); however, bias may arise during visual analyses. The aim of this study was to evaluate and compare MRI and PET post-processing techniques, automated quantitative hippocampal volume (Q-volume), and fluid-attenuated inversion-recovery (FLAIR) signal (Q-FLAIR) and glucose metabolism (Q-PET) analyses in patients with HS.

Methods: We collected MRI and ¹⁸FDG-PET images from 54 patients with HS and 22 healthy controls and independently performed conventional visual analyses (CVA) of PET (CVA-PET) and MRI (CVA-MRI) images. During the subsequent quantitative analyses, the hippocampus was segmented from the 3D T1 image, and the mean volumetric, FLAIR intensity and standardized uptake value ratio (SUVR) values of the left and right hippocampus were assessed in each subject. Threshold confidence levels calculated from the mean volumetric, FLAIR intensity and SUVR values of the controls were used to identify healthy subjects or subjects with HS. The performance of the three methods was assessed using receiver operating characteristic (ROC) curves, and the detection rates of CVA-MRI, CVA-PET, Q-volume, Q-FLAIR, and Q-PET were statistically compared.

Results: The areas under the curves (AUCs) for the Q-volume, Q-FLAIR, and Q-PET ROC analyses were 0.88, 0.41, and 0.98, which suggested a diagnostic method with moderate, poor, and high accuracy, respectively. Although Q-PET had the highest detection rate among the two CVA methods and three quantitative methods, the difference between Q-volume and Q-PET did not reach statistical significance. Regarding the HS subtypes, CVA-MRI, CVA-PET, Q-volume, and Q-PET had similar detection rates for type 1 HS, and Q-PET was the most sensitive method for detecting types 2 and 3 HS.

Conclusions: In MRI or ¹⁸FDG-PET images that have been visually assessed by experts, the quantification of hippocampal volume or glucose uptake can increase the detection of HS and appear to be additional valuable diagnostic tools for evaluating patients with epilepsy who are suspected of having HS.

Brain lesion segmentation through image synthesis and outlier detection

Cerebral small vessel disease (SVD) can manifest in a number of ways. Many of these result in hyperintense regions visible on T2-weighted magnetic resonance (MR) images. The automatic segmentation of these lesions has been the focus of many studies. However, previous methods tended to be limited to certain types of pathology, as a consequence of either restricting the search to the white matter, or by training on an individual pathology. Here we present an unsupervised abnormality detection method which is able to detect abnormally hyperintense regions on FLAIR regardless of the underlying pathology or location. The method uses a combination of image synthesis, Gaussian mixture models and one class support vector machines, and needs only be trained on healthy tissue. We evaluate our method by comparing segmentation results from 127 subjects with SVD with three established methods and report significantly superior performance across a number of metrics.

Visualization of heterogeneity and regional grading of gliomas by multiple features using magnetic resonance-based clustered images

Preoperative glioma grading is important for therapeutic strategies and influences prognosis. Intratumoral heterogeneity can cause an underestimation of grading because of the sampling error in biopsies. We developed a voxel-based unsupervised clustering method with multiple magnetic resonance imaging (MRI)-derived features using a self-organizing map followed by K-means. This method produced novel magnetic resonance-based clustered images (MRcIs) that enabled the visualization of glioma grades in 36 patients. The 12-class MRcIs revealed the highest classification performance for the prediction of glioma grading (area under the receiver operating characteristic curve = 0.928; 95% confidential interval = 0.920–0.936). Furthermore, we also created 12-class MRcIs in four new patients using the previous data from the 36 patients as training data and obtained tissue sections of the classes 11 and 12, which were significantly higher in high-grade gliomas (HGGs), and those of classes 4, 5 and 9, which were not significantly different between HGGs and low-grade gliomas (LGGs), according to a MRcI-based navigational system. The tissues of classes 11 and 12 showed features of malignant glioma, whereas those of classes 4, 5 and 9 showed LGGs without anaplastic features. These results suggest that the proposed voxel-based clustering method provides new insights into preoperative regional glioma grading.

Brain imaging in the assessment for epilepsy surgery

Brain imaging has a crucial role in the presurgical assessment of patients with epilepsy. Structural imaging reveals most cerebral lesions underlying focal epilepsy. Advances in MRI acquisitions including diffusion-weighted imaging, post-acquisition image processing techniques, and quantification of imaging data are increasing the accuracy of lesion detection. Functional MRI can be used to identify areas of the cortex that are essential for language, motor function, and memory, and tractography can reveal white matter tracts that are vital for these functions, thus reducing the risk of epilepsy surgery causing new morbidities. PET, SPECT, simultaneous EEG and functional MRI, and electrical and magnetic source imaging can be used to infer the localisation of epileptic foci and assist in the design of intracranial EEG recording strategies. Progress in semi-automated methods to register imaging data into a common space is enabling the creation of multimodal three-dimensional patient-specific datasets. These techniques show promise for the demonstration of the complex relations between normal and abnormal structural and functional data and could be used to direct precise intracranial navigation and surgery for individual patients.

White matter abnormalities differentiate severe from benign temporal lobe epilepsy

OBJECTIVE

Temporal and extratemporal white matter abnormalities have been identified frequently in patients with refractory mesial temporal lobe epilepsy (rMTLE). However, the identification of potential water diffusion abnormalities in patients with drug-responsive, benign MTLE (bMTLE) is still missing. The aim of this study was to identify markers of refractoriness in MTLE.

METHODS

The study group included 48 patients with bMTLE (mean age 42.8 + 13.5 years), 38 with rMTLE (mean age 41.7 + 14.1 years) and 54 healthy volunteers. Diffusion tensor imaging (DTI) was performed to measure mean diffusivity (MD) and fractional anisotropy (FA) in a regions-of-interest analysis comprising hippocampi and temporal lobe gray and white matter regions. The presence of hippocampal sclerosis (Hs) was assessed using automated magnetic resonance imaging (MRI) evaluation. For statistics we used chi-square test; two-tailed, two-sample t-test; and stratified linear regression.

RESULTS

The significant demographic differences between the two patient groups were sex (p = 0.003), duration of epilepsy (p = 0.003) and complex febrile convulsions (p = 0.0001). In rMTLE, temporal white matter MD was higher and FA lower, as compared to bMTLE. The analysis of diagnostic accuracy (area under the receiver operator characteristic [ROC] curve [AUC]) showed that FA had an AUC for discriminating patients affected from those unaffected by refractory MTLE of 74.0% (p < 0.001), a value that was higher than that of temporal MD (64.0%), hippocampus volume (65.0%), and Hs (66.0%).

SIGNIFICANCE

We performed DTI measurements in MTLE and found a significant reduction of FA along the white matter of the temporal lobes in rMTLE, suggesting it as a valuable measure of refractoriness in MTLE.

Post-processing of structural MRI for individualized diagnostics

Currently, a relevant proportion of all histopathologically proven focal cortical dysplasia (FCD) escape visual detection; this shows the need for additional improvements in analyzing MRI data. A positive MRI is still the strongest prognostic factor for postoperative freedom of seizures. Among several post-processing methods voxel-based morphometry (VBM) of T1- and T2-weighted sequences and T2 relaxometry are routinely applied in pre-surgical diagnostics of cryptogenic epilepsy in epilepsy centers. VBM is superior to conventional visual analysis with 9-15% more identified epileptogenic foci, while T2 relaxometry has its main application in (mesial) temporal lobe epilepsy. Further methods such as surface-based morphometry (SBM) or diffusion tensor imaging are promising but there is a lack of current studies comparing their individual diagnostic value. Post-processing methods represent an important addition to conventional visual analysis but need to be interpreted with expertise and experience so that they should be apprehended as a complementary tool within the context of the multi-modal evaluation of epilepsy patients. This review will give an overview of existing post-processing methods of structural MRI and outline their clinical relevance in detection of epileptogenic structural changes.

T2 mapping outperforms normalised FLAIR in identifying hippocampal sclerosis

RATIONALE

Qualitatively, FLAIR MR imaging is sensitive to the detection of hippocampal sclerosis (HS). Quantitative analysis of T2 maps provides a useful objective measure and increased sensitivity over visual inspection of T2-weighted scans. We aimed to determine whether quantification of normalised FLAIR is as sensitive as T2 mapping in detection of HS.

METHOD

Dual echo T2 and FLAIR MR images were retrospectively analysed in 27 patients with histologically confirmed HS and increased T2 signal in ipsilateral hippocampus and 14 healthy controls. Regions of interest were manually segmented in all hippocampi aiming to avoid inclusion of CSF. Hippocampal T2 values and measures of normalised FLAIR Signal Intensity (nFSI) were compared in healthy and sclerotic hippocampi.

RESULTS

HS was identified on T2 values with 100% sensitivity and 100% specificity. HS was identified on nFSI measures with 60% sensitivity and 93% specificity.

CONCLUSION

T2 mapping is superior to nFSI for identification of HS.

3-T magnetic resonance imaging simultaneous automated multimodal approach improves detection of ambiguous visual hippocampal sclerosis

BACKGROUND AND PURPOSE

To evaluate if an automatic magnetic resonance imaging (MRI) processing system may improve detection of hippocampal sclerosis (Hs) in patients with mesial temporal lobe epilepsy (MTLE).

METHODS

Eighty consecutive patients with a diagnosis of MTLE and 20 age- and sex-matched controls were prospectively recruited and included in our study. The entire group had 3-T MRI visual assessment of Hs analysed by two blinded imaging epilepsy experts. Logistic regression was used to evaluate the performances of neuroradiologists and multimodal analysis.

RESULTS

The multimodal automated tool gave no evidence of Hs in all 20 controls and classified the 80 MTLE patients as follows: normal MRI (54/80), left Hs (14/80), right Hs (11/80) and bilateral Hs (1/80). Of note, this multimodal automated tool was always concordant with the side of MTLE, as determined by a comprehensive electroclinical evaluation. In comparison with standard visual assessment, the multimodal automated tool resolved five ambiguous cases, being able to lateralize Hs in four patients and detecting one case of bilateral Hs. Moreover, comparing the performances of the three logistic regression models, the multimodal approach overcame performances obtained with a single image modality for both the hemispheres, reaching a global accuracy value of 0.97 for the right and 0.98 for the left hemisphere.

CONCLUSIONS

Multimodal quantitative automated MRI is a reliable and useful tool to depict and lateralize Hs in patients with MTLE, and may help to lateralize the side of MTLE especially in subtle and uncertain cases.

Is the type and extent of hippocampal sclerosis measurable on high-resolution MRI?

INTRODUCTION

The purpose of this study is to relate hippocampal volume and FLAIR signal intensity to Wyler grading of hippocampal sclerosis (HS).

METHODS

Of 100 consecutive patients with temporal lobe epilepsy and HS as histopathological diagnosis, 32 had high-resolution 3 Tesla MRI and anatomically well-preserved hippocampi following amygdalo-hippocampectomy. Hippocampal volume on 3D T1-weighted gradient echo and signal intensity on coronal FLAIR sequences were determined using FreeSurfer and SPM tools and related to Wyler grading. Seizure outcome was determined after 1 year.

RESULTS

Histopathology showed four Wyler II, 19 Wyler III, and 9 Wyler IV HS. Hippocampal volumes were 3.08 ml for Wyler II (Wyler II/contralateral side: p > 0.05), 2.19 ml for Wyler III (p < 0.01), 2.62 ml for Wyler IV (p = 0.01), and 3.08 ml for the contralateral side. Normalized FLAIR signals were 1,354 (p = 0.0004), 1,408 (p < 0.0001), 1,371 (p < 0.04), and 1,296, respectively. Wyler II hippocampi were visually normal. Two of four (50%) Wyler II, 16/19 (84%) Wyler III, and 6/9 (66%) Wyler IV patients achieved Engel I outcome.

CONCLUSIONS

Combined volumetry and quantitative FLAIR signal analysis clearly identifies Wyler III and IV HS. Quantitative FLAIR signal analysis may be helpful to identify Wyler II HS.

Quantitative FLAIR analysis indicates predominant affection of the amygdala in antibody-associated limbic encephalitis

PURPOSE

Limbic encephalitis is an autoimmune-mediated disease leading to temporal lobe epilepsy, mnestic deficits, and affective disturbances. Magnetic resonance imaging (MRI) usually shows signal and volume changes of the temporomesial structures. However, these abnormalities may be subtle, thereby hampering the diagnosis by conventional visual assessment. In the present study we evaluated the diagnostic value of a fully automated MRI postprocessing technique in limbic encephalitis and hippocampal sclerosis.

METHODS

The MRI postprocessing was based largely on a recently described method allowing for an observer-independent quantification of the fluid-attenuated inversion recovery (FLAIR) signal intensities of amygdala and hippocampus. A 95% confidence region was calculated from the FLAIR intensities of 100 healthy controls. We applied this analysis to the MRI data of 39 patients with antibody-associated limbic encephalitis and 63 patients with hippocampal sclerosis. Moreover, the results were compared to those of visual assessment by an experienced neuroradiologist.

KEY FINDINGS

The method detected limbic encephalitis and hippocampal sclerosis with a high sensitivity of 85% and 95%, respectively. The detection rate of the automated approach in limbic encephalitis was significantly superior to visual analysis (85% vs. 51%; p = 0.001), whereas no statistically significant difference for the detection rate in hippocampal sclerosis was found. Patients with limbic encephalitis had significantly higher absolute intensity values of the amygdala and a significantly higher percentage fell outside of the amygdalar confidence region compared to those with hippocampal sclerosis (79% vs. 27%; p < 0.001), whereas we found opposite results in the hippocampal analysis (38% vs. 95%; p < 0.001).

SIGNIFICANCE

The FLAIR analysis applied in this study is a powerful tool to quantify signal changes of the amygdala and hippocampus in limbic encephalitis and hippocampal sclerosis. It significantly increases the diagnostic sensitivity in limbic encephalitis in comparison to conventional visual analysis. Furthermore, the method provides an interesting insight into the distinct properties of these two disease entities on MRI, indicating a predominant affection of the amygdala in limbic encephalitis, whereas the affection of the hippocampus is far less pronounced when compared to hippocampal sclerosis.