Dataset Information

Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study.

ABSTRACT: One outstanding challenge for machine learning in diagnostic biomedical imaging is algorithm interpretability. A key application is the identification of subtle epileptogenic focal cortical dysplasias (FCDs) from structural MRI. FCDs are difficult to visualize on structural MRI but are often amenable to surgical resection. We aimed to develop an open-source, interpretable, surface-based machine-learning algorithm to automatically identify FCDs on heterogeneous structural MRI data from epilepsy surgery centres worldwide. The Multi-centre Epilepsy Lesion Detection (MELD) Project collated and harmonized a retrospective MRI cohort of 1015 participants, 618 patients with focal FCD-related epilepsy and 397 controls, from 22 epilepsy centres worldwide. We created a neural network for FCD detection based on 33 surface-based features. The network was trained and cross-validated on 50% of the total cohort and tested on the remaining 50% as well as on 2 independent test sites. Multidimensional feature analysis and integrated gradient saliencies were used to interrogate network performance. Our pipeline outputs individual patient reports, which identify the location of predicted lesions, alongside their imaging features and relative saliency to the classifier. On a restricted 'gold-standard' subcohort of seizure-free patients with FCD type IIB who had T1 and fluid-attenuated inversion recovery MRI data, the MELD FCD surface-based algorithm had a sensitivity of 85%. Across the entire withheld test cohort the sensitivity was 59% and specificity was 54%. After including a border zone around lesions, to account for uncertainty around the borders of manually delineated lesion masks, the sensitivity was 67%. This multicentre, multinational study with open access protocols and code has developed a robust and interpretable machine-learning algorithm for automated detection of focal cortical dysplasias, giving physicians greater confidence in the identification of subtle MRI lesions in individuals with epilepsy.

SUBMITTER: Spitzer H

PROVIDER: S-EPMC9679165 | biostudies-literature | 2022 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study.

Spitzer Hannah H Ripart Mathilde M Whitaker Kirstie K D'Arco Felice F Mankad Kshitij K Chen Andrew A AA Napolitano Antonio A De Palma Luca L De Benedictis Alessandro A Foldes Stephen S Humphreys Zachary Z Zhang Kai K Hu Wenhan W Mo Jiajie J Likeman Marcus M Davies Shirin S Güttler Christopher C Lenge Matteo M Cohen Nathan T NT Tang Yingying Y Wang Shan S Chari Aswin A Tisdall Martin M Bargallo Nuria N Conde-Blanco Estefanía E Pariente Jose Carlos JC Pascual-Diaz Saül S Delgado-Martínez Ignacio I Pérez-Enríquez Carmen C Lagorio Ilaria I Abela Eugenio E Mullatti Nandini N O'Muircheartaigh Jonathan J Vecchiato Katy K Liu Yawu Y Caligiuri Maria Eugenia ME Sinclair Ben B Vivash Lucy L Willard Anna A Kandasamy Jothy J McLellan Ailsa A Sokol Drahoslav D Semmelroch Mira M Kloster Ane G AG Opheim Giske G Ribeiro Letícia L Yasuda Clarissa C Rossi-Espagnet Camilla C Hamandi Khalid K Tietze Anna A Barba Carmen C Guerrini Renzo R Gaillard William Davis WD You Xiaozhen X Wang Irene I González-Ortiz Sofía S Severino Mariasavina M Striano Pasquale P Tortora Domenico D Kälviäinen Reetta R Gambardella Antonio A Labate Angelo A Desmond Patricia P Lui Elaine E O'Brien Terence T Shetty Jay J Jackson Graeme G Duncan John S JS Winston Gavin P GP Pinborg Lars H LH Cendes Fernando F Theis Fabian J FJ Shinohara Russell T RT Cross J Helen JH Baldeweg Torsten T Adler Sophie S Wagstyl Konrad K

Brain : a journal of neurology 20221101 11

One outstanding challenge for machine learning in diagnostic biomedical imaging is algorithm interpretability. A key application is the identification of subtle epileptogenic focal cortical dysplasias (FCDs) from structural MRI. FCDs are difficult to visualize on structural MRI but are often amenable to surgical resection. We aimed to develop an open-source, interpretable, surface-based machine-learning algorithm to automatically identify FCDs on heterogeneous structural MRI data from epilepsy s ...[more]

PMID: 35953082

Similar Datasets

Project description:One third of patients with medically refractory focal epilepsy have normal-appearing MRI scans. This poses a problem as identification of the epileptogenic region is required for surgical treatment. This study performs a multimodal voxel-based analysis (VBA) to identify brain abnormalities in MRI-negative focal epilepsy. Data was collected from 69 focal epilepsy patients (42 with discrete lesions on MRI scans, 27 with no visible findings on scans), and 62 healthy controls. MR images comprised T1-weighted, fluid-attenuated inversion recovery (FLAIR), fractional anisotropy (FA) and mean diffusivity (MD) from diffusion tensor imaging, and neurite density index (NDI) from neurite orientation dispersion and density imaging. These multimodal images were coregistered to T1-weighted scans, normalized to a standard space, and smoothed with 8 mm FWHM. Initial analysis performed voxel-wise one-tailed t-tests separately on grey matter concentration (GMC), FLAIR, FA, MD, and NDI, comparing patients with epilepsy to controls. A multimodal non-parametric combination (NPC) analysis was also performed simultaneously on FLAIR, FA, MD, and NDI. Resulting p-maps were family-wise error rate corrected, threshold-free cluster enhanced, and thresholded at p < 0.05. Sensitivity was established through visual comparison of results to manually drawn lesion masks or seizure onset zone (SOZ) from stereoelectroencephalography. A leave-one-out cross-validation with the same analysis protocols was performed on controls to determine specificity. NDI was the best performing individual modality, detecting focal abnormalities in 38% of patients with normal MRI and conclusive SOZ. GMC demonstrated the lowest sensitivity at 19%. NPC provided superior performance to univariate analyses with 50% sensitivity. Specificity in controls ranged between 96 and 100% for all analyses. This study demonstrated the utility of a multimodal VBA utilizing NPC for detecting epileptogenic lesions in MRI-negative focal epilepsy. Future work will apply this approach to datasets from other centres and will experiment with different combinations of MR sequences.

Project description:ObjectivesAccurate segmentation of focal cortical dysplasia (FCD) lesions from MR images plays an important role in surgical planning and decision but is still challenging for radiologists and clinicians. In this study, we introduce a novel transformer-based model, designed for the end-to-end segmentation of FCD lesions from multi-channel MR images.MethodsThe core innovation of our proposed model is the integration of a convolutional neural network-based encoder-decoder structure with a multiscale transformer to augment the feature representation of lesions in the global field of view. Transformer pathways, composed of memory- and computation-efficient dual-self-attention modules, leverage feature maps from varying depths of the encoder to discern long-range interdependencies among feature positions and channels, thereby emphasizing areas and channels relevant to lesions. The proposed model was trained and evaluated on a public-open dataset including MR images of 85 patients using both subject-level and voxel-level metrics.ResultsExperimental results indicate that our model offers superior performance both quantitatively and qualitatively. It successfully identified lesions in 82.4% of patients, with a low false-positive lesion cluster rate of 0.176 ± 0.381 per patient. Furthermore, the model achieved an average Dice coefficient of 0.410 ± 0.288, outperforming five established methods.ConclusionIntegration of the transformer could enhance the feature presentation and segmentation performance of FCD lesions. The proposed model has the potential to serve as a valuable assistive tool for physicians, enabling rapid and accurate identification of FCD lesions. The source code and pre-trained model weights are available at https://github.com/zhangxd0530/MS-DSA-NET .Critical relevance statementThis multiscale transformer-based model performs segmentation of focal cortical dysplasia lesions, aiming to help radiologists and clinicians make accurate and efficient preoperative evaluations of focal cortical dysplasia patients from MR images.Key pointsThe first transformer-based model was built to explore focal cortical dysplasia lesion segmentation. Integration of global and local features enhances the segmentation performance of lesions. A valuable benchmark for model development and comparative analyses was provided.

Project description:ImportanceIt is controversial whether epilepsy is a static or progressive disease. Evidence of progressive gray matter loss in epilepsy would support early diagnosis, rapid treatment, and early referral for surgical interventions.ObjectiveTo demonstrate progressive cortical thinning in patients with focal epilepsy distinct from cortical thinning associated with normal aging.Design, setting, and participantsA case-control neuroimaging study was conducted from August 3, 2004, to January 26, 2016, among 190 patients with focal epilepsy at a tertiary epilepsy referral center (epilepsy data) and 3 independent comparison cohorts matched for age and sex (healthy volunteer data; n = 141).ExposuresTwo or more high-resolution T1-weighted magnetic resonance imaging scans at least 6 months apart (mean [SD] interval, 2.5 [1.6] years).Main outcomes and measuresGlobal and vertexwise rate of progressive cortical thinning.ResultsA total of 190 people with focal epilepsy (99 women and 91 men; mean [SD] age, 36 [11] years; 396 magnetic resonance imaging scans) were compared with 141 healthy volunteers (76 women and 65 men; mean [SD] age, 35 [17] years; 282 magnetic resonance imaging scans). Widespread highly significant progressive cortical thinning exceeding normal aging effects, mainly involving the bilateral temporal lobes, medial parietal and occipital cortices, pericentral gyri, and opercula, was seen in 146 individuals with epilepsy (76.8%; 95% CI, 58%-95%). The mean (SD) annualized rate of global cortical thinning in patients with epilepsy was twice the rate of age-associated thinning observed in healthy volunteers (0.024 [0.061] vs 0.011 [0.029] mm/y; P = .01). Progression was most pronounced in adults older than 55 years and during the first 5 years after the onset of seizures. Areas of accelerated cortical thinning were detected in patients with early onset of epilepsy and in patients with hippocampal sclerosis. Accelerated thinning was not associated with seizure frequency, history of generalized seizures, or antiepileptic drug load and did not differ between patients with or without ongoing seizures. Progressive atrophy in temporal (n = 101) and frontal (n = 28) lobe epilepsy was most pronounced ipsilaterally to the epileptic focus but also affected a widespread area extending beyond the focus and commonly affected the contralateral hemisphere. For patients with temporal lobe epilepsy, accelerated cortical thinning was observed within areas structurally connected with the ipsilateral hippocampus.Conclusions and relevanceWidespread progressive cortical thinning exceeding that seen with normal aging may occur in patients with focal epilepsy. These findings appear to highlight the need to develop epilepsy disease-modifying treatments to disrupt or slow ongoing atrophy. Longitudinal cortical thickness measurements may have the potential to serve as biomarkers for such studies.

Project description:ObjectiveFocal cortical dysplasia (FCD) is the most common pathological cause for pediatric epilepsy, with frontal lobe epilepsy (FLE) being the most prevalent in the pediatric population. We attempted to utilize radiomic and morphological methods on MRI and PET to detect FCD in children with FLE.MethodsThirty-seven children with FLE and 20 controls were included in the primary cohort, and a five-fold cross-validation was performed. In addition, we validated the performance in an independent site of 12 FLE children. A two-stage experiments including frontal lobe and subregions were employed to detect the lesion area of FCD, incorporating the asymmetric feature between the left and right hemispheres. Specifically, for the radiomics approach, we used gray matter (GM), white matter (WM), GM and WM, and the gray-white matter boundary regions of interest to extract features. Then, we employed a Multi-Layer Perceptron classifier to achieve FCD lesion localization based on both radiomic and morphological methods.ResultsThe Multi-Layer Perceptron model based on the asymmetric feature exhibited excellent performance both in the frontal lobe and subregions. In the primary cohort and independent site, the radiomics analysis with GM and WM asymmetric features had the highest sensitivity (89.2 and 91.7%) and AUC (98.9 and 99.3%) in frontal lobe. While in the subregions, the GM asymmetric features had the highest sensitivity (85.6 and 79.7%). Furthermore, relying on the highest sensitivity of GM and WM asymmetric features in frontal lobe, when integrated with the subregions results, our approach exhibited overlaps with GM asymmetric features (55.4 and 52.4%), as well as morphological asymmetric features (54.4 and 53.8%), both in the primary cohort and at the independent site.SignificanceThis study demonstrates that a two-stage design based on the asymmetry of radiomic and morphological features can improve FCD detection. Specifically, incorporating regions of interest for GM, WM, GM, and WM, and the gray-white matter boundary significantly enhances the localization capabilities for lesion detection within the radiomics approach.

Project description:ObjectiveTissue abnormalities in focal epilepsy may extend beyond the presumed focus. The underlying pathophysiology of these broader changes is unclear, and it is not known whether they result from ongoing disease processes or treatment-related side effects, or whether they emerge earlier. Few studies have focused on the period of onset for most focal epilepsies, childhood. Fewer still have utilized quantitative magnetic resonance imaging (MRI), which may provide a more sensitive and interpretable measure of tissue microstructural change. Here, we aimed to determine common spatial modes of changes in cortical architecture in children with heterogeneous drug-resistant focal epilepsy and, secondarily, whether changes were related to disease severity.MethodsTo assess cortical microstructure, quantitative T1 and T2 relaxometry (qT1 and qT2) was measured in 43 children with drug-resistant focal epilepsy (age range = 4-18 years) and 46 typically developing children (age range = 2-18 years). We assessed depth-dependent qT1 and qT2 values across the neocortex, as well as their gradient of change across cortical depths. We also determined whether global changes seen in group analyses were driven by focal pathologies in individual patients. Finally, as a proof-of-concept, we trained a classifier using qT1 and qT2 gradient maps from patients with radiologically defined abnormalities (MRI positive) and healthy controls, and tested whether this could classify patients without reported radiological abnormalities (MRI negative).ResultsWe uncovered depth-dependent qT1 and qT2 increases in widespread cortical areas in patients, likely representing microstructural alterations in myelin or gliosis. Changes did not correlate with disease severity measures, suggesting they may represent antecedent neurobiological alterations. Using a classifier trained with MRI-positive patients and controls, sensitivity was 71.4% at 89.4% specificity on held-out MRI-negative patients.SignificanceThese findings suggest the presence of a potential imaging endophenotype of focal epilepsy, detectable irrespective of radiologically identified abnormalities.

Dataset Information

Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study.

Publications

Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets