Atlas of lesion locations and postsurgical seizure freedom in focal cortical dysplasia: A MELD study

Wagstyl, Konrad; Whitaker, Kirstie; Raznahan, Armin; Seidlitz, Jakob; Vértes, Petra E.; Foldes, Stephen T.; Humphreys, Zachary; Hu, Wenhan; Mo, Jiajie; Likeman, Marcus; Davies, S; Lenge, Matteo; Nl, Cohen; Tang, Yingying; Wang, Shan; Ripart, Mathilde; Chari, Ajai; Tisdall, Martin; Bargalló, Núria; Conde‐Blanco, Estefanía; Pariente, José; Pascual‐Diaz, Saül; Delgado‐Martínez, Ignacio; Pérez‐Enríquez, Carmen; Lagorio, Ilaria; Abela, Eugenio; Mullatti, Nandini; O’Muircheartaigh, Jonathan; Vecchiato, Katy; Liu, Yawu; Caligiuri, Maria Eugenia; Sinclair, Ben; Vivash, Lucy; Willard, Anna; Kandasamy, Jothy; McLellan, Ailsa; Sokol, Drahoslav; Semmelroch, Mira; Kloster, Ane G.; Opheim, Giske; Yasuda, Clarissa Lin; Zhang, Kai; Hamandi, Khalid; Barba, Carmen; Guerrini, Renzo; Gaillard, William D.; You, Xiaozhen; Wang, Irène; González‐Ortiz, Sofía; Severino, Mariasavina; Striano, Pasquale; Tortora, Domenico; Kälviäinen, Reetta; Gambardella, Antonio; Labate, Angelo; Desmond, Patricia; Lui, Elaine; O'Brien, Terry; Shetty, Jay; Jackson, Graeme D.; Duncan, John; Winston, Gavin P.; Pinborg, Lars H.; Cendes, Fernando; Cross, J. Helen; Baldeweg, Torsten; Adler, Sophie

doi:10.1111/epi.17130

Cited by 45 publications

(58 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Images were processed using a standardized pipeline (MELD protocol 8 ) and FCDs masked with manually drawn 3D regions of interest (ROI) on T1 or fluid‐attenuated inversion recovery (FLAIR) images. The ROIs were projected onto individual FreeSurfer 19 surfaces then registered to fsaverage_sym, 20 a bilaterally symmetrical template.…”

Section: Methodsmentioning

confidence: 99%

“…7 The international, Multi-Centre Epilepsy Lesion Detection (MELD project) of 580 patients found that FCD type II are more commonly frontal and types I and III are more commonly temporal; larger lesions and lesions in primary somatosensory or motor areas have earlier age of epilepsy onset compared to lesions in higherorder regions. 8 Sulcal depth may play a role in FCD epileptogenesis. Functional connectivity variability correlates with sulcal depth variability with increased gyrification in association areas and decreased gyrification in older unimodal cortex; whereas cortical thickness is not related to network functional connectivity.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Networks Underlie Temporal Onset of Dysplasia‐Related Epilepsy: A MELD Study

You

Krishnamurthy

et al. 2022

Annals of Neurology

Self Cite

View full text Add to dashboard Cite

Objective The purpose of this study was to evaluate if focal cortical dysplasia (FCD) co‐localization to cortical functional networks is associated with the temporal distribution of epilepsy onset in FCD. Methods International (20 center), retrospective cohort from the Multi‐Centre Epilepsy Lesion Detection (MELD) project. Patients included if >3 years old, had 3D pre‐operative T1 magnetic resonance imaging (MRI; 1.5 or 3 T) with radiologic or histopathologic FCD after surgery. Images processed using the MELD protocol, masked with 3D regions‐of‐interest (ROI), and co‐registered to fsaverage_sym (symmetric template). FCDs were then co‐localized to 1 of 7 distributed functional cortical networks. Negative binomial regression evaluated effect of FCD size, network, histology, and sulcal depth on age of epilepsy onset. From this model, predictive age of epilepsy onset was calculated for each network. Results Three hundred eighty‐eight patients had median age seizure onset 5 years (interquartile range [IQR] = 3–11 years), median age at pre‐operative scan 18 years (IQR = 11–28 years). FCDs co‐localized to the following networks: limbic (90), default mode (87), somatomotor (65), front parietal control (52), ventral attention (32), dorsal attention (31), and visual (31). Larger lesions were associated with younger age of onset (p = 0.01); age of epilepsy onset was associated with dominant network (p = 0.04) but not sulcal depth or histology. Sensorimotor networks had youngest onset; the limbic network had oldest age of onset (p values <0.05). Interpretation FCD co‐localization to distributed functional cortical networks is associated with age of epilepsy onset: sensory neural networks (somatomotor and visual) with earlier onset, and limbic latest onset. These variations may reflect developmental differences in synaptic/white matter maturation or network activation and may provide a biological basis for age‐dependent epilepsy onset expression. ANN NEUROL 2022;92:503–511

show abstract

Section: Methodsmentioning

confidence: 99%

mentioning

confidence: 99%

Networks Underlie Temporal Onset of Dysplasia‐Related Epilepsy: A MELD Study

You

Krishnamurthy

et al. 2022

Annals of Neurology

Self Cite

View full text Add to dashboard Cite

show abstract

“…The likelihood of seizure freedom decreased with a longer duration of epilepsy. 17 Finally, AI, including ML and DL, has been applied to neuroimaging data for predicting clinical diagnosis, that is, clinical phenotyping from imaging, 32,33 and predicting response to antiseizure medications. 34 For example, dMRI measures and connectome profiles may identify patients with juvenile myoclonic epilepsy 32 and distinguish patients with focal epilepsy vs healthy controls.…”

Section: Artificial Intelligence Applications For Epilepsy Diagnosis ...mentioning

confidence: 99%

“…[13][14][15][16] Another is the multicenter epilepsy lesion detection (MELD) project. 17 These datasets include hundreds to thousands of MRI and clinical data in patients with epilepsy and are ripe for AI applications.…”

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence Applications in the Imaging of Epilepsy and Its Comorbidities: Present and Future

Cendes

McDonald

2022

Epilepsy Curr

Self Cite

View full text Add to dashboard Cite

Artificial intelligence (AI) is increasingly used in medical image analysis and has accelerated scientific discoveries across fields of medicine. In this review, we highlight how AI has been applied to neuroimaging in patients with epilepsy to enhance classification of clinical diagnosis, prediction of treatment outcomes, and the understanding of cognitive comorbidities. We outline the strengths and shortcomings of current AI research and the need for future studies using large datasets that test the reproducibility and generalizability of current findings, as well as studies that test the clinical utility of AI approaches.

show abstract

“…Here, as part of the MELD Project (Wagstyl et al, 2021), we aimed to collate a cohort of patients from multiple epilepsy surgery centres, across multiple MRI scanners including both 1.5T and 3T field strengths; create protocols for de-centralised MRI post-processing; and develop an open-access, robust and interpretable surface-based classifier to detect FCD.…”

Section: Introductionmentioning

confidence: 99%

Interpretable surface-based detection of focal cortical dysplasias: a MELD study

Spitzer

Ripart

Whitaker

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Introduction: 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 visualise 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. Methods: The Multi-centre Epilepsy Lesion Detection (MELD) Project collated and harmonised 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. Results: Our pipeline outputs individual patient reports, which identify the location of predicted lesions, alongside their imaging features and relative saliency to the classifier. Overall, after including a border-zone around lesions, the developed MELD FCD surface-based algorithm had a sensitivity of 67% and a specificity of 54% on the withheld test cohort, and a sensitivity of 85% on a restricted subcohort of seizure free patients with FCD type IIB who had T1 and FLAIR MRI data. Conclusions: 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.

show abstract

Atlas of lesion locations and postsurgical seizure freedom in focal cortical dysplasia: A MELD study

Abstract: This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Cited by 45 publications

References 36 publications

Networks Underlie Temporal Onset of Dysplasia‐Related Epilepsy: A MELD Study

Networks Underlie Temporal Onset of Dysplasia‐Related Epilepsy: A MELD Study

Artificial Intelligence Applications in the Imaging of Epilepsy and Its Comorbidities: Present and Future

Interpretable surface-based detection of focal cortical dysplasias: a MELD study

Contact Info

Product

Resources

About