Plate IV.227 · Publications
Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study
Article in Brain, 10 August 2022.
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.
Versions2
- PublishedBrain10.1093/brain/awac224
- PreprintmedRxiv10.1101/2021.12.13.21267721
Cite
@article{spitzer2022interpretable,
title = {{Interpretable surface-based detection of focal cortical dysplasias: a Multi-centre Epilepsy Lesion Detection study}},
author = {Spitzer, Hannah and Ripart, Mathilde and Whitaker, Kirstie and D’Arco, Felice and Mankad, Kshitij and Chen, Andrew A and Napolitano, Antonio and De Palma, Luca and De Benedictis, Alessandro and Foldes, Stephen and Humphreys, Zachary and Zhang, Kai and Hu, Wenhan and Mo, Jiajie and Likeman, Marcus and Davies, Shirin and Güttler, Christopher and Lenge, Matteo and Cohen, Nathan T and Tang, Yingying and Wang, Shan and Chari, Aswin and Tisdall, Martin and Bargallo, Nuria and Conde-Blanco, Estefanía and Pariente, Jose Carlos and Pascual-Diaz, Saül and Delgado-Martínez, Ignacio and Pérez-Enríquez, Carmen and Lagorio, Ilaria and Abela, Eugenio and Mullatti, Nandini and O’Muircheartaigh, Jonathan and Vecchiato, Katy and Liu, Yawu and Caligiuri, Maria Eugenia and Sinclair, Ben and Vivash, Lucy and Willard, Anna and Kandasamy, Jothy and McLellan, Ailsa and Sokol, Drahoslav and Semmelroch, Mira and Kloster, Ane G and Opheim, Giske and Ribeiro, Letícia and Yasuda, Clarissa and Rossi-Espagnet, Camilla and Hamandi, Khalid and Tietze, Anna and Barba, Carmen and Guerrini, Renzo and Gaillard, William Davis and You, Xiaozhen and Wang, Irene and González-Ortiz, Sofía and Severino, Mariasavina and Striano, Pasquale and Tortora, Domenico and Kälviäinen, Reetta and Gambardella, Antonio and Labate, Angelo and Desmond, Patricia and Lui, Elaine and O’Brien, Terence and Shetty, Jay and Jackson, Graeme and Duncan, John S and Winston, Gavin P and Pinborg, Lars H and Cendes, Fernando and Theis, Fabian J and Shinohara, Russell T and Cross, J Helen and Baldeweg, Torsten and Adler, Sophie and Wagstyl, Konrad},
journal = {Brain},
publisher = {Oxford University Press (OUP)},
year = {2022},
month = aug,
doi = {10.1093/brain/awac224},
url = {https://doi.org/10.1093/brain/awac224},
pmid = {35953082},
pmcid = {PMC9679165}
}