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

Hannah Spitzer, Institute of Computational Biology, Helmholtz Center Munich, Munich 85764, Germany.
Mathilde Ripart, Department of Developmental Neuroscience, UCL Great Ormond Street Institute for Child Health, London WC1N 1EH, UK.
Kirstie Whitaker, The Alan Turing Institute, London NW1 2DB, UK.
Felice D'Arco, Great Ormond Street Hospital NHS Foundation Trust, London WC1N 3JH, UK.
Kshitij Mankad, Great Ormond Street Hospital NHS Foundation Trust, London WC1N 3JH, UK.
Andrew A. Chen, Penn Statistics in Imaging and Visualization Center, Department of Biostatistics, Epidemiology, and Informatics, University of Pennsylvania, Philadelphia, PA 19104, USA.
Antonio Napolitano, Medical Physics Department, Bambino Gesù Children's Hospital, Rome 00165, Italy.
Luca De Palma, Rare and Complex Epilepsies, Department of Neurosciences, Bambino Gesù Children's Hospital, IRCCS, Rome 00165, Italy.
Alessandro De Benedictis, Neurosurgery Unit, Department of Neurosciences, Bambino Gesù Children's Hospital, IRCCS, Rome 00165, Italy.
Stephen Foldes, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ 85016, USA.Follow
Zachary Humphreys, Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, AZ 85016, USA.
Kai Zhang, Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing 100054, China.
Wenhan Hu, Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing 100054, China.
Jiajie Mo, Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing 100054, China.
Marcus Likeman, Bristol Royal Hospital for Children, Bristol BS2 8BJ, UK.
Shirin Davies, School of Psychology, Cardiff University Brain Research Imaging Centre, Cardiff CF24 4HQ, UK.
Christopher Güttler, Charité University Hospital, Berlin 10117, Germany.
Matteo Lenge, Neuroscience Department, Children's Hospital Meyer-University of Florence, Florence 50139, Italy.
Nathan T. Cohen, Center for Neuroscience, Children's National Hospital, Washington, DC 20012, USA.
Yingying Tang, Department of Neurology, West China Hospital of Sichuan University, Chengdu 610093, China.
Shan Wang, Epilepsy Center, Cleveland Clinic, Cleveland, OH 44106, USA.
Aswin Chari, Department of Developmental Neuroscience, UCL Great Ormond Street Institute for Child Health, London WC1N 1EH, UK.
Martin Tisdall, Department of Developmental Neuroscience, UCL Great Ormond Street Institute for Child Health, London WC1N 1EH, UK.
Nuria Bargallo, Department of Neuroradiology, Hospital Clinic Barcelona and Magnetic Resonance Imaging, Core Facility, IDIBAPS, Barcelona 08036, Spain.
Estefanía Conde-Blanco, Magnetic Resonance Imaging, Core Facility, IDIBAPS, Barcelona 08036, Spain.
Jose Carlos Pariente, Magnetic Resonance Imaging, Core Facility, IDIBAPS, Barcelona 08036, Spain.
Saül Pascual-Diaz, Magnetic Resonance Imaging, Core Facility, IDIBAPS, Barcelona 08036, Spain.
Ignacio Delgado-Martínez, Department of Neurosurgery, Hospital del Mar, Barcelona 08003, Spain.
Carmen Pérez-Enríquez, Department of Neurology, Hospital del Mar, Barcelona 08003, Spain.
Ilaria Lagorio, IRCCS Istituto Giannina Gaslini, Genova 16147, Italy.
Eugenio Abela, Center for Neuropsychiatry and Intellectual Disability, Psychiatrische Dienste Aargau AG, Windisch 5120, Switzerland.
Nandini Mullatti, Institute of Psychiatry, Psychology and Neuroscience, King's College, London SE5 8AF, UK.

Document Type

Article

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.

Keywords

epilepsy, focal cortical dysplasia, machine learning, structural MRI

Medical Subject Headings

Humans; Retrospective Studies; Malformations of Cortical Development (complications, diagnostic imaging); Epilepsy (diagnostic imaging); Magnetic Resonance Imaging (methods); Machine Learning; Epilepsies, Partial (diagnostic imaging)

Publication Date

11-21-2022

Publication Title

Brain : a journal of neurology

E-ISSN

1460-2156

Volume

145

Issue

11

First Page

3859

Last Page

3871

PubMed ID

35953082

Digital Object Identifier (DOI)

10.1093/brain/awac224

Link to Full Text

Share

COinS