Artificial intelligence for classification of temporal lobe epilepsy with ROI-level MRI data: A worldwide ENIGMA-Epilepsy study

Autor(es):Mostrar menos -

Gleichgerrcht, Ezequiel ^[1] ; Munsell, Brent C. ^{[2, 3]} ; Alhusaini, Saud ^{[4, 5]} ; Alvim, Marina K. M. ^{[6, 7]} ; Bargallo, Nuria ^{[8, 9]} ; Bender, Benjamin ^[10] ; Bernasconi, Andrea ^[11] ; Bernasconi, Neda ^[11] ; Bernhardt, Boris ^[12] ; Blackmon, Karen ^[13] ; Caligiuri, Maria Eugenia ^[14] ; Cendes, Fernando ^{[6, 7]} ; Concha, Luis ^[15] ; Desmond, Patricia M. ^[16] ; Devinsky, Orrin ^[17] ; Doherty, Colin P. ^{[18, 19]} ; Domin, Martin ^[20] ; Duncan, John S. ^[21] ; Focke, Niels K. ^[22] ; Gambardella, Antonio ^{[23, 14]} ; Gong, Bo ^[24] ; Guerrini, Renzo ^[25] ; Hatton, Sean N. ^[26] ; Kalviainen, Reetta ^{[27, 28]} ; Keller, Simon S. ^{[29, 30]} ; Kochunov, Peter ^[31] ; Kotikalapudi, Raviteja ^{[32, 10, 33]} ; Kreilkamp, Barbara A. K. ^{[30, 22]} ; Labate, Angelo ^{[23, 14]} ; Langner, Soenke ^{[34, 35]} ; Lariviere, Sara ^[12] ; Lenge, Matteo ^{[36, 37]} ; Lui, Elaine ^[16] ; Martin, Pascal ^[32] ; Mascalchi, Mario ^[38] ; Meletti, Stefano ^{[39, 40]} ; O'Brien, Terence J. ^{[41, 42, 43]} ; Pardoe, Heath R. ^[17] ; Pariente, Jose C. ^[8] ; Rao, Jun Xian ^[44] ; Richardson, Mark P. ^[45] ; Rodriguez-Cruces, Raul ^{[46, 15]} ; Ruber, Theodor ^[47] ; Sinclair, Ben ^{[41, 42, 43]} ; Soltanian-Zadeh, Hamid ^{[48, 49]} ; Stein, Dan J. ^[50] ; Striano, Pasquale ^{[51, 52]} ; Taylor, Peter N. ^{[53, 51]} ; Thomas, Rhys H. ^[54] ; Vaudano, Anna Elisabetta ^{[39, 40]} ; Vivash, Lucy ^{[41, 42, 43]} ; von Podewills, Felix ^[55] ; Vos, Sjoerd B. ^{[56, 57]} ; Weber, Bernd ^[58] ; Yao, Yi ^[58] ; Yasuda, Clarissa Lin ^{[6, 7]} ; Zhang, Junsong ^[59] ; Thompson, Paul M. ^[60] ; Sisodiya, Sanjay M. ^{[61, 62]} ; McDonald, Carrie R. ^[44] ; Bonilha, Leonardo ^[1] ; Grp, ENIGMA-Epilepsy Working

Afiliação do(s) autor(es):Mostrar menos -

^[1] Med Univ South Carolina, Dept Neurol, Charleston, SC 29425 - USA

^[2] Univ N Carolina, Dept Psychiat, Chapel Hill, NC - USA

^[3] Univ N Carolina, Dept Comp Sci, Chapel Hill, NC - USA

^[4] Royal Coll Surgeons Ireland, Dept Mol & Cellular Therapeut, Dublin - Ireland

^[5] Yale Univ, Sch Med, Neurol Dept, New Haven, CT - USA

^[6] Univ Campinas UNICAMP, Dept Neurol, Campinas, SP - Brazil

^[7] Univ Campinas UNICAMP, Neuroimaging Lab, Campinas, SP - Brazil

^[8] Univ Barcelona, Inst Invest Biomed August Pi & Sunyer IDIBAPS, Magnet Resonance Image Core Facil, Barcelona - Spain

^[9] Hosp Clin Barcelona, Dept Radiol, Ctr Image Diag CDIC, Barcelona - Spain

^[10] Univ Hosp Tubingen, Dept Diagnost & Intervent Neuroradiol, Tubingen - Germany

^[11] McGill Univ, Montreal Neurol Inst, Neuroimaging Epilepsy Lab, Montreal, PQ - Canada

^[12] McGill Univ, McConnell Brain Imaging Ctr, Montreal Neurol Inst, Montreal, PQ - Canada

^[13] Mayo Clin, Psychiat & Psychol, Jacksonville, FL - USA

^[14] Magna Graecia Univ Catanzaro, Neurosci Res Ctr, Dept Med & Surg Sci, Catanzaro - Italy

^[15] Univ Nacl Autonoma Mexico, Inst Neurobiol, Mexico City, DF - Mexico

^[16] Univ Melbourne, Royal Melbourne Hosp, Dept Radiol, Melbourne, Vic - Australia

^[17] NYU, Langone Sch Med, Dept Neurol, New York, NY - USA

^[18] Trinity Coll Dublin, Sch Med, Dublin - Ireland

^[19] FutureNeuro SFI Res Ctr Rare & Chron Neurol Dis, Dublin - Ireland

^[20] Univ Med Greifswald, Dept Diagnost Radiol & Neuroradiol, Funct Imaging Unit, Greifswald - Germany

^[21] UCL Queen Sq Inst Neurol, Dept Clin & Expt Epilepsy, London - England

^[22] Univ Med Gottingen, Clin Neurophysiol, Gottingen - Germany

^[23] Magna Graecia Univ Catanzaro, Inst Neurol, Catanzaro - Italy

^[24] Univ British Columbia, BC Childrens Hosp, Dept Radiol, Vancouver, BC - Canada

^[25] Univ Florence, Neurosci Dept, Florence - Italy

^[26] Univ Calif San Diego, Ctr Multimodal Imaging & Genet, La Jolla, CA 92093 - USA

^[27] Kuopio Univ Hosp, EpiCARE ERN, Kuopio - Finland

^[28] Univ Eastern Finland, Inst Clin Med, Neurol, Kuopio - Finland

^[29] Walton Ctr NHS Fdn Trust, Liverpool, Merseyside - England

^[30] Univ Liverpool, Inst Syst Mol & Integrat Biol, Liverpool, Merseyside - England

^[31] Univ Maryland, Sch Med, Dept Psychiat, Baltimore, MD 21201 - USA

^[32] Univ Hosp Tubingen, Hertie Inst Clin Brain Res, Dept Neurol & Epileptol, Tubingen - Germany

^[33] Univ Hosp Gottingen, Dept Clin Neurophysiol, Gottingen - Germany

^[34] Univ Med Greifswald, Inst Diagnost Radiol & Neuroradiol, Greifswald - Germany

^[35] Univ Med Ctr Rostock, Inst Diagnost & Intervent Radiol, Pediat & Neuroradiol, Rostock - Germany

^[36] Childrens Hosp A Meyer Univ Florence, Pediat Neurol Neurogenet & Neurobiol Unit & Labs, Florence - Italy

^[37] Childrens Hosp A Meyer Univ Florence, Neurosurg Dept, Funct & Epilepsy Neurosurg Unit, Florence - Italy

^[38] Univ Florence, Mario Serio Dept Clin & Expt Med Sci, Florence - Italy

^[39] Univ Modena & Reggio Emilia, Dept Biomed Metab & Neural Sci, Modena - Italy

^[40] AOU Modena, Neurol Unit, OCB Hosp, Modena - Italy

^[41] Monash Univ, Dept Neurosci, Melbourne, Vic - Australia

^[42] Univ Melbourne, Dept Med, Royal Melbourne Hosp, Parkville, Vic - Australia

^[43] Alfred Hlth, Dept Neurol, Melbourne, Vic - Australia

^[44] Univ Calif San Diego, Dept Psychiat, La Jolla, CA 92093 - USA

^[45] Kings Coll London, Div Neurosci, London - England

^[46] McGill Univ, Montreal Neurol Inst & Hosp, Montreal, PQ - Canada

^[47] Univ Hosp Bonn, Dept Epileptol, Bonn - Germany

^[48] Henry Ford Hlth Syst, Radiol & Res Adm, Detroit, MI - USA

^[49] Univ Tehran, Coll Engn, Sch Elect & Comp Engn, Tehran - Iran

^[50] Univ Cape Town, Dept Psychiat & Neurosci Inst, SA MRC Unit Risk & Resilience Mental Disorders, Cape Town - South Africa

^[51] Univ Genoa, Dept Neurosci Rehabil Ophthalmol Genet Maternal &, Genoa - Italy

^[52] IRCCS Ist G Gaslini, Genoa - Italy

^[53] Newcastle Univ, Sch Comp, Newcastle Upon Tyne, Tyne & Wear - England

^[54] Newcastle Univ, Inst Translat & Clin Res, Newcastle Upon Tyne, Tyne & Wear - England

^[55] Univ Med Greifswald, Epilepsy Ctr, Dept Neurol, Greifswald - Germany

^[56] UCL, Ctr Med Image Comp, Dept Comp Sci, London - England

^[57] UCL, Neuroradiol Acad Unit, UCL Queen Sq Inst Neurol, London - England

^[58] Univ Bonn, Inst Expt Epileptol & Cognit Res, Bonn - Germany

^[59] Xiamen Univ, Sch Informat, Cognit Sci Dept, Xiamen - Peoples R China

^[60] Univ Southern Calif, Imaging Genet Ctr, Keck Sch Med, Mark & Mary Stevens Inst Neuroimaging & Informat, Marina Del Rey, CA - USA

^[61] Chalfont Ctr Epilepsy, Gerrards Cross, Bucks - England

^[62] UCL Queen Sq Inst Neurol, London - England

• Tipo de documento: Artigo Científico
• Fonte: NEUROIMAGE-CLINICAL; v. 31, 2021.
• Citações Web of Science: 1

Resumo

Artificial intelligence has recently gained popularity across different medical fields to aid in the detection of diseases based on pathology samples or medical imaging findings. Brain magnetic resonance imaging (MRI) is a key assessment tool for patients with temporal lobe epilepsy (TLE). The role of machine learning and artificial intelligence to increase detection of brain abnormalities in TLE remains inconclusive. We used support vector machine (SV) and deep learning (DL) models based on region of interest (ROI-based) structural (n = 336) and diffusion (n = 863) brain MRI data from patients with TLE with ({''}lesional{''}) and without ({''}non-lesional{''}) radiographic features suggestive of underlying hippocampal sclerosis from the multinational (multi-center) ENIGMA-Epilepsy consortium. Our data showed that models to identify TLE performed better or similar (68-75%) compared to models to lateralize the side of TLE (56-73%, except structural-based) based on diffusion data with the opposite pattern seen for structural data (67-75% to diagnose vs. 83% to lateralize). In other aspects, structural and diffusion-based models showed similar classification accuracies. Our classification models for patients with hippocampal sclerosis were more accurate (68-76%) than models that stratified non-lesional patients (53-62%). Overall, SV and DL models performed similarly with several instances in which SV mildly outperformed DL. We discuss the relative performance of these models with ROI-level data and the implications for future applications of machine learning and artificial intelligence in epilepsy care. (AU)

• Processo FAPESP: 13/07559-3 - Instituto Brasileiro de Neurociência e Neurotecnologia - BRAINN
• Beneficiário: Fernando Cendes
• Modalidade de apoio: Auxílio à Pesquisa - Centros de Pesquisa, Inovação e Difusão - CEPIDs
• Processo FAPESP: 15/17066-0 - Avaliação da relação entre marcadores inflamatórios e padrão de atrofia hipocampal e extra-hipocampal em pacientes com epilepsia de lobo temporal
• Beneficiário: Marina Koutsodontis Machado Alvim
• Modalidade de apoio: Bolsas no Brasil - Doutorado Direto