Skip to main content
Advertisement
  • Loading metrics

Correction: Evaluating the generalisability of region-naïve machine learning algorithms for the identification of epilepsy in low-resource settings

  • Ioana Duta,
  • Symon M. Kariuki,
  • Anthony K. Ngugi,
  • Angelina Kakooza Mwesige,
  • Honorati Masanja,
  • Daniel M. Mwanga,
  • Seth Owusu-Agyei,
  • Ryan Wagner,
  • J. Helen Cross,
  • Josemir W. Sander,
  • Charles R. Newton,
  • Arjune Sen,
  • Gabriel Davis Jones
  • Article
  • Metrics
  • Comments
  • Media Coverage

There are errors in the authors’ affiliations. The correct affiliations are listed below:

Ioana Duta1,2, Symon M. Kariuki3,4,5, Anthony K. Ngugi4,6, Angelina Kakooza Mwesige7, Honorati Masanja8, Daniel M. Mwanga9,10, Seth Owusu-Agyei11,12, Ryan Wagner13, J. Helen Cross14, Josemir W. Sander15,16,17, Charles R. Newton1,3,4,18, Arjune Sen1, Gabriel Davis Jones1,2,19

1 Oxford Epilepsy Research Group, Nuffield Department of Clinical Neurosciences, John Radcliffe Hospital, Oxford, United Kingdom, 2 Oxford Digital Health Labs, Nuffield Department of Women’s and Reproductive Health, The University of Oxford, John Radcliffe Hospital, Oxford, United Kingdom, 3 KEMRI/Wellcome Trust Research Programme, Centre for Geographic Medicine Research – Coast, Kilifi, Kenya, 4 Studies of Epidemiology of Epilepsy in Demographic Surveillance Systems (SEEDS) – INDEPTH Network, Accra, Ghana, 5 Department of Public Health, Pwani University, Kilifi, Kenya, 6 Department of Population Health, Aga Khan University, Nairobi, Kenya, 7 Department of Paediatrics and Child Health, Makerere University College of Health Sciences, Kampala, Uganda, 8 Ifakara Health Institute, Ifakara, Tanzania, 9 Data Synergy and Evaluations, African Population and Health Research Center, Nairobi, Kenya, 10 Department of Mathematics, University of Nairobi, Nairobi, Kenya, 11 Kintampo Health Research Centre, Kintampo, Ghana, 12 Institute of Health Research, University of Health and Allied Sciences, Ho, Ghana, 13 MRC/Wits Rural Public Health & Health Transitions Research Unit (Agincourt), School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa, 14 Developmental Neurosciences, University College London, UCL- Great Ormond Street Institute of Child Health, London, United Kingdom, 15 Department of Clinical & Experimental Epilepsy, UCL Queen Square Institute of Neurology, London, & Chalfont Centre for Epilepsy, Chalfont St Peter, United Kingdom, 16 Stichting Epilepsie Instellingen Nederland, Heemstede, Netherlands, 17 Department of Neurology, West China Hospital, Sichuan University, Chengdu, China, 18 Department of Psychiatry, University of Oxford, Oxford, United Kingdom, 19 The Alan Turing Institute, London, United Kingdom.

Reference

  1. 1. Duta I, Kariuki SM, Ngugi AK, Mwesige AK, Masanja H, Mwanga DM, et al. Evaluating the generalisability of region-naïve machine learning algorithms for the identification of epilepsy in low-resource settings. PLOS Digit Health. 2025;4(2): e0000491. pmid:39937713