TY - JOUR
T1 - Development and validation of a diagnostic aid for convulsive epilepsy in sub-Saharan Africa
T2 - a retrospective case-control study
AU - EPInA Study Group
AU - Jones, Gabriel Davis
AU - Kariuki, Symon M.
AU - Ngugi, Anthony K.
AU - Mwesige, Angelina Kakooza
AU - Masanja, Honorati
AU - Owusu-Agyei, Seth
AU - Wagner, Ryan
AU - Cross, J. Helen
AU - Sander, Josemir W.
AU - Newton, Charles R.
AU - Sen, Arjune
AU - Abban, Hanna
AU - Adjei, Patrick
AU - Ae-Ngibise, Ken
AU - Agbokey, Francis
AU - Aissaoui, Lisa
AU - Akpalu, Albert
AU - Akpalu, Bright
AU - Asiamah, Sabina
AU - Asiki, Gershim
AU - Atieno, Mercy
AU - Bauni, Evasius
AU - Bhwana, Dan
AU - Bitta, Mary
AU - Bottomley, Christian
AU - Chabi, Martin
AU - Chengo, Eddie
AU - Chowdhary, Neerja
AU - Connor, Myles
AU - Cross, Helen
AU - Collinson, Mark
AU - Darkwa, Emmanuel
AU - Denison, Timothy
AU - Doku, Victor
AU - Dua, Tarun
AU - Egesa, Isaac
AU - Godi, Tony
AU - Gómez-Olivé, F. Xavier
AU - Grassi, Simone
AU - Iddi, Samuel
AU - Junior, Daniel Nana Yaw Abankwah
AU - Kahn, Kathleen
AU - Kakooza, Angelina
AU - Kariuki, Symon
AU - Kamuyu, Gathoni
AU - Khalayi, Clarah
AU - Kimambo, Henrika
AU - Kleinschmidt, Immo
AU - Kwasa, Thomas
AU - Mahone, Sloan
N1 - Funding Information:
This work was supported by The Alan Turing Institute's Enrichment Scheme. JWS is based at the National Institute of Health Research (NIHR) University College London (UCL) Hospitals Biomedical Research Centre, which receives funding from the UK Department of Health's Research Centres funding scheme. JWS receives research support from the Dr Marvin Weil Epilepsy Research Fund and from the UK Epilepsy Society and the Christelijke Vereniging voor de Verpleging van Lijders aan Epilepsie, Netherlands. JWS or his department have also received grants from Esai, Angelina Pharma, and UCB Pharma. The UK Epilepsy Society endows the current position of JWS and he serves as Editorial Board member for The Lancet Neurology. JHC's research is supported by the NIHR Biomedical Research Centre at Great Ormond Street Hospital. JHC holds an endowed chair at UCL Great Ormond Street Institute of Child Health, has received grants from the NIHR, Engineering and Physical Sciences Research Council, Great Ormond Street Hospital Children's Charity, Epilepsy Research UK, the Waterloo Foundation, and the Great Ormond Street Hospital Biomedical Research Centre. AS is supported by the Oxford Biomedical Research Centre. This work is funded by the NIHR and the Oxford Martin Programme on Global Epilepsy. This research was commissioned by the National Institute for Health Research using Official Development Assistance funding. The views expressed in this publications are those of the authors and not necessarily those of the UK National Health Service, the NIHR, nor the Department of Health and Social Care. The Wellcome Trust, UK, funded the SEEDS study through a Wellcome Trust Senior Fellowship in Clinical Tropical Medicine (083744) and the EPInA project is funded by the UK NIHR. This research was commissioned by the National Institute for Health Research (NIHR200134) using Official Development Assistance funding. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the department of Health and Social Care.
Funding Information:
This work was supported by The Alan Turing Institute's Enrichment Scheme. JWS is based at the National Institute of Health Research (NIHR) University College London (UCL) Hospitals Biomedical Research Centre, which receives funding from the UK Department of Health's Research Centres funding scheme. JWS receives research support from the Dr Marvin Weil Epilepsy Research Fund and from the UK Epilepsy Society and the Christelijke Vereniging voor de Verpleging van Lijders aan Epilepsie, Netherlands. JWS or his department have also received grants from Esai, Angelina Pharma, and UCB Pharma. The UK Epilepsy Society endows the current position of JWS and he serves as Editorial Board member for The Lancet Neurology. JHC's research is supported by the NIHR Biomedical Research Centre at Great Ormond Street Hospital. JHC holds an endowed chair at UCL Great Ormond Street Institute of Child Health, has received grants from the NIHR, Engineering and Physical Sciences Research Council, Great Ormond Street Hospital Children's Charity, Epilepsy Research UK, the Waterloo Foundation, and the Great Ormond Street Hospital Biomedical Research Centre. AS is supported by the Oxford Biomedical Research Centre. This work is funded by the NIHR and the Oxford Martin Programme on Global Epilepsy. This research was commissioned by the National Institute for Health Research using Official Development Assistance funding. The views expressed in this publications are those of the authors and not necessarily those of the UK National Health Service, the NIHR, nor the Department of Health and Social Care. The Wellcome Trust, UK, funded the SEEDS study through a Wellcome Trust Senior Fellowship in Clinical Tropical Medicine (083744) and the EPInA project is funded by the UK NIHR. This research was commissioned by the National Institute for Health Research (NIHR200134) using Official Development Assistance funding. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research or the department of Health and Social Care.
Publisher Copyright:
© 2023 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license
PY - 2023/4
Y1 - 2023/4
N2 - Background: Identification of convulsive epilepsy in sub-Saharan Africa relies on access to resources that are often unavailable. Infrastructure and resource requirements can further complicate case verification. Using machine-learning techniques, we have developed and tested a region-specific questionnaire panel and predictive model to identify people who have had a convulsive seizure. These findings have been implemented into a free app for health-care workers in Kenya, Uganda, Ghana, Tanzania, and South Africa. Methods: In this retrospective case-control study, we used data from the Studies of the Epidemiology of Epilepsy in Demographic Sites in Kenya, Uganda, Ghana, Tanzania, and South Africa. We randomly split these individuals using a 7:3 ratio into a training dataset and a validation dataset. We used information gain and correlation-based feature selection to identify eight binary features to predict convulsive seizures. We then assessed several machine-learning algorithms to create a multivariate prediction model. We validated the best-performing model with the internal dataset and a prospectively collected external-validation dataset. We additionally evaluated a leave-one-site-out model (LOSO), in which the model was trained on data from all sites except one that, in turn, formed the validation dataset. We used these features to develop a questionnaire-based predictive panel that we implemented into a multilingual app (the Epilepsy Diagnostic Companion) for health-care workers in each geographical region. Findings: We analysed epilepsy-specific data from 4097 people, of whom 1985 (48·5%) had convulsive epilepsy, and 2112 were controls. From 170 clinical variables, we initially identified 20 candidate predictor features. Eight features were removed, six because of negligible information gain and two following review by a panel of qualified neurologists. Correlation-based feature selection identified eight variables that demonstrated predictive value; all were associated with an increased risk of an epileptic convulsion except one. The logistic regression, support vector, and naive Bayes models performed similarly, outperforming the decision-tree model. We chose the logistic regression model for its interpretability and implementability. The area under the receiver operator curve (AUC) was 0·92 (95% CI 0·91–0·94, sensitivity 85·0%, specificity 93·7%) in the internal-validation dataset and 0·95 (0·92–0·98, sensitivity 97·5%, specificity 82·4%) in the external-validation dataset. Similar results were observed for the LOSO model (AUC 0·94, 0·93–0·96, sensitivity 88·2%, specificity 95·3%). Interpretation: On the basis of these findings, we developed the Epilepsy Diagnostic Companion as a predictive model and app offering a validated culture-specific and region-specific solution to confirm the diagnosis of a convulsive epileptic seizure in people with suspected epilepsy. The questionnaire panel is simple and accessible for health-care workers without specialist knowledge to administer. This tool can be iteratively updated and could lead to earlier, more accurate diagnosis of seizures and improve care for people with epilepsy. Funding: The Wellcome Trust, the UK National Institute of Health Research, and the Oxford NIHR Biomedical Research Centre.
AB - Background: Identification of convulsive epilepsy in sub-Saharan Africa relies on access to resources that are often unavailable. Infrastructure and resource requirements can further complicate case verification. Using machine-learning techniques, we have developed and tested a region-specific questionnaire panel and predictive model to identify people who have had a convulsive seizure. These findings have been implemented into a free app for health-care workers in Kenya, Uganda, Ghana, Tanzania, and South Africa. Methods: In this retrospective case-control study, we used data from the Studies of the Epidemiology of Epilepsy in Demographic Sites in Kenya, Uganda, Ghana, Tanzania, and South Africa. We randomly split these individuals using a 7:3 ratio into a training dataset and a validation dataset. We used information gain and correlation-based feature selection to identify eight binary features to predict convulsive seizures. We then assessed several machine-learning algorithms to create a multivariate prediction model. We validated the best-performing model with the internal dataset and a prospectively collected external-validation dataset. We additionally evaluated a leave-one-site-out model (LOSO), in which the model was trained on data from all sites except one that, in turn, formed the validation dataset. We used these features to develop a questionnaire-based predictive panel that we implemented into a multilingual app (the Epilepsy Diagnostic Companion) for health-care workers in each geographical region. Findings: We analysed epilepsy-specific data from 4097 people, of whom 1985 (48·5%) had convulsive epilepsy, and 2112 were controls. From 170 clinical variables, we initially identified 20 candidate predictor features. Eight features were removed, six because of negligible information gain and two following review by a panel of qualified neurologists. Correlation-based feature selection identified eight variables that demonstrated predictive value; all were associated with an increased risk of an epileptic convulsion except one. The logistic regression, support vector, and naive Bayes models performed similarly, outperforming the decision-tree model. We chose the logistic regression model for its interpretability and implementability. The area under the receiver operator curve (AUC) was 0·92 (95% CI 0·91–0·94, sensitivity 85·0%, specificity 93·7%) in the internal-validation dataset and 0·95 (0·92–0·98, sensitivity 97·5%, specificity 82·4%) in the external-validation dataset. Similar results were observed for the LOSO model (AUC 0·94, 0·93–0·96, sensitivity 88·2%, specificity 95·3%). Interpretation: On the basis of these findings, we developed the Epilepsy Diagnostic Companion as a predictive model and app offering a validated culture-specific and region-specific solution to confirm the diagnosis of a convulsive epileptic seizure in people with suspected epilepsy. The questionnaire panel is simple and accessible for health-care workers without specialist knowledge to administer. This tool can be iteratively updated and could lead to earlier, more accurate diagnosis of seizures and improve care for people with epilepsy. Funding: The Wellcome Trust, the UK National Institute of Health Research, and the Oxford NIHR Biomedical Research Centre.
UR - http://www.scopus.com/inward/record.url?scp=85150846055&partnerID=8YFLogxK
U2 - 10.1016/S2589-7500(22)00255-2
DO - 10.1016/S2589-7500(22)00255-2
M3 - Article
C2 - 36963908
AN - SCOPUS:85150846055
SN - 2589-7500
VL - 5
SP - e185-e193
JO - The Lancet Digital Health
JF - The Lancet Digital Health
IS - 4
ER -