• Deep learning architectures fo...

Treffer: Deep learning architectures for modeling and forecasting stroke cases in Ghana.

Title:
Deep learning architectures for modeling and forecasting stroke cases in Ghana.
Authors:
Iddrisu AK; Department of Statistics, University of Botswana, Gaborone, Botswana. Electronic address: karim@aims.ac.za., Gabanakgosi M; Department of Statistics, University of Botswana, Gaborone, Botswana., Siddick AH; Department of Statistics and Actuarial Science, University of Ghana, Accra, Ghana.
Source:
Journal of stroke and cerebrovascular diseases : the official journal of National Stroke Association [J Stroke Cerebrovasc Dis] 2026 Mar; Vol. 35 (3), pp. 108570. Date of Electronic Publication: 2026 Jan 21.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Saunders Country of Publication: United States NLM ID: 9111633 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1532-8511 (Electronic) Linking ISSN: 10523057 NLM ISO Abbreviation: J Stroke Cerebrovasc Dis Subsets: MEDLINE
Imprint Name(s):
Publication: Philadelphia, PA : Saunders
Original Publication: New York, NY : Demos Publications, [1991-
MeSH Terms:
Stroke*/epidemiology , Stroke*/diagnosis , Deep Learning*, Ghana/epidemiology ; Diabetes Mellitus/epidemiology ; Humans ; Incidence ; Time Factors ; Forecasting ; Risk Factors ; Prevalence ; Bayes Theorem ; Risk Assessment
Contributed Indexing:
Keywords: Bayesian Long Short-Term Memory (BLSTM); Bayesian convolution LSTM(BConv); deep learning; stroke
Entry Date(s):
Date Created: 20260123 Date Completed: 20260220 Latest Revision: 20260220
Update Code:
20260221
DOI:
10.1016/j.jstrokecerebrovasdis.2026.108570
PMID:
41577309
Database:
MEDLINE

Weitere Informationen

Introduction: Stroke remains a leading cause of global morbidity and mortality, ranking second in deaths and third in disability-adjusted life years (DALYs). Its burden is particularly severe in low- and middle-income countries such as Ghana, where stroke is currently the leading cause of death. However, local data and predictive modeling remain limited, hindering effective health planning and intervention. This study aimed to model and forecast stroke incidence in Ghana using advanced deep learning techniques to support data-driven public health strategies.
Material and Methods: Monthly stroke case data from 2018 to 2023 were obtained from Ghana Health Service. Four deep learning models; Long Short-Term Memory (LSTM), Bayesian LSTM (BLSTM), Convolutional LSTM (ConvLSTM), and Bayesian ConvLSTM (BConvLSTM), were employed to capture spatiotemporal patterns in stroke incidence. Diabetes prevalence was included as a covariate. Model performance was evaluated using mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE). Data analyses were carried out using python version 3.13.7 and R version software.
Results: LSTM and BLSTM models showed strong forecasting performance, with LSTM yielding the lowest errors. ConvLSTM and BConvLSTM models underperformed significantly. Forecasts from 2024 to 2028 reveal initial variability in 2024, with monthly cases between 1,694 and 2,007, followed by gradual stabilization through 2028, where values converge between 1,774 and 1,781.
Conclusion: The study highlights a persistently high but stabilizing stroke burden in Ghana. It underscores the urgent need for targeted interventions addressing modifiable risk factors, particularly diabetes, and supports LSTM as the most effective model for forecasting in this context.
(Copyright © 2026. Published by Elsevier Inc.)

Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.