Artificial Intelligence for Improving the Management of People With Epilepsy in Low-and-Middle Income Countries.
Аннотация
BACKGROUND
About 80% of people with epilepsy live in low-and-middle-income countries (LMICs) where the treatment gap is high. Limited access to neurologists, diagnostic tools, and antiseizure medications, combined with persistent stigma, contribute to poor outcomes, including premature mortality. Artificial intelligence (AI) offers potential to address these gaps through scalable, low-cost solutions for diagnosis, investigation, management, and monitoring.
METHODS
This review examines three recent and complementary AI applications in epilepsy care for LMICs: a smartphone-based diagnostic tool for convulsive epilepsy developed using population-based data from five sub-Saharan African countries; an automated EEG interpretation system based on a deep learning model (SCORE-AI) validated across multicenter datasets; and a wearable-based deep learning model for detecting generalized convulsive seizures using low-cost smartwatches.
RESULTS
The smartphone diagnostic tool achieved area under the curve (AUC) 0.92-0.95 with sensitivity 85.0%-97.5% for identifying epilepsy with convulsive seizures using eight binary clinical features. SCORE-AI demonstrated expert-level performance (AUC 0.89-0.96, accuracy 85%-92%) for automated EEG classification across multiple validation datasets. The wearable seizure detection algorithm achieved 96% sensitivity with approximately one false alarm per 8 days. All three solutions were designed for deployment on widely accessible platforms.
CONCLUSIONS
AI-driven approaches demonstrate feasibility for addressing diagnostic and monitoring gaps in resource-limited settings. However, implementation faces substantial challenges including infrastructure constraints, limited digital literacy, ethical considerations, and sociocultural factors. Successful deployment requires validation with large locally relevant datasets, context-adapted solutions, task-sharing strategies, implementation research, appropriate regulatory frameworks/certification, and community engagement to reduce the global epilepsy care gap.
Предметные рубрики MeSH
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.