Annals of Indian Academy of Neurology · 2026
Deep Learning Detection of Facial Nerve Enhancement in Bell's Palsy.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
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.
Akkaya Habip Eser, Polat Önder, Hürsoy Nur, Akpınar Dorukan İnanç
Аннотация
BACKGROUND AND OBJECTIVES: The facial nerve, or seventh cranial nerve, with its complex course and association with multiple pathologies, remains a critical area of study. Bell's palsy, an idiopathic condition leading to unilateral facial weakness, is frequently diagnosed through magnetic resonance imaging (MRI), which can reveal inflammation or pathology within the facial nerve. However, physiological enhancement can complicate diagnosis, particularly in healthy individuals. This study aimed to assess facial nerve enhancement in Bell's palsy patients using machine learning algorithms.
METHODS: We retrospectively analyzed 306 MR images from subjects, including 156 Bell's palsy patients and 150 controls, and trained a DenseNet-based deep learning model to identify pathological enhancement with an 80% accuracy (area under the curve (AUC) 87%).
RESULTS: Our model achieved an accuracy of 80% and an AUC of 87% as compared to neuroradiologist's decision, demonstrating its potential to aid clinicians in diagnosis.
CONCLUSIONS: Future studies with external validation and larger datasets may further enhance the clinical applicability of AI-based assessment tools for facial nerve pathologies.
METHODS: We retrospectively analyzed 306 MR images from subjects, including 156 Bell's palsy patients and 150 controls, and trained a DenseNet-based deep learning model to identify pathological enhancement with an 80% accuracy (area under the curve (AUC) 87%).
RESULTS: Our model achieved an accuracy of 80% and an AUC of 87% as compared to neuroradiologist's decision, demonstrating its potential to aid clinicians in diagnosis.
CONCLUSIONS: Future studies with external validation and larger datasets may further enhance the clinical applicability of AI-based assessment tools for facial nerve pathologies.