World neurosurgery · 2025
Automated Detection and Localization of Focal Cortical Dysplasia Type II: The Optimized Detection Framework (ODF).
СтатьяПоддержка NIH: внешняя программаНазвание и аннотация: Перевод в очереди
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.
Ganji Zohreh, Hakkak Mohsen Aghaee, Zare Hoda
Аннотация
OBJECTIVE: Focal cortical dysplasia (FCD) type II is a significant contributor to drug-resistant epilepsy. The accurate identification of lesions via magnetic resonance imaging (MRI) presents considerable challenges, often impeding effective surgical planning. This study aims to develop an automated detection framework, the optimized detection framework (ODF), to enhance the diagnosis of FCD type II.
METHODS: The ODF merges surface-based morphometry with machine learning techniques. A dataset comprising 58 MRI scans (30 FCD type II patients, 28 controls) underwent preprocessing (bias field correction, skull stripping), processing (segmentation), and feature extraction of both morphological and intensity-based metrics. Three classification algorithms-artificial neural network, decision tree, and support vector machine-were evaluated for their effectiveness in lesion detection and localization.
RESULTS: The artificial neural network classifier, part of the ODF, exhibited superior performance with an overall accuracy of 98.6%, attaining sensitivity of 97.5% and specificity of 100% for lesion detection. The localization accuracy for lesions was 84.2% for hemispheric and 77.3% for lobar classification.
CONCLUSIONS: The ODF represents a significant advancement in the automated detection and localization of FCD type II lesions. Its high precision and efficiency support presurgical evaluations, particularly in MRI-negative cases, and may optimize epilepsy management. Prospective integration with intraoperative navigation systems could enhance surgical outcomes.
METHODS: The ODF merges surface-based morphometry with machine learning techniques. A dataset comprising 58 MRI scans (30 FCD type II patients, 28 controls) underwent preprocessing (bias field correction, skull stripping), processing (segmentation), and feature extraction of both morphological and intensity-based metrics. Three classification algorithms-artificial neural network, decision tree, and support vector machine-were evaluated for their effectiveness in lesion detection and localization.
RESULTS: The artificial neural network classifier, part of the ODF, exhibited superior performance with an overall accuracy of 98.6%, attaining sensitivity of 97.5% and specificity of 100% for lesion detection. The localization accuracy for lesions was 84.2% for hemispheric and 77.3% for lobar classification.
CONCLUSIONS: The ODF represents a significant advancement in the automated detection and localization of FCD type II lesions. Its high precision and efficiency support presurgical evaluations, particularly in MRI-negative cases, and may optimize epilepsy management. Prospective integration with intraoperative navigation systems could enhance surgical outcomes.