Multi-centre Epilepsy Lesion Detection

The Multi-centre Epilepsy Lesion Detection (MELD) Project aimed to develop an open-source, interpretable machine-learning algorithm to identify subtle focal cortical dysplasias (FCDs) in structural MRI scans, addressing the challenge of detecting these lesions which are often difficult to visualize. The algorithm utilizes surface-based features to enhance the detection and interpretation of FCDs from a diverse dataset collected from multiple epilepsy centers. The goal is to improve diagnostic accuracy and confidence in identifying subtle lesions in patients with epilepsy.