MELD Project

Multi-centre Epilepsy Lesion Detection (MELD) Project

About the MELD project

Objectives

Epilepsy is one of the most common neurological conditions where 20-30% of patients will not be controlled with medications. In many patients, the seizures are caused by a focal cerebral lesion, and surgery is a safe and effective treatment. Accurate detection of lesions on MRI and complete neurosurgical resection are important for surgery to be successful. 

The development of machine-learning algorithms to detect these lesions requires large multi-centre efforts. We collaborate with over 25 hospitals internationally!

The overall aim of the MELD project is to improve the outcomes of patients with drug-resistant, surgically-remediable epilepsy through:

1.    Collation of a large multicentre cohort of clinical and MRI data from patients with focal epilepsy

2.    Creation of atlases of the spatial locations of focal epilepsies

3.    Development of deep learning algorithms for automated lesion detection

4.    Development of deep learning algorithms to automatically identify the focal epilepsy pathology

5.    Generation of models for predicting post-surgical seizure freedom 

6.    Translation of algorithms into clinical practice