Every year, 120,000 people in the UK with drug-resistant focal epilepsy face a stark choice: invasive brain surgery or life-threatening seizures. For 30% of them, standard 3T MRI scans fail to locate the tiny brain lesion causing the seizures, leaving surgery impossible without a risky, £30,000 procedure that implants electrodes directly into the brain for two weeks. This project tests whether a new type of ultra-high-field 7T MRI—called parallel transmit (7T-pTx)—can replace that invasive test. The researchers have already shown in a pilot study that 7T-pTx found hidden lesions in 29% of patients whose standard scans were inconclusive. Some of those patients then went straight to curative surgery. Now, across two NHS sites, the team will scan 100 patients who would otherwise need electrode implantation. They will measure how often 7T-pTx identifies new lesions, whether it changes clinical decisions, and whether the findings are confirmed when removed tissue is examined by pathologists. An AI tool will help radiologists spot subtle abnormalities in the high-resolution images. If successful, 7T-pTx could fast-track epilepsy surgery for nearly a third of patients currently stuck in diagnostic limbo, saving the NHS tens of thousands of pounds per patient while giving people back their working and social lives.
View original technical description
Context: Uncontrolled epilepsy is a major health burden. In the UK, 120,000 people have drug-resistant focal epilepsy (DRFE), which causes unpredictable and life-threatening seizures (Wigglesworth, 2023). Surgery is the only curative treatment but will only stop seizures if it removes or isolates the diseased tissue driving the seizures – called the “epilepsy lesion”. Current NHS practice uses “3T MRI” and “FDG-PET” scans to find epilepsy lesions, but in 30% of patients the scans do not show the location of the epilepsy lesion clearly enough to justify surgery. At present, the remaining option is to surgically implant electrodes and record brain activity for 2 weeks (SEEG). This costs ~£30,000, is uncomfortable, and also carries a risk of complications. Ultra-high field (7T) MRI is more sensitive to detect epilepsy lesions (Opheim 2021), but so far, it has only been implemented in a few single-site studies using an older form of 7T MRI called single transmit (7T-sTx) which means that the temporal lobes - a key site for epilepsy lesions – are frequently obscured by signal-dropouts. We are leaders in parallel transmit (7T-pTx) technology, which images the whole brain with exceptional sensitivity and sub-millimetre resolution, including the temporal lobes. We have developed a 7T-pTx protocol for epilepsy. In a pilot study in 31 DRFE patients, we found epilepsy lesions in 29% of cases when normal scans had been inconclusive. Some patients were referred for surgery without further tests, while for others the 7T-pTx scan showed that surgery was not possible, avoiding further invasive tests and focusing scarce SEEG capacity for other patients. Challenge: This project addresses the challenge of establishing a clinically-viable and cost-effective alternative to scarce SEEG testing so that DRFE patients can access life-changing epilepsy surgery. Aims: We propose a two-site, four-year prospective study to prove that 7T-pTx MRI is ready for NHS implementation in cost-effective epilepsy surgery pathways. Our project has three work packages: First, we will harmonise 7T-pTx MRI protocols and demonstrate reproducibility across sites. Second, we will recruit 100 patients for 7T-pTx MRI whose clinical 3T MRI and FDG-PET scans have not provided sufficient information to proceed to surgery and who would otherwise need SEEG. We will measure the number of epilepsy lesions newly identified by consultant neuroradiologists based on 7T-pTx MRI as our primary outcome measure. We will measure the percentage of cases where 7T-pTx scans alter the clinical team’s recommendations (e.g. towards surgery instead of SEEG) as a secondary outcome. For patients who have surgery, pathologists will check the removed tissue to confirm whether the 7T-pTx scan findings were correct. We will measure the health benefits to patients and economic benefits to the NHS. Third, we will implement an AI-powered workflow for automatically processing 7T-pTx images by updating a tool called “MELD” for use on 7T-pTx scans. This will assist neuroradiologists to search through the many images produced by a 7T-pTx scan looking for small abnormalities by giving them hints on where to look. We will combine this with complementary information about brain metabolism from PET scans. Potential benefit: By establishing 7T-pTx MRI in standard clinical practice, we expect to fast-track curative neurosurgery in 29% of patients with otherwise untreatable focal epilepsy, without needing invasive SEEG. This will accrue significant health economic benefits by letting these often younger patients have a productive working and social life.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know