Using new techniques and artificial intelligence to predict patient outcomes following surgery
Scientific Title: Hippocampal shape analysis: a novel practical predictive biomarker of surgical outcome?
Lead investigator: Dr Marian Galovic, University College London
Co-investigators: Dr Matthias Koepp, UCL
Pilot grant: £29,744, 15 months
“Epilepsy is frequently treated with a one-size-fits-all approach. We hope that analysing the hippocampus in detail using shape analysis will reveal why some people are more likely to fail epilepsy surgery or to develop cognitive deficits and will allow us to develop tailor-made treatments. We aim to automate this method and make it available to other clinicians and researchers in future.” Dr Marian Galovic, University College London
An area of the brain, called the hippocampus, is most frequently affected in people with epilepsy. The hippocampus is typically evaluated by measuring its volume, but this does not identify which part of the hippocampus is most affected by epilepsy.
Dr Galovic and Professor Koepp will use a new technique called hippocampal shape analysis to analyse the shape and appearance of the hippocampus in high detail. They aim to determine, whether distinct shape patterns of the hippocampus cause different types of epilepsy, and whether the shape of the hippocampus can predict if someone will benefit from epilepsy surgery or if they will develop cognitive problems as a result. The team will use available data from a large dataset to develop an automated pipeline to analyse the hippocampus and to generate reports on its appearance. They will use artificial intelligence (AI) to predict the likelihood of becoming seizure-free or of having memory problems after surgery.
This knowledge will be important in improving our understanding of the hippocampus and to selecting optimal candidates for epilepsy surgery. Upon completion, they will also make the AI-pipeline freely available online and it can be used for evaluation of people with epilepsy.