Multimodal Evaluation and Decoding of Impaired Consciousness (MEDIC)
Research type
Research Study
Full title
Multimodal Evaluation and Decoding of Brain Function in Disorders of Consciousness using Functional Neuroimaging and Electrophysiology
IRAS ID
361588
Contact name
Alexander L Green
Contact email
Sponsor organisation
Research Governance, Ethics and Assurance Team, University of Oxford
Duration of Study in the UK
3 years, 11 months, 31 days
Research summary
Advances in intensive care medicine mean that more people are surviving various brain injuries, including head injury and stroke, periods of low brain oxygenation. However, patients with severe injuries can develop disorders of consciousness (DoC), meaning that they are left with substantially reduced consciousness. These patients have inconsistent or absent awareness or interaction with their surroundings, leading to poor quality of life and dependence on healthcare and social services. Unfortunately, it is very difficult to predict if and when an individual patient will recover consciousness and no treatments have been proven to improve long-term outcomes.
Current assessment of consciousness level is challenging and inconsistent, involving observing behaviours that suggest awareness of their environment. Outside the UK, doctors use a brain imaging technology called functional magnetic resonance imaging (fMRI) and a technique that measures electrical activity in the brain ('brain waves') called electroencephalography (EEG). However, these are not currently recommended by UK guidelines.
To develop a more objective and reliable test for consciousness level, we have developed a ‘neural decoding’ approach based on fMRI and EEG. Neural decoding works by analysing patterns of brain activity (from fMRI and EEG) to figure out (or ‘decode’) what sounds or words someone is hearing. 20 DoC patients (unconscious for at least 4 weeks after a brain injury) and 20 healthy volunteers will have fMRI scans and EEG recordings. During these sessions, they will wear headphones and simply listen to different sounds, including beeps, speech from audiobook recordings, and distorted/mumbled speech. By comparing how closely an individual patient’s brain activity matches fully conscious people’s brain activity, we can measure how much they are consciously recognising and processing speech – without needing them to move, speak or follow commands.
This will be the first research study ever aiming to decode full speech from unconscious brain injury patients.
REC name
East of England - Essex Research Ethics Committee
REC reference
26/EE/0114
Date of REC Opinion
21 May 2026
REC opinion
Further Information Favourable Opinion