Decoding neural responses to sounds in patients with disorders of consciousness for diagnostics and unlocking communication

Command following (i.e., the ability to voluntarily execute a simple command) is an important criteria in the diagnosis of patients with disorders of consciousness (DOC). Indeed, it is assessed clinically through the use of behavioural scales such as the Coma Recovery Scale Revised (CRS-R) and allows the distinction between patients with unresponsive wakefulness syndrom (UWS) and patients in a minimally conscious state (MCS). However, DOC patients may have limited neuromuscular capacity, which will alter some behavioural responses but, lack of behavioural response does not necessarily mean lack of consciousness. Moreover, the behavioural assessment of these patients may also be altered by the subjectivity of the examiners. It is therefore essential to develop objective and motor-independent diagnostic and communication tools for these patients. This would improve rehabilitation strategies, quality of life and prognosis for this sensitive population. Brain-computer interface technology based on electroencephalography (BCI-EEG) is a method used in conscious but severely motor-disabled patients to allow them to interact with the environment regardless of their motor abilities. Recent studies have focused on the use of BCI technology in DOC patients, but currently it is still difficult to transpose this technology to patients with severe brain injuries. Indeed, these electrophysiological (and behavioural) assessments are based on instructions that must be understood to succeed in the task. However, little is known about each patient's ability to hear, perceive and understand the instructions given. Furthermore, the patient's motivation and level of vigilance will also influence their ability to perform the requested task. Failure to take into account the clinical reality of these patients, which can vary considerably from one person to another, would lead to a misdiagnosis of their level of consciousness. The aim of this project is to develop a new, objective and motor-independent diagnostic tool based on brain-computer interface technology coupled to the EEG. This tool will integrate information concerning the patient's ability to hear, perceive and understand instructions, and will adapt to non-stationarities in the state of the patient (e.g., fluctuation of vigilance, motivation, and alertness) and signal (e.g., signal-to-noise, artifact). The overall goal of this project being to better understand and manage these patients.