Clinical utility of in-ear EEG electrodes in hospitalized patients: comparison with standard scalp EEG.

Objective: Clinical electroencephalography (EEG) is resource-intensive, requiring expensive equipment and trained specialists to attach scalp electrodes. Consequently, there is interest in alternative EEG technologies that could provide low-cost, user-friendly, and unobtrusive recording in clinical settings. Here, we aimed to evaluate the clinical utility of in-ear EEG electrodes in hospitalized patients through direct comparison with routine scalp EEG. Methods: For thirty routine adult in-patient clinical EEG recordings at Charing Cross Hospital (London, United Kingdom), participants were recruited and had in-ear EEG electrodes placed alongside standard scalp electrodes. Recordings from ear and scalp electrodes were independently and blindly interpreted by a consultant clinical neurophysiologist using a bespoke assessment protocol. Scalp EEG served as the gold-standard reference. Results: Compared with scalp EEG, interpretation of ear EEG achieved a sensitivity of 83.3%, specificity of 80.0%, and accuracy of 82.8% for the detection of any neurophysiological abnormality. There was substantial agreement between ear and scalp EEG for encephalopathy grading (quadratic weighted kappa = 0.72). One left posterior quadrant seizure, present in scalp EEG, was not detected by ear EEG interpretation alone. Conclusion: In-ear EEG demonstrated good agreement with conventional scalp EEG for the detection of clinically relevant abnormalities, with substantial agreement for encephalopathy severity grading. While reduced spatial coverage may limit sensitivity to focal phenomena, in-ear EEG appears well-suited to detecting diffuse cerebral dysfunction. Significance: This study paves the way for broader clinical development and validation of ear EEG-based devices as a pragmatic and scalable technology for use in acute hospital settings.