Epilepsy

A wireless implantable neural interface, the Brain Interchange (BIC), paired with artifact-removal algorithms captured about 82% of the high-frequency oscillations (HFOs) that a clinical-grade amplifier detected across 24-hour intracranial EEG recordings in 10 patients with drug-resistant epilepsy, according to a preprint. Using those HFOs, the system localized the seizure onset zone with 84% sensitivity and 90% specificity, comparable to clinical equipment. The authors present the work as a translational framework for chronic tracking of epileptogenic networks and for future biomarker-guided adaptive neuromodulation.

A multicenter trial has begun in China to test closed-loop brain-computer interface neuromodulation for treatment-resistant depression, run jointly by the Shanghai Mental Health Center and two Beijing hospitals, Anding Hospital and Xuanwu Hospital. The study will compare closed-loop deep brain stimulation and transcutaneous auricular vagus nerve stimulation against sham stimulation, enrolling patients who have responded inadequately to standard antidepressant treatment. It uses a patient-preference design with randomized assignment within each intervention group.

The US Food and Drug Administration cleared CN-Suite, source localization software from FIND Surgical Sciences Inc., doing business as FIND Neuro, on August 16, 2026, finding it substantially equivalent under submission number K260563 via the Traditional 510(k) pathway. FDA lists the device as Source Localization Software for Electroencephalograph or Magnetoencephalograph, a Class II device under regulation number 882.1400, reviewed by the agency's neurology advisory committee. Such software works back from electrical or magnetic fields recorded at the scalp to the intracranial origin of the signal, supporting presurgical planning such as locating epileptic foci.

The researchers developed a deep learning architecture that analyzes multichannel interictal intracranial EEG to localize epileptogenic zones without requiring seizure-period recordings or manual annotation. The model combines a Morlet-wavelet temporal Transformer with a spatial attention encoder. It was evaluated with 50.5 hours of recordings from 161 patients and 17,012 channels at 7 independent centers. Leave-one-center-out validation produced a pooled AUROC of 0.778 with a 95% confidence interval of 0.748 to 0.808, and discrimination remained above chance at every held-out center. The results indicate performance comparable to established electrophysiological baselines across centers and implantation modalities, but prospective clinical validation is still needed. This medRxiv preprint has not been peer reviewed.

Nova Southeastern University researchers combine intracranial ECoG network features with a support vector machine and Bayesian updating for real-time continuous seizure risk forecasting.

The FDA cleared the Ceribell Monitor on June 30, 2026, finding it substantially equivalent under submission number K261101 through the traditional 510(k) pathway. The agency classifies it as a Reduced-Montage Standard Electroencephalograph, a Class II device under regulation 882.1400, with neurology as the assigned review panel. It is one of several Ceribell clearances in the same period; the product line is built around rapid reduced-montage setup for bedside recording in emergency and critical care.

The FDA cleared the Ceribell Neurology Review Software on June 24, 2026, finding it substantially equivalent under submission number K260998 through the traditional 510(k) pathway. The agency classifies it as Automatic Event Detection Software for Full-Montage Electroencephalograph, a Class II device under regulation 882.1400, with neurology as the assigned review panel. The category covers software that automatically flags events within full-montage EEG recordings, surfacing candidate segments from long records for clinician review.

The FDA cleared Clouds of Care's PreOp v3 on February 13, 2026, finding it substantially equivalent under submission number K252565 through the special 510(k) pathway. The agency classifies it as Source Localization Software for Electroencephalograph or Magnetoencephalograph, a Class II device under regulation 882.1400, with neurology as the assigned review panel. Source localization software infers the intracranial origin of signals from scalp-recorded electric or magnetic fields, a step in neurosurgical planning such as presurgical localization of epileptic foci.

The FDA cleared the Ceribell Delirium Monitor System on December 8, 2025, finding it substantially equivalent under submission number K251936 through the traditional 510(k) pathway. The agency classifies it as a Neuropsychiatric Interpretative Electroencephalograph Assessment Aid, a Class II device under regulation 882.1440, with neurology as the assigned review panel. The category covers devices that derive neuropsychiatric assessments from EEG signals; Ceribell's earlier products in this line target seizure monitoring in emergency and intensive care settings.

The US Food and Drug Administration cleared Ceribell, Inc.'s Ceribell Infant Seizure Detection Software on November 21, 2025, finding it substantially equivalent under submission number K252070 via the Traditional 510(k) pathway. FDA lists it as Automatic Event Detection Software for Full-Montage Electroencephalograph, a Class II device under regulation number 882.1400, reviewed by the agency's neurology advisory committee. Neonatal EEG looks markedly different from adult morphology, which has long made automated interpretation difficult; the classification covers software that flags events automatically in full-montage recordings.

Chinese registry ChiCTR2400081380 describes a 96-patient trial of an implantable closed-loop recording and stimulation system for focal drug-resistant epilepsy in people aged 6 to 70. All participants are to receive the implant and then be randomized one month later to active or sham stimulation, 48 per group. The registry provides no evidence that enrollment had passed 50%.