BCIwiki — Brain-computer interface news, research and industry database

Onward Medical said the first participant in its global Empower BP pivotal study was enrolled and implanted with the investigational ARC-IM system at Craig Hospital in Denver during the first quarter of 2026. The randomized, double-blind, sham-controlled trial is evaluating management of blood-pressure instability after spinal cord injury across about 20 planned centers. Empower BP is not an upper-limb BCI trial, and the first implant did not take place in the Netherlands.

The Jiangxi Provincial Health Commission in eastern China issued Guidelines for Ethics Review of Brain-Computer Interface Clinical Research Projects on May 26, 2026, document number Gan Wei Ke Jiao Fa [2026] No. 8, providing implementing detail for the mutual recognition mechanism proposed in the province's November 2025 BCI measures. For multicentre projects the lead institution's ethics committee conducts the first full review; participating institutions should finish within 10 working days of receiving complete materials and examine only local research conditions, staff qualifications and the informed-consent process; and the lead institution must file with the provincial medical ethics expert committee within 10 working days. Invasive projects treating neurological or psychiatric disease need provincial committee review within 20 working days, with an opinion issued within a further seven.

Can a non-invasive brain-computer interface read motor intention and, combined with spinal cord stimulation and exoskeleton training, improve function after spinal cord injury? Xuanwu Hospital, Capital Medical University has registered a single-arm exploratory trial that plans to enroll 10 patients aged 18 to 65 with spinal cord injury. The intervention combines all three into one rehabilitation regimen. The primary outcome is the overall clinical efficacy rate 6 months after surgery; secondary outcomes include improvement in the Visual Analogue Scale for pain and the incidence of adverse events related to the trial devices.

The multicenter randomized controlled trial plans to enroll 222 patients who are 1 to 3 months past their first stroke and have motor impairment in the affected lower limb. The experimental group will receive 4 weeks of motor imagery-based BCI-driven active and passive lower limb training, while the control group gets 4 weeks of conventional training. The primary outcome is the Fugl-Meyer Assessment Lower Extremity Section, with balance, daily living ability and walking classification as secondary outcomes. First enrolment is set for May 31, 2026.

The multicenter randomized controlled trial pits closed-loop BCI ankle training against conventional active and passive ankle training in subacute stroke patients who are 4 to 12 weeks past onset and have less than 10 degrees of active ankle dorsiflexion. Primary endpoints are active ankle dorsiflexion range of motion and Manual Muscle Testing of the ankle; secondary endpoints include Functional Ambulation Classification, the Berg Balance Scale and the Fugl-Meyer Motor Assessment of the Lower Limb. The trial plans to enroll 120 patients, 60 per arm, with first enrolment set for May 31, 2026.

The study introduces RLED, a reinforcement learning-enhanced EEG diffusion framework for adaptive data augmentation in endogenous EEG tasks such as motor imagery and emotion recognition. Reinforcement learning dynamically regulates the diffusion training process to balance temporal, spectral and category-related features. Across four datasets, the high-quality synthetic EEG signals it generated consistently improved classification performance.

A Chinese team has registered a clinical trial of a multimodal brain-computer interface system for intelligent diagnosis and treatment, aimed at early screening and non-invasive modulation of Alzheimer's disease. The entry, ChiCTR2600125240, was filed with the Chinese Clinical Trial Registry on May 22, 2026 and is carried by the WHO International Clinical Trials Registry Platform; recruitment has not begun.

The trial tests whether a steady-state visual evoked potential (SSVEP) brain-computer interface can drive functional electrical stimulation of swallowing muscles more effectively than conventional neuromuscular electrical stimulation. It plans to enroll 90 patients, 45 per arm, all within 6 months of a first ischemic or hemorrhagic stroke and aged 18 to 80. The primary outcome is the Standard Swallowing Assessment, with secondary measures including the Functional Oral Intake Scale and the Penetration-Aspiration Scale. First enrollment is listed for June 1, 2026.

Burning mouth syndrome produces months or years of burning oral pain with no visible lesion on examination, and treatment options are limited, leaving patients shuttling between oral medicine and psychiatry. A randomized, sham-controlled interventional trial registered by Shanghai Ninth People's Hospital, Shanghai JiaoTong University School of Medicine, will apply transcranial direct current stimulation over the left dorsolateral prefrontal cortex while collecting resting-state brain function and structural imaging, tracking pain, anxiety, depression, sleep and salivary flow. The trial is funded by the hospital's Brain-Computer Interface Special Research Program and registered as ChiCTR2600124987 on May 20, 2026. It plans to enroll 38 participants, 19 in the active group and 19 in the sham group, with first enrollment set for June 1, 2026; recruitment has not yet started.

Researchers have built a hybrid batch-stream framework for EEG-based emotion recognition, which until now has had to choose between accurate but slow offline batch processing and fast but less accurate real-time streaming. The system estimates whether the streaming module is currently reliable from its past performance, then decides probabilistically whether to use the batch or the streaming output. It achieved classification accuracies of 85%, 94% and 74% on the public DEAP, AMIGOS and SEED benchmarks, and was also tested under concept drift, when data distributions shift over time. Results come from offline evaluation on public datasets and have not been validated with real users.