中 / EN

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

Latest

October 2026
10-14 Papers Foundations & Methods

Researchers Detect Events and Their Boundaries in Continuous EEG

Researchers proposed NeuroDETR, a framework that identifies event classes and estimates their onsets and offsets directly from continuous EEG, without predefined event boundaries at inference time. It combines information across temporal scales and spatial relationships among EEG channels, using predicted event presence to guide feature processing. The authors report state-of-the-art performance on eight EEG datasets spanning diverse paradigms. These dataset experiments address event recognition for continuous interaction and monitoring; they do not directly establish performance in deployed brain-computer interface systems.
Why it matters The task goes beyond assigning a label to an EEG segment: it also requires locating when an event starts and ends. Not needing predefined boundaries at inference time distinguishes it from classification of presegmented EEG.
View details →
10-14 Papers Foundations & Methods

Imagined Speech Decoders Perform Near Chance Across People in Three EEG Corpora

Researchers analyzed three public corpora to test whether imagined speech decoders trained on one person's scalp EEG could work for another. Across 482 ordered participant pairs, none remained significant after multiple-comparison correction; mean transfer accuracy was within 0.006 of chance in every corpus. A signal distinguishing the imagery task from pre-stimulus rest instead reflected a post-stimulus state. Only the competition corpus exceeded the within-subject threshold for every participant, but its frequency profile failed an artifact audit. The findings apply to these corpora and the representations examined, not to all imagined speech decoding.
Why it matters Recognizing a task state is not the same as recognizing imagined words. The decoder also frequently labeled the interval before imagery as imagery, highlighting the need to distinguish task-state signals from speech-content signals.
View details →
10-14 Papers Foundations & Methods

Study Finds Muscle Signals Dominate Overt-Speech Controls in EEG Decoding

Researchers audited three public corpora and found that decoding overt speech may not establish a pipeline's ability to recognize brain-derived speech signals. In ds003626, two lip electrodes achieved an accuracy of 0.7430, compared with 0.4231 for the 128-channel scalp montage; no decodable signal remained in the 8-30 Hz band. By injecting effects of known amplitude into real recordings, the team also measured detection floors and found differences between methods. The authors conclude that muscle signals dominate the overt-speech positive controls in these corpora and recommend reporting band-limited performance, artifact-channel accuracy, and measured detection floors.
Why it matters A null result needs to be interpreted alongside a method's sensitivity. Injecting effects of known amplitude into real recordings tests how small a change the pipeline can detect, a different question from whether it can classify overt speech.
View details →
10-13 · Huazhong University of Science and Technology China Papers Foundations & Methods

Review Maps Functional Ultrasound BCI Applications and Highlights Response Delays

A review examines the principles of functional ultrasound brain-computer interfaces and their research uses and potential clinical value in stroke assessment, epilepsy monitoring, and early identification of neurodegenerative diseases. Functional ultrasound detects cerebral blood volume changes through the relationship between neural activity and blood flow, with a spatial resolution of approximately 100 μm. Although imaging frames are acquired at intervals of 200–500 ms, the hemodynamic response limits effective temporal resolution to 1–2 s. The author notes that bidirectional closed-loop systems combining acquisition and neuromodulation remain at the proof-of-concept stage. The biosafety of chronic ultrasound exposure is still unclear, and decoding reliability under pathological conditions needs improvement.
Why it matters The review distinguishes how frequently images are acquired from how quickly blood flow reflects neural activity. This distinction helps explain why faster image acquisition does not necessarily provide equally fast brain-signal feedback.
View details →
10-13 United States Clinical Launch & Registration

Registered Study Will Compare BCI Communication With Eye Tracking in ALS

A registered study that is not yet recruiting will compare a wearable brain-computer interface communication device with the eye-tracking devices already used by people with ALS. It plans to enroll 30 adults who use Tobii Dynavox or Smartbox eye-tracking devices. Each participant will try three options at home over 12 weeks: their own eye-tracking device and Cognixion ONE with either a head-worn augmented-reality display or a screen-based display. The main measure is the time participants voluntarily choose to use each option. Additional measures include communication speed, accuracy, and caregiver burden. No results are available. Cognixion ONE is investigational and is not FDA-approved or cleared for any medical use.
Why it matters The study prioritizes voluntary use at home rather than communication speed alone. Some sessions assign an option, while others let participants choose, allowing researchers to examine which setup people prefer to spend time using. Whether that preference translates into practical benefit remains to be tested.
View details →
10-13 · Arbor Neuroscience United States Media Coverage

Arbor Neuroscience Plans Brain Maps to Link Symptoms With Neuromodulation Targets

Arbor Neuroscience plans to develop brain maps that track signal changes after neuromodulation and connect specific symptoms with neural circuits. Its goal is to guide treatment targeting by symptoms rather than diagnostic categories alone. Its predecessor, Forest Neurotech, developed Forest 17, a wearable functional ultrasound imaging device that observes brain activity through localized blood flow changes. A future version is intended to add focused ultrasound neuromodulation. These are research instruments for people who already have ultrasound-transparent skull-replacement implants, not devices that anyone can simply put on and use. Symptom-to-circuit mapping and treatment planning based on it remain organizational goals.
Why it matters Wearable does not mean usable without specific surgical conditions: these research instruments depend on an ultrasound-transparent skull-replacement implant. That requirement separates the device's form factor from its eligible users and proposed therapeutic applications.
View details →
10-13 Industry News

Muse Plans Weekly Brain Age Estimates Using EEG and Cognitive Tests

Muse announced Brain Age, a feature combining headband-recorded EEG with four one-time cognitive tests. It will show whether estimated brain age is younger than, aligned with, or older than chronological age, expressed as a difference in years. According to the company, the first reading requires 10 qualifying minutes on each of 7 different days and completion of all four tests, with updates every Monday thereafter. The feature supports Muse 2, Muse S, and Muse S Athena. The first reading is free; subsequent readings require Muse Premium. Muse describes it as a general wellness feature whose readings are not for medical use; the company does not promise that practice changes brain age.
Why it matters Recording EEG directly does not mean directly measuring a definitive brain age. The output is a model estimate, and Muse says cognitive test scores can be affected by practice. The company presents cognitive performance separately from EEG readings to distinguish which input changed.
View details →
10-13 · Beijing Sport University China Papers Foundations & Methods

Beijing Sport University Model Gains Lose Significance Under Broader Statistical Correction

Researchers at Beijing Sport University developed a motor imagery EEG decoding model and evaluated it offline on three public datasets using subject-specific and cross-subject protocols. Against the baseline with the highest mean accuracy, subject-specific gains on two datasets were 3.56 and 1.62 percentage points; another cross-subject comparison showed a 2.08-point difference. These three comparisons had statistical support after correction within their respective settings, but none remained significant under a broader correction covering 56 comparisons. The findings offer limited support for integrating local and longer-range EEG information in stages, rather than establishing benefits for online control or clinical use.
Why it matters Performance for a new user is distinct from performance after person-specific training. The study used leave-one-subject-out evaluation, holding one participant out for testing while training on the others, to examine cross-subject applicability alongside individualized results.
View details →
10-12 Papers Foundations & Methods

Stage-Aware EEG Window Selection Improves Motor Imagery Decoding in Benchmark Tests

Researchers propose an adaptive EEG window-selection framework for motor imagery classification. Instead of independently retaining or discarding each segment, it considers event phases and relationships between successive windows. Each window receives a Discard, Onset, Sustain, or Offset role, with constraints blocking selections that violate phase order. The authors report consistent decoding improvements across multiple benchmark datasets in both cross-session and cross-subject settings. Visualizations further suggest that the learned policy aligns with neurophysiological principles. These dataset experiments do not establish equivalent gains during online BCI use. This study is a preprint and has not been peer reviewed.
Why it matters The method expands window selection from deciding which segment to keep to identifying its role within an event. Sequence constraints aim to reduce fragmented selections, while compatibility with different motor imagery classifiers allows the approach to target window selection rather than replace the entire classification system.
View details →
10-11 Papers Foundations & Methods

EEG Transform Reduces Cross-Session Identity Recognition at a Cost to Motor Imagery Decoding

Researchers developed an EEG transform that requires no trained privacy model to reduce features linking recordings from the same person across sessions. Tests on two public datasets used five attackers spanning four covariance-space representations. The strongest attacker's accuracy fell from 90.8% to 15.6% on BCI Competition IV-2a and from 90.8% to 9.9% on OpenBMI. Relative to the 8-30 Hz input, motor imagery decoding accuracy decreased by 7.0 and 2.1 percentage points, respectively. The transform was fitted on one session and applied unchanged to a held-out later session. This study is a preprint and has not been peer reviewed.
Why it matters The study tests whether privacy processing fitted on one recording session carries over to a later session, rather than only working within the same batch of data. Ablation tests also isolate the main contributors: mean alignment and scatter whitening.
View details →

© 2026 BCIwiki.com Digest Topics Tips Subscribe Revisions About