Pith. sign in

REVIEW 4 major objections 5 minor 72 references

Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Saccade-related potentials are recordable and decodable through a blood-vessel brain implant in a human, with fixation versus saccade reaching a mean AUC of 0.88.

desk verdict First human endovascular oculomotor BCI proof-of-concept, but the headline onset-decoding AUC is likely inflated by the blocked task design. read the letter →

arxiv 2506.07481 v3 pith:F5EXRRSK submitted 2025-06-09 eess.SP q-bio.NC

Suleman Rasheed , James Bennett , Peter E. Yoo , Anthony N. Burkitt , David B. Grayden This is my paper
classification eess.SPq-bio.NC
keywords oculomotorbrain-computerinterfaceendovascularneuralStentrodesaccadedecodingevent-relatedpotentialsamyotrophiclateralsclerosisrandomforestclassificationfree-viewingsaccades
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper reports a proof-of-concept that a minimally invasive endovascular brain implant a stent-electrode array placed inside a blood vessel near the supplementary motor area can pick up neural signals tied to saccadic eye movements in a human with ALS. It aims to show that an oculomotor brain-computer interface is feasible without opening the skull, using self-paced free-viewing saccades to avoid the visual-cue confounds of earlier paradigms. The central result is that fixation can be separated from saccade with a mean AUC of 0.88 within sessions and 0.86 across sessions, while saccade direction is harder, at 0.67 for four classes and up to 0.75 for the best binary pairs. If the claim holds, endovascular interfaces could support eye-movement-based communication and control for people with severe paralysis, without craniotomy.

What carries the argument

The load-bearing setup is the Stentrode: a 16-electrode endovascular array implanted in the superior sagittal sinus near the supplementary motor area, which records broadband cortical signals without craniotomy. The analytic machinery combines independent component analysis to reveal source-space potentials, xDAWN spatial filtering plus time-frequency features for onset classification, and direct downsampled time series for direction classification, with leave-one-run-out block-wise cross-validation to address temporal correlation between trials.

What would settle it

Run the free-viewing task again with fixation and saccade trials interleaved in random order within each run, or take fixation trials from the gaps between saccades inside the saccade interval; if fixation-versus-saccade AUC drops toward 0.5 while the original blocked design still scores high, the onset-decoding claim is an artefact of block order rather than saccade-specific neural activity.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that saccade-related potentials appear in endovascular recordings from a human, beginning roughly 50 ms before eye movement, peaking about 50 ms after onset, and lasting about 200 ms, with a low-frequency synchronisation below 15 Hz. In the free-viewing task these potentials are consistent across directions and visible in single trials, whereas in the visually-guided task they overlap with a P300 evoked by the cue. Decoding fixation versus saccade is reliable within and across sessions, and decoding direction works above chance but remains limited; the pre-saccade component carries enough information that a classifier using only pre-movement data performs comparably to one using post-movement data.

Load-bearing premise

In the free-viewing task, all fixation trials were collected in a 15 s block at the start of each run and all saccade trials afterward, so the classifier might be separating two different time blocks rather than two brain states; the authors themselves flag this temporal separation as a possible reason discrimination was easier.

Editorial extensions

If this is right

  • A two-stage real-time oculomotor BCI, first detecting saccade onset and then classifying direction, is a plausible next step for endovascular interfaces.
  • Pre-saccade decoding at near-comparable accuracy suggests that covert or imagined saccades, usable by people without voluntary eye control, may be reachable with the same implant.
  • The free-viewing paradigm provides an asynchronous, cue-free control channel that avoids the P300 confound seen in visually-guided tasks.
  • Because visually-guided cues produce a clear P300 in Stentrode signals, the same implant could support a P300-based speller.
  • Cross-session direction decoding near chance indicates that generalisation across days remains an open problem for this implant.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The high onset-decoding AUC may partly reflect slow drifts in arousal, fatigue, impedance, or other state variables correlated with the 15 s fixation block at the start of each run, rather than saccade-specific activity; the paper's own design note invites this concern.
  • Direction decoding might improve with larger visual angles, since the current setup used about 5 degrees whereas earlier EEG studies found sharp accuracy gains at larger angles, or with implant sites closer to the frontal eye fields.
  • If the onset results survive an interleaved design where fixation and saccade trials are mixed within each run, an endovascular oculomotor BCI could become a faster, more intuitive complement to motor-imagery BCIs in assistive communication.
  • The broad low-frequency synchronisation below 15 Hz, rather than a narrow band, appears to carry the discriminative signal, which may guide the design of future online decoders.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper reports an offline proof-of-concept study of an oculomotor brain-computer interface using an endovascular Stentrode in a single participant with ALS. Two tasks were used: a visually guided saccade task, in which arrow cues produced P300 potentials that overlapped with initial-saccade responses, and a free-viewing task, in which the participant made self-paced saccades without external cues. The authors characterize saccade-related activity in the time, frequency, and time-frequency domains, and report classification results for saccade onset (fixation vs. saccade) and saccade direction. The main claims are that free-viewing saccades produce distinct saccade-related potentials beginning shortly before eye movement, that onset classification reaches AUC 0.88 within sessions and 0.86 across sessions, and that direction decoding reaches AUC 0.67 for four classes and up to 0.75 for selected binary comparisons.

Significance. If the results hold, this would be the first demonstration of an oculomotor BCI from an endovascular human implant, which is clinically relevant because the Stentrode does not require a craniotomy. The study has several strengths: it uses AUC-ROC rather than accuracy to handle class imbalance, it applies block-wise leave-one-run-out cross-validation to reduce temporal-correlation leakage between neighboring trials, it explicitly compares multiple classifiers, and it candidly acknowledges the P300 confound in the visually guided task and the blocked fixation/saccade structure in the free-viewing task. However, the central onset-classification claim is weakened by a design confound that the authors themselves identify in Section 3.2, and the direction-decoding results lack statistical uncertainty estimates. The paper is a valuable single-participant feasibility report, but the load-bearing quantitative claims need stronger support before they can be accepted.

major comments (4)
  1. [2.2.2, 2.5.2, 3.2, Figure 8] The Free-Viewing Task places all fixation trials in a 15 s block at the start of each run and all saccade trials in the subsequent 3-4 min block. Leave-one-run-out cross-validation keeps each run intact, so within every test run the fixation block always precedes the saccade block. The reported onset AUC values of 0.88 (within-session) and 0.86 (cross-session) may therefore reflect slow drift, arousal/fatigue fluctuations, impedance changes, or autonomic state rather than saccade-specific neural activity. The authors explicitly acknowledge this temporal separation in Section 3.2. This is load-bearing because the abstract and conclusion base the feasibility claim primarily on this onset-classification result. A concrete remedy would be to re-run the onset classification using fixation epochs drawn from within the saccade block (e.g., inter-saccade intervals), or to include a control feature such as time elapsed within the run and show that the AUC is not explained by it. The Visually-Guided Task does not independently rescue the claim because its initial saccades are contaminated by cue-evoked P300 activity and its back-saccade onset results are weaker.
  2. [Table 1 and Section 4.3] The claim that pre-saccade decoding proves the features are neural rather than EOG artifacts is not fully supported. The pre-saccade epochs are still drawn from the saccade block, so the same blocked-design confound applies to the fixation-vs-saccade comparison in Table 1. In addition, the eye tracker operates at 30-60 Hz (Section 2.3), giving onset-label jitter of roughly 16-33 ms; the [-500, 0] ms pre-saccade window can therefore contain a few samples of post-onset activity for the earliest labeled onsets. The authors should quantify this by repeating the pre-saccade analysis with a window that ends at least 50-100 ms before the labeled onset, or by demonstrating that the classification is insensitive to label jitter.
  3. [3.4, Figure 8, Table 1, Section 4.4.2] The direction-decoding results are reported as point estimates without confidence intervals, per-fold variability, or permutation tests against chance. This matters because Figure 8 shows cross-session direction decoding often falling below the 0.5 chance line, and the abstract highlights only the best binary comparisons (left vs. up and left vs. down) from a larger set. The reader cannot tell whether the four-class AUC of 0.67 or the binary AUC of 0.75 is statistically reliable. The authors should report the distribution of fold AUCs, include error bars or confidence intervals, and test the pairwise binary results against chance with correction for multiple comparisons.
  4. [4.4.2, Table A1, Figure C1] The manuscript reports accuracy values in Section 4.4.2 for comparison with prior invasive studies, but the main results are AUC values. The mapping between these metrics is not given, and the reported chance accuracies for binary comparisons differ from the 0.5 AUC reference used elsewhere. For reproducibility, the authors should state which operating point was used to convert AUC to accuracy, or report the accuracy values separately with their corresponding AUC values. This is a reporting gap rather than a central error, but it affects the comparability of the headline numbers.
minor comments (5)
  1. [Table 1] The table lists 'Free-Viewing S1' twice in the Within-Session Testing block; the second row appears to be Free-Viewing S2. Please correct the label.
  2. [Figures 8, A1, C1] The x-axis labels in these figures appear as raw glyph codes (e.g., '/uni00000013/...') rather than human-readable text. Please replace them with proper task and cross-validation labels.
  3. [Figure 6 caption] The caption contains the typo 'disriminable' for 'discriminable.'
  4. [Methods, Section 2.5.2] The text and Figure 3 use 'xDA WN' with a space; the standard algorithm name is xDAWN. Please make the notation consistent.
  5. [General] The manuscript does not include a data/code availability statement. Given the single-participant nature of the study and the central role of the onset-classification result, sharing the de-identified features or analysis code would substantially increase confidence in the reported AUC values.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the reported AUC scores are empirical held-out classification results, not quantities derived from the inputs by construction, and self-citations concern device hardware and participant identity rather than the saccade-decoding claim.

full rationale

The paper's central claims are empirical classification outcomes (fixation vs. saccade AUC 0.88 within session, 0.86 across sessions; direction AUC 0.67 four-class, up to 0.75 binary) obtained with Leave-One-Run-Out cross-validation on sensor-space Stentrode recordings. The class labels come from an eye tracker, which is an independent measurement modality, and no fitted parameter or derived constant is renamed as a prediction. The only equations in the paper (point-biserial correlation and the xDAWN mixing model) are standard analysis tools, not derivation chains that map inputs to the reported results by definition. Self-citations to the Stentrode device [18], signal quality comparisons [19], the SWITCH trial [20], and implant details [24, 25] support hardware and participant identity claims; they are externally published evidence and are not invoked to justify the saccade-decoding conclusion. The paper's own acknowledgment that all fixation trials in the Free-Viewing Task occurred together, followed by all saccade trials, and that this temporal separation 'might have made the discrimination between fixation and saccade trials easier' (Section 3.2) is a genuine experimental validity concern about drift or state confounds, but it is not circularity in the sense of the derivation being equivalent to its inputs. Similarly, selecting discriminative ICs, restricting r2 analysis to 0-30 Hz based on exploratory analysis, and highlighting the best binary direction pairs after seeing results are selective-analysis and multiple-comparison risks, not definitional circularity, because the classifier is still evaluated on held-out runs. No self-definitional, fitted-input-as-prediction, or self-citation-load-bearing step can be exhibited from the text. The appropriate finding is therefore no significant circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

No new physical entities, forces, or conserved quantities are introduced. The free parameters are analysis choices, not derived constants. The most load-bearing unstated premise is that the blocked fixation and saccade intervals differ only by eye movement, which the paper itself partially concedes may not hold in Section 3.2.

free parameters (4)
  • xDAWN ncomponents = 2
    Chosen because it yielded the best decoding results (Section 2.5.2); no nested cross-validation is described, so this number is effectively fit to the data.
  • Random Forest hyperparameters (n_estimators, max_depth) = not reported
    Optimized by grid search (Section 2.5.2) without reporting selected values or a nested procedure; affects all classification results.
  • Channel subsets for ICA and TS-CAR = 13 and 11 channels from 15 active
    Two noisy channels removed before ICA and two more excluded before TS-CAR based on visual inspection (Sections 2.4.3 and 2.5.2); all results depend on these selections.
  • Frequency range for r2 analysis = 0 to 30 Hz
    Restricted to 0-30 Hz based on exploratory analysis (Section 2.5.1), which uses the same data to select the analysis range.
assumptions (5)
  • domain assumption Stentrode signals near the supplementary motor area contain oculomotor planning and execution activity.
    The study infers from anatomy and prior work (Sections 2.1, references [21-23]) that superior sagittal sinus recordings adjacent to precentral gyrus capture SEF/FEF-related activity; this is a precondition for interpreting the ERPs as saccade-related.
  • domain assumption Eye-tracker-derived cursor positions provide accurate saccade onset times.
    Labels come from a 30/60 Hz cursor signal plus manual visual inspection (Section 2.4.1); if onset times are biased late, pre-saccade windows may include movement-related activity, weakening the EOG argument in Section 4.3.
  • domain assumption Fixation and saccade blocks are comparable except for eye movement.
    The design assumes that differences between the initial 15 s fixation block and the later saccade interval are due to saccades, but time-in-run confounds are possible and acknowledged in Section 3.2.
  • standard math FastICA's linear instantaneous mixing model applies to endovascular recordings.
    ICA is used as blind source separation on 13 channels (Section 2.4.3); the validity of this mixing model for the electrode geometry and volume conduction is unverified.
  • domain assumption EOG contamination is negligible for these recordings.
    The interpretation depends on the implant being anatomically distant from the corneo-retinal dipole (Section 4.3), which is plausible but not directly measured in this study.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface." pith.science (2026). https://pith.science/paper/F5EXRRSK

@misc{pith2026250607481,
  author       = {Pith},
  title        = {Pith review of: Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F5EXRRSK}},
  note         = {Machine review of arXiv:2506.07481}
}
read the original abstract

An Oculomotor Brain-Computer Interface (BCI) records neural activity from regions of the brain involved in planning eye movements and translates this activity into control commands. While previous successful oculomotor BCI studies primarily relied on invasive microelectrode implants in non-human primates, this study investigates the feasibility of an oculomotor BCI using a minimally invasive endovascular Stentrode device implanted near the supplementary motor area in a patient with amyotrophic lateral sclerosis (ALS). To achieve this, self-paced visually-guided and free-viewing saccade tasks were designed, in which the participant performed saccades in four directions (left, right, up, down), with simultaneous recording of endovascular EEG and eye gaze. The visually guided saccades were cued with visual stimuli, whereas the free-viewing saccades were self-directed without explicit cues. The results showed that while the neural responses of visually guided saccades overlapped with the cue-evoked potentials, the free-viewing saccades exhibited distinct saccade-related potentials that began shortly before eye movement, peaked approximately 50 ms after saccade onset, and persisted for around 200 ms. In the frequency domain, these responses appeared as a low-frequency synchronisation below 15 Hz. Classification of 'fixation vs. saccade' was robust, achieving mean area under the receiver operating characteristic curve (AUC) scores of 0.88 within sessions and 0.86 between sessions. In contrast, classifying saccade direction proved more challenging, yielding within-session AUC scores of 0.67 for four-class decoding and up to 0.75 for the best-performing binary comparisons (left vs. up and left vs. down). This proof-of-concept study demonstrates the feasibility of an endovascular oculomotor BCI in an ALS patient, establishing a foundation for future oculomotor BCI studies in human subjects.

Figures

Figures reproduced from arXiv: 2506.07481 by the authors.

Figure 1
Figure 1. Anatomical placement of the Stentrode implant relative to major [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Task description (a) Visually-Guided Saccade Task: Each trial began [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Signal processing pipelines for feature visualisation and decoding. [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Evoked responses of Independent Components (ICs) plotted against time, [PITH_FULL_IMAGE:figures/full_fig_p017_4.png]
Figure 5
Figure 5. Figure 5: Evoked responses for the Visually-Guided (a-d) and Free-Viewing Tasks [PITH_FULL_IMAGE:figures/full_fig_p019_5.png]
Figure 6
Figure 6. Figure 6: Squared Point Biserial Correlation Coefficient ( [PITH_FULL_IMAGE:figures/full_fig_p023_6.png]
Figure 7
Figure 7. Figure 7: Spectrograms of Visually-Guided Task (IC4; panels a–c) and Free-Viewing [PITH_FULL_IMAGE:figures/full_fig_p025_7.png]
Figure 8
Figure 8. Figure 8: Saccade classification performance using sensor space data. Each panel [PITH_FULL_IMAGE:figures/full_fig_p027_8.png]
Figure 9
Figure 9. Figure 9: Distribution of wait times between the appearance of the visual cue and [PITH_FULL_IMAGE:figures/full_fig_p031_9.png]
Figure 5
Figure 5. Figure 5: figure 5 [PITH_FULL_IMAGE:figures/full_fig_p044_5.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

72 extracted references · 72 canonical work pages

  1. [1]

    Motor imagery and direct brain-computer communication

    Gert Pfurtscheller and Christa Neuper. Motor imagery and direct brain-computer communication. Proceedings of the IEEE, 89(7):1123–1134, 2001

    work page 2001

  2. [2]

    Neuronal ensemble control of prosthetic devices by a human with tetraplegia

    Leigh R Hochberg, Mijail D Serruya, Gerhard M Friehs, Jon A Mukand, Maryam Saleh, Abraham H Caplan, Almut Branner, David Chen, Richard D Penn, and John P Donoghue. Neuronal ensemble control of prosthetic devices by a human with tetraplegia. Nature, 442(7099):164–171, 2006

    work page 2006

  3. [3]

    Decoding of intended saccade direction in an oculomotor brain–computer interface

    Nan Jia, Scott L Brincat, Andr´ es F Salazar-G´ omez, Mikhail Panko, Frank H Guenther, and Earl K Miller. Decoding of intended saccade direction in an oculomotor brain–computer interface. Journal of Neural Engineering , 14(4):046007, 2017

    work page 2017

  4. [4]

    Towards EEG-based eye-tracking for interaction design in head-mounted devices

    Marc Philipp Dietrich, G¨ otz Winterfeldt, and Sebastian von Mammen. Towards EEG-based eye-tracking for interaction design in head-mounted devices. In 2017 IEEE 7th International Conference on Consumer Electronics-Berlin (ICCE-Berlin) , pages 227–232. IEEE, 2017

    work page 2017

  5. [5]

    Brain–machine interface for eye movements

    Arnulf BA Graf and Richard A Andersen. Brain–machine interface for eye movements. Proceedings of the National Academy of Sciences , 111(49):17630–17635, 2014

    work page 2014

  6. [6]

    Decoding the timing and target locations of saccadic eye movements from neuronal activity in macaque oculomotor areas

    Shogo Ohmae, Toshimitsu Takahashi, Xiaofeng Lu, Yasunori Nishimori, Yasushi Kodaka, Ichiro Takashima, and Shigeru Kitazawa. Decoding the timing and target locations of saccadic eye movements from neuronal activity in macaque oculomotor areas. Journal of Neural Engineering, 12(3):036014, 2015

    work page 2015

  7. [7]

    Single-trial decoding of intended eye movement goals from lateral prefrontal cortex neural ensembles

    Chadwick B Boulay, Florian Pieper, Matthew Leavitt, Julio Martinez-Trujillo, and Adam J Sachs. Single-trial decoding of intended eye movement goals from lateral prefrontal cortex neural ensembles. Journal of Neurophysiology , 115(1):486–499, 2016

    work page 2016

  8. [8]

    Decoding saccadic directions using epidural ECoG in non-human primates

    Jeyeon Lee, Hoseok Choi, Seho Lee, Baek Hwan Cho, Kyoung-ha Ahn, In Young Kim, Kyoung-Min Lee, and Dong-Pyo Jang. Decoding saccadic directions using epidural ECoG in non-human primates. Journal of Korean Medical Science , 32(8):1243, 2017

    work page 2017

Show all 72 references
  1. [9]

    EEG signal processing for eye tracking

    Mohammad Reza Haji Samadi and Neil Cooke. EEG signal processing for eye tracking. In 2014 22nd European Signal Processing Conference (EUSIPCO) , pages 2030–2034. IEEE, 2014

    2014

  2. [10]

    Using deep learning to classify saccade direction from brain activity

    Ard Kastrati, Martyna Beata Plomecka, Roger Wattenhofer, and Nicolas Langer. Using deep learning to classify saccade direction from brain activity. In ACM Symposium on Eye Tracking Research and Applications, pages 1–6, 2021

    2021

  3. [11]

    Using deep neural networks for natural saccade classification from electroencephalograms

    Alexandre Drouin-Picaro and Tiago H Falk. Using deep neural networks for natural saccade classification from electroencephalograms. In 2016 IEEE EMBS International Student Conference (ISC), pages 1–4. IEEE, 2016

    2016

  4. [12]

    Detecting eye movements in EEG for controlling devices

    Soumya Sen Gupta, Sumit Soman, P Govind Raj, Rishi Prakash, S Sailaja, and Rupam Borgohain. Detecting eye movements in EEG for controlling devices. In 2012 IEEE International Conference on Computational Intelligence and Cybernetics (CyberneticsCom) , pages 69–73. IEEE, 2012

    2012

  5. [13]

    Single trial analysis on saccade-related EEG signal

    Arao Funase, Tohru Yagi, Allan K Barros, Andrzej Cichocki, and Ichi Takumi. Single trial analysis on saccade-related EEG signal. In 2007 3rd International IEEE/EMBS Conference on Neural Engineering , pages 371–374. IEEE, 2007. Decoding Saccadic Eye Movements from Brain Signals...

    2007

  6. [14]

    Analysis of neural correlates of saccadic eye movements

    Jan-Philipp Tauscher, Fabian Wolf Schottky, Steve Grogorick, Marcus Magnor, and Maryam Mustafa. Analysis of neural correlates of saccadic eye movements. In Proceedings of the 15th ACM Symposium on Applied Perception , pages 1–9, 2018

    2018

  7. [15]

    Research on relationship between saccadic eye movements and EEG signals in the case of free movements and cued movements

    Arao Funase, Andrzej Cichocki, and Ichi Takumi. Research on relationship between saccadic eye movements and EEG signals in the case of free movements and cued movements. In Neural Information Processing: 18th International Conference, ICONIP 2011, Shanghai, China, November 13-...

    2011

  8. [16]

    Classification of four eye directions from EEG signals for eye-movement-based communication systems

    Abdelkader Nasreddine Belkacem, Hideaki Hirose, Natsue Yoshimura, Duk Shin, and Yasuharu Koike. Classification of four eye directions from EEG signals for eye-movement-based communication systems. Life, 1:3, 2014

    2014

  9. [17]

    Tsipouras, Nikolaos Giannakeas, and Alexandros T Tzallas

    Evangelos Antoniou, Pavlos Bozios, Vasileios Christou, Katerina D Tzimourta, Konstantinos Kalafatakis, Markos G. Tsipouras, Nikolaos Giannakeas, and Alexandros T Tzallas. EEG-based eye movement recognition using brain–computer interface and random forests. Sensors, 21(7):2339, 2021

    2021

  10. [18]

    Minimally invasive endovascular stent-electrode array for high-fidelity, chronic recordings of cortical neural activity

    Thomas J Oxley, Nicholas L Opie, Sam E John, Gil S Rind, Stephen M Ronayne, Tracey L Wheeler, Jack W Judy, Alan J McDonald, Anthony Dornom, Timothy JH Lovell, et al. Minimally invasive endovascular stent-electrode array for high-fidelity, chronic recordings of cortical neural ...

    2016

  11. [19]

    Signal quality of simultaneously recorded endovascular, subdural and epidural signals are comparable

    Sam E John, Nicholas L Opie, Yan T Wong, Gil S Rind, Stephen M Ronayne, Giulia Gerboni, Sebastien H Bauquier, Terence J O’Brien, Clive N May, David B Grayden, et al. Signal quality of simultaneously recorded endovascular, subdural and epidural signals are comparable. Scientifi...

    2018

  12. [20]

    Peter Mitchell, Sarah CM Lee, Peter E Yoo, Andrew Morokoff, Rahul P Sharma, Daryl L Williams, Christopher MacIsaac, Mark E Howard, Lou Irving, Ivan Vrljic, et al. Assessment of safety of a fully implanted endovascular brain-computer interface for severe paralysis in 4 patients...

    2023

  13. [21]

    Human eye fields in the frontal lobe as studied by epicortical recording of movement-related cortical potentials

    Junichi Yamamoto, Akio Ikeda, Takeshi Satow, Masao Matsuhashi, Koichi Baba, Fumitaka Yamane, Susumu Miyamoto, Tadahiro Mihara, Tomokatsu Hori, Waro Taki, et al. Human eye fields in the frontal lobe as studied by epicortical recording of movement-related cortical potentials. Br...

    2004

  14. [22]

    Neurophysiology and neuroanatomy of reflexive and volitional saccades: evidence from studies of humans

    Jennifer E McDowell, Kara A Dyckman, Benjamin P Austin, and Brett A Clementz. Neurophysiology and neuroanatomy of reflexive and volitional saccades: evidence from studies of humans. Brain and Cognition , 68(3):255–270, 2008

    2008

  15. [23]

    The cortex is in overall control of ‘voluntary’eye movement

    P Pouget. The cortex is in overall control of ‘voluntary’eye movement. Eye, 29(2):241–245, 2015

    2015

  16. [24]

    Motor neuroprosthesis implanted with neurointerventional surgery improves capacity for activities of daily living tasks in severe paralysis: first in-human experience

    Thomas J Oxley, Peter E Yoo, Gil S Rind, Stephen M Ronayne, CM Sarah Lee, Christin Bird, Victoria Hampshire, Rahul P Sharma, Andrew Morokoff, Daryl L Williams, et al. Motor neuroprosthesis implanted with neurointerventional surgery improves capacity for activities of daily liv...

    2021

  17. [25]

    Motor activity in gamma and high gamma bands recorded with a stentrode from the human motor cortex in two people with als

    Kriti Kacker, Nikole Chetty, Ariel K Feldman, James Bennett, Peter E Yoo, Adam Fry, David Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface49 Lacomis, Noam Y Harel, Raul G Nogueira, Shahram Majidi, et al. Motor activity in gamma and high...

    2025

  18. [26]

    PsychoPy2: Experiments in behavior made easy

    Jonathan Peirce, Jeremy R Gray, Sol Simpson, Michael MacAskill, Richard H¨ ochenberger, Hiroyuki Sogo, Erik Kastman, and Jonas Kristoffer Lindeløv. PsychoPy2: Experiments in behavior made easy. Behavior Research Methods, 51:195–203, 2019

    2019

  19. [27]

    Lab Streaming Layer (LSL)

    Christian Kothe. Lab Streaming Layer (LSL)

  20. [28]

    k-fold cross-validation can significantly over-estimate true classification accuracy in common EEG-based passive BCI experimental designs: an empirical investigation

    Jacob White and Sarah D Power. k-fold cross-validation can significantly over-estimate true classification accuracy in common EEG-based passive BCI experimental designs: an empirical investigation. Sensors, 23(13):6077, 2023

    2023

  21. [29]

    Independent component analysis of electroencephalographic data

    Scott Makeig, Anthony Bell, Tzyy-Ping Jung, and Terrence J Sejnowski. Independent component analysis of electroencephalographic data. Advances in Neural Information Processing Systems, 8, 1995

    1995

  22. [30]

    MEG and EEG data analysis with mne-python

    Alexandre Gramfort, Martin Luessi, Eric Larson, Denis A Engemann, Daniel Strohmeier, Christian Brodbeck, Roman Goj, Mainak Jas, Teon Brooks, Lauri Parkkonen, et al. MEG and EEG data analysis with mne-python. Frontiers in Neuroinformatics, 7:267, 2013

    2013

  23. [31]

    Faster independent component analysis by preconditioning with hessian approximations

    Pierre Ablin, Jean-Fran¸ cois Cardoso, and Alexandre Gramfort. Faster independent component analysis by preconditioning with hessian approximations. IEEE Transactions on Signal Processing, 66(15):4040–4049, 2018

    2018

  24. [32]

    Group-level component analyses of eeg: validation and evaluation

    Rene J Huster, Sergey M Plis, and Vince D Calhoun. Group-level component analyses of eeg: validation and evaluation. Frontiers in neuroscience, 9:254, 2015

    2015

  25. [33]

    Eegift: Group independent component analysis for event-related eeg data

    Tom Eichele, Srinivas Rachakonda, Brage Brakedal, Rune Eikeland, and Vince D Calhoun. Eegift: Group independent component analysis for event-related eeg data. Computational intelligence and neuroscience , 2011(1):129365, 2011

    2011

  26. [34]

    Stability of ica decomposition across within-subject EEG datasets

    Romain Grandchamp, Claire Braboszcz, Scott Makeig, and Arnaud Delorme. Stability of ica decomposition across within-subject EEG datasets. In 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society , pages 6735–6739. IEEE, 2012

    2012

  27. [35]

    A large scale screening study with a smr-based BCI: Categorization of BCI users and differences in their smr activity

    Claudia Sannelli, Carmen Vidaurre, Klaus-Robert M¨ uller, and Benjamin Blankertz. A large scale screening study with a smr-based BCI: Categorization of BCI users and differences in their smr activity. PloS One , 14(1):e0207351, 2019

    2019

  28. [36]

    Decoding neural activity to anticipate eye movements

    Neda Nategh. Decoding neural activity to anticipate eye movements. In 2020 54th Asilomar Conference on Signals, Systems, and Computers , pages 375–378. IEEE, 2020

    2020

  29. [37]

    EEG is better when cleaning effectively targets artifacts

    Neil W Bailey, Aron T Hill, Kate Godfrey, M Prabhavi N Perera, Nigel C Rogasch, Bernadette M Fitzgibbon, and Paul B Fitzgerald. EEG is better when cleaning effectively targets artifacts. bioRxiv, pages 2024–06, 2024

    2024

  30. [38]

    Decoding Sensorimotor Rhythms for Brain-Computer Interfaces

    James Bennett. Decoding Sensorimotor Rhythms for Brain-Computer Interfaces. PhD thesis, The University of Melbourne, 2021

    2021

  31. [39]

    Makoto’s preprocessing pipeline

    Makoto Miyakoshi. Makoto’s preprocessing pipeline. https://sccn.ucsd.edu/wiki/ Makoto’s_preprocessing_pipeline. [Accessed 10-01-2025]

    2025

  32. [40]

    xDA WN algorithm to enhance evoked potentials: application to brain–computer interface

    Bertrand Rivet, Antoine Souloumiac, Virginie Attina, and Guillaume Gibert. xDA WN algorithm to enhance evoked potentials: application to brain–computer interface. IEEE Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface50 Transactions on B...

    2009

  33. [41]

    Theoretical analysis of xdawn algorithm: application to an efficient sensor selection in a p300 BCI

    Bertrand Rivet, Hubert Cecotti, Antoine Souloumiac, Emmanuel Maby, and J´ er´ emie Mattout. Theoretical analysis of xdawn algorithm: application to an efficient sensor selection in a p300 BCI. In 2011 19th European Signal Processing Conference , pages 1382–1386. IEEE, 2011

    2011

  34. [42]

    Pedregosa, G

    F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine learning in Python. Journal of Machine Learning...

    2011

  35. [43]

    Machine learning seizure prediction: one problematic but accepted practice

    Joseph West, Zahra Dasht Bozorgi, Jeffrey Herron, Howard J Chizeck, Jordan D Chambers, and Lyra Li. Machine learning seizure prediction: one problematic but accepted practice. Journal of Neural Engineering , 20(1):016008, 2023

    2023

  36. [44]

    The p300 wave of the human event-related potential

    Terence W Picton et al. The p300 wave of the human event-related potential. Journal of Clinical Neurophysiology, 9:456–456, 1992

    1992

  37. [45]

    One out of ten independent components shows flipped polarity with poorer data quality: Eeg database study

    Makoto Miyakoshi, Hyeonseok Kim, Masaki Nakanishi, Jason Palmer, and Noriaki Kanayama. One out of ten independent components shows flipped polarity with poorer data quality: Eeg database study. Technical report, Wiley Online Library, 2024

    2024

  38. [46]

    Jobin T Philip and S Thomas George. Visual p300 mind-speller brain-computer interfaces: a walk through the recent developments with special focus on classification algorithms.Clinical EEG and Neuroscience , 51(1):19–33, 2020

    2020

  39. [47]

    Advances in p300 brain–computer interface spellers: toward paradigm design and performance evaluation

    Jiahui Pan, XueNing Chen, Nianming Ban, JiaShao He, Jiayi Chen, and Haiyun Huang. Advances in p300 brain–computer interface spellers: toward paradigm design and performance evaluation. Frontiers in Human Neuroscience, 16:1077717, 2022

    2022

  40. [48]

    Complete locked-in and locked-in patients: command following assessment and communication with vibro-tactile P300 and motor imagery brain-computer interface tools

    Christoph Guger, Rossella Spataro, Brendan Z Allison, Alexander Heilinger, Rupert Ortner, Woosang Cho, and Vincenzo La Bella. Complete locked-in and locked-in patients: command following assessment and communication with vibro-tactile P300 and motor imagery brain-computer inte...

    2017

  41. [49]

    Ssvep bci and eye tracking use by individuals with late-stage als and visual impairments

    Betts Peters, Steven Bedrick, Shiran Dudy, Brandon Eddy, Matt Higger, Michelle Kinsella, Deirdre McLaughlin, Tab Memmott, Barry Oken, Fernando Quivira, et al. Ssvep bci and eye tracking use by individuals with late-stage als and visual impairments. Frontiers in Human Neuroscie...

    2020

  42. [50]

    Ocular artifact elimination from electroencephalography signals: A systematic review

    Rakesh Ranjan, Bikash Chandra Sahana, and Ashish Kumar Bhandari. Ocular artifact elimination from electroencephalography signals: A systematic review. Biocybernetics and Biomedical Engineering, 41(3):960–996, 2021

    2021

  43. [51]

    A novel wearable forehead eog measurement system for human computer interfaces

    Jeong Heo, Heenam Yoon, and Kwang Suk Park. A novel wearable forehead eog measurement system for human computer interfaces. Sensors, 17(7):1485, 2017

    2017

  44. [52]

    Combining EEG and eye tracking: identification, characterization, and correction of eye movement artifacts in electroencephalographic data

    Michael Pl¨ ochl, Jos´ e P Ossand´ on, and Peter K¨ onig. Combining EEG and eye tracking: identification, characterization, and correction of eye movement artifacts in electroencephalographic data. Frontiers in Human Neuroscience, 6:278, 2012

    2012

  45. [53]

    Ocular reduction in EEG signals based on adaptive filtering, regression and blind source separation

    S Romero, MA Ma˜ nanas, and Manel J Barbanoj. Ocular reduction in EEG signals based on adaptive filtering, regression and blind source separation. Annals of Biomedical Engineering, 37:176–191, 2009. Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neura...

    2009

  46. [54]

    Measurement of saccadic eye movements by electrooculography for simultaneous EEG recording

    Yingxin Jia and Christopher W Tyler. Measurement of saccadic eye movements by electrooculography for simultaneous EEG recording. Behavior Research Methods , 51:2139–2151, 2019

    2019

  47. [55]

    Saccadic eye movement related potentials

    Fedor Jagla, M Jergelova, and I Rieˇ cansk` y. Saccadic eye movement related potentials. Physiological Research, 56(6), 2007

    2007

  48. [56]

    Cerebral events preceding self-paced and visually triggered saccades

    GW Thickbroom and FL Mastaglia. Cerebral events preceding self-paced and visually triggered saccades. a study of presaccadic potentials. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section, 62(4):277–289, 1985

    1985

  49. [57]

    The presaccadic cortical negativity prior to self-paced saccades with and without visual guidance

    W Klostermann, D K¨ ompf, W Heide, R Verleger, B Wauschkuhn, and T Seyfert. The presaccadic cortical negativity prior to self-paced saccades with and without visual guidance. Electroencephalography and Clinical Neurophysiology, 91(3):219–228, 1994

    1994

  50. [58]

    Presaccadic spike potential to horizontal eye movements

    C Boylan and HR Doig. Presaccadic spike potential to horizontal eye movements. Electroencephalography and Clinical Neurophysiology, 70(6):559–562, 1988

    1988

  51. [59]

    Presaccadic spike potential with vertical saccades

    H Ross Doig, Catheryn J Nesfield, and Christine Boylan. Presaccadic spike potential with vertical saccades. Ophthalmic and Physiological Optics , 10(2):182–185, 1990

    1990

  52. [60]

    Analysis on saccade-related EEG with independent component analysis

    Arao Funase, Tohru Yagi, Allan K Barros, Yoshiaki Kuno, and Yoshiki Uchikawa. Analysis on saccade-related EEG with independent component analysis. International Journal of Applied Electromagnetics and Mechanics , 14(1-4):353–358, 2002

    2002

  53. [61]

    Imaging the Future Medicine

    Kazuyuki Ohno, A Funase, A Cichocki, and I Takumi. Analysis of EEG signals in memory guided saccade tasks. In World Congress on Medical Physics and Biomedical Engineering 2006: August 27–September 1, 2006 COEX Seoul, Korea “Imaging the Future Medicine” , pages 2664–2667. Sprin...

    2006

  54. [62]

    Brain-computer interfaces patient preferences: a systematic review

    Jamie FM Brannigan, Kishan Liyanage, Hugo Layard Horsfall, Luke Bashford, William Muirhead, and Adam Fry. Brain-computer interfaces patient preferences: a systematic review. Journal of Neural Engineering , 2024

    2024

  55. [63]

    Data augmentation for learning predictive models on eeg: a systematic comparison

    C´ edric Rommel, Joseph Paillard, Thomas Moreau, and Alexandre Gramfort. Data augmentation for learning predictive models on eeg: a systematic comparison. Journal of Neural Engineering , 19(6):066020, 2022

    2022

  56. [64]

    Online and offline domain adaptation for reducing BCI calibration effort

    Dongrui Wu. Online and offline domain adaptation for reducing BCI calibration effort. IEEE Transactions on Human-Machine Systems , 47(4):550–563, 2016

    2016

  57. [65]

    Transfer learning for motor imagery based brain–computer interfaces: A tutorial

    Dongrui Wu, Xue Jiang, and Ruimin Peng. Transfer learning for motor imagery based brain–computer interfaces: A tutorial. Neural Networks, 153:235–253, 2022

    2022

  58. [66]

    Primate frontal eye fields

    Charles J Bruce and Michael E Goldberg. Primate frontal eye fields. i. single neurons discharging before saccades. Journal of Neurophysiology , 53(3):603–635, 1985

    1985

  59. [67]

    EEG SIGNAL PROCESSING FOR EYE TRACKING

    Mohammad Reza, Haji Samadi, and Neil Cooke. EEG SIGNAL PROCESSING FOR EYE TRACKING. Technical report, 2014

    2014

  60. [68]

    Neural correlates of motor imagery and execution in real-world dynamic behavior: evidence for similarities and differences

    Magda Mustile, Dimitrios Kourtis, Martin G Edwards, David I Donaldson, and Magdalena Ietswaart. Neural correlates of motor imagery and execution in real-world dynamic behavior: evidence for similarities and differences. Frontiers in Human Neuroscience, 18:1412307, 2024

    2024

  61. [69]

    Hmm-based gesture recognition for eye-swipe typing

    Matthew Mifsud, Tracey A Camilleri, and Kenneth P Camilleri. Hmm-based gesture recognition for eye-swipe typing. Biomedical Signal Processing and Control, 86:105161, 2023

    2023

  62. [70]

    The state of clinical Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface52 trials of implantable brain–computer interfaces

    K Michelle Patrick-Krueger, Ian Burkhart, and Jose L Contreras-Vidal. The state of clinical Decoding Saccadic Eye Movements from Brain Signals Using an Endovascular Neural Interface52 trials of implantable brain–computer interfaces. Nature Reviews Bioengineering, 3(1):50–67, 2025

    2025

  63. [71]

    A comparative review of detection methods in SSVEP-based brain-computer interfaces

    Amin Besharat, Nasser Samadzadehaghdam, and Reyhaneh Afghan. A comparative review of detection methods in SSVEP-based brain-computer interfaces. IEEE Access, 2024

    2024

  64. [72]

    Hmm based automated wheelchair navigation using eog traces in EEG

    Fayeem Aziz, Hamzah Arof, Norrima Mokhtar, and Marizan Mubin. Hmm based automated wheelchair navigation using eog traces in EEG. Journal of Neural Engineering, 11(5):056018, 2014

    2014

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.