Pith. sign in

REVIEW 3 major objections 7 minor 1 cited by

Explainable MST-ECoGNet Decode Visual Information from ECoG Signal

T0 review · 3 major / 7 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read A compact convolutional net with MST time-frequency features decodes visual categories from ECoG signals more accurately than a baseline model ten times its size.

desk verdict Frequency-range selection on test folds is the load-bearing flaw; the compact architecture and exploratory findings still warrant a second look. read the letter →

arxiv 2411.16165 v1 pith:DMKNB5MH submitted 2024-11-25 math.NA cs.NA

Changqing JI This is my paper · ORCID
classification math.NAcs.NA
keywords ECoGdecodingvisualinformationModifiedSTransformtime-frequencyanalysisspatialfilterbrain-computerinterfaceexplainabledeeplearningmacaquecortex
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 proposes MST-ECoGNet, a decoder that turns electrocorticogram (ECoG) recordings of macaque visual cortex into a six-way classification of viewed images (building, body part, face, fruit, insect, tool). It aims to show that a small, interpretable pipeline—Modified S Transform time-frequency analysis followed by a single spatial-filter convolution—extracts the visual information that larger black-box models miss. The reported result is higher accuracy than the baseline model at roughly one tenth of the parameters: 54.15% vs 36.8% for one monkey and 35.98% vs 27.59% for the other. The paper also claims this architecture reveals where visual information lives: low-frequency bands, spatial patterns across electrodes, and complementary real/imaginary parts of the complex transform, with a roughly 50 ms delay after stimulus onset. These findings matter for building lightweight, explainable brain-computer interfaces and for guiding future electrode design.

What carries the argument

The load-bearing object is the Modified S Transform (MST), a time-frequency transform whose window width adapts to frequency through an arctangent function; it converts each 128-electrode ECoG trial into a 3D complex feature cube indexed by frequency, time, and space. The classifier side is minimal: a batch-normalized ELU network whose first conv3d kernel (1 x 128 x 1) acts as a spatial filter over electrodes, followed by average pooling and two small convolutions, with identical real and imaginary encoders running in parallel and concatenating before a fully connected layer. MST supplies the frequency localization; the spatial filter supplies the channel selection; the parallel encoders supply the complementary real/imaginary information. Together they make the pipeline inspectable at each stage.

What would settle it

Retrain MST-ECoGNet and the baseline with the frequency band for each subject selected on the training folds only (or fixed in advance to e.g. 0-60 Hz), then evaluate on held-out test folds; if the accuracy gap over [5] collapses, the reported advantage is an artifact of test-time band selection rather than the architecture.

Watch

Extended reading notes

Core claim

The central claim is that visual information in ECoG signals is recoverable from the complex time-frequency representation produced by the Modified S Transform, and that a deliberately simple network can exploit it. The model feeds the real and imaginary parts of the MST output through two parallel encoders whose core is a single conv3d layer acting as a spatial filter across the 128 electrodes; the encoded vectors are concatenated and classified by one fully connected layer. On the macaque ECoG dataset, the paper reports six-class accuracies of 54.15±1.15% (MonC) and 35.98±1.15% (MonJ) against baseline accuracies of 36.8±0.51% and 27.59±0.73%, with 0.0396M and 0.0488M parameters versus 0.451M. Ablation experiments attribute the gain to three properties: low-frequency dominance (optimal bands 0-38 Hz and 0-52 Hz), a spatial pattern selected by the spatial filter, and complementarity between real and imaginary coefficients, which the parallel encoders combine. A temporal-importance test finds decodable visual information appearing about 50 ms after stimulus onset.

Load-bearing premise

The per-subject frequency range was chosen by training the model on progressively truncated bands and keeping the one with the best accuracy, and the paper never states that this selection used a separate validation split rather than the same test folds reported in Table II.

Editorial extensions

If this is right

  • If the reported accuracy holds, the same time-frequency plus spatial-filter recipe can be applied to other high-density ECoG arrays, yielding lightweight decoders for BCI use.
  • The per-subject optimal bands (0-52 Hz, 0-38 Hz) imply that future recording systems can concentrate sampling or filtering in low frequencies without losing decodable visual information.
  • The superiority of the spatial filter suggests that electrode array geometry and density, not just signal processing, can be optimized to capture the visual pattern.
  • The real/imaginary complementarity result implies that treating complex time-frequency coefficients as magnitude-only discards usable information, so neural decoders should keep phase-related structure.
  • The observed 50 ms lag between stimulus onset and decodable signal provides a concrete timing anchor for studying visual processing in inferior temporal cortex.

Reading between the lines

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

  • A natural test the paper does not run is whether the per-subject frequency band can be selected inside each training fold; if the advantage persists under that protocol, the method would be cross-validated rather than tuned to the test set.
  • The spatial filter's learned weights could be visualized as a 128-electrode map; comparing that map across monkeys might reveal whether the temporal-cortex spatial pattern is a general visual-code property or an individual implant artifact.
  • If the 50 ms onset delay is a neural signature rather than an artifact of the transform window, then the same ablation protocol applied to other modalities (e.g., EEG or spiking data) should reproduce a comparable delay.
  • Because the paper uses only two subjects and six categories, a direct extension would test whether the real/imaginary parallel encoder helps on multi-subject pooled data and on finer-grained visual categories.
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

3 major / 7 minor

Summary. This paper proposes MST-ECoGNet, a lightweight CNN decoder for six-class visual decoding from ECoG signals. It first applies the Modified S Transform to 128-channel ECoG segments, producing a complex three-dimensional time-frequency-spatial feature volume; separate encoders process the real and imaginary parts with a spatial-filter convolutional layer, and the concatenated features are classified by a fully connected layer. Experiments are reported on two macaque subjects with five-fold cross-validation. The paper claims accuracy improvements over baseline [5] of 6.63 percentage points for MonJ and 16.63 points for MonC while reducing parameter count to about 9-11% of [5], and it reports exploratory analyses on frequency-range selection, spatial/temporal/frequency feature importance, and real-imaginary complementarity, including a 50 ms lag in informative ECoG content after stimulus onset.

Significance. The proposed pipeline is attractive for BCI because it is simple, physically interpretable, and potentially much smaller than the prior model; the exploratory frequency/time importance analyses are useful hypotheses for neuroscience. The paper also makes the comparison falsifiable by specifying the dataset and a concrete baseline. However, the central accuracy claim is not reliably established: the per-subject frequency range is selected on the basis of model performance without a described held-out validation, the baseline numbers are quoted rather than re-run under matched conditions, and several reported values are internally inconsistent. These issues directly affect the headline comparison, so the strength of the contribution cannot be assessed until they are resolved.

major comments (3)
  1. [Sec. IV-A and Table II] The per-subject frequency range is chosen by training the network on reduced data and selecting the range 'based on the performance of the network' (Sec. IV-A), but the paper never states that this selection used validation folds separate from the 5-fold test partitions described in Sec. III-C. Because the frequency range determines the input dimension F and is a hyperparameter, selecting it on the same folds used for the reported accuracies in Table II causes optimistic bias in the headline 6.63% and 16.63% improvements over [5]. A nested validation scheme, or an explicit statement that a separate validation split was used, is required before the accuracy comparison can be accepted.
  2. [Sec. III-E, Table II, and Abstract] The arithmetic of the reported improvements is inconsistent. Table II reports MonC accuracy 54.15±1.15%, which is 17.35 percentage points above the 36.8±0.51% baseline, not 16.63%; MonJ improves by 8.39 points, not 6.63. Section III-E states MonC accuracy 53.43%, differing from Table II's 54.15%. Also, the MonJ model-size reduction is described in Sec. III-E as 'down to 1.08%' whereas 0.0488/0.451 = 10.82%. The central quantitative claims must be corrected and made consistent.
  3. [Sec. III-E and Table II] The comparison with [5] uses accuracy numbers quoted from the prior paper without re-running that model under the same preprocessing, cross-validation folds, and evaluation protocol used for MST-ECoGNet. Unless the implementation and protocol are matched, the reported gains and the parameter-count comparison could reflect differences in the experimental setup rather than the proposed architecture. The authors should re-run [5] on the same preprocessed inputs and folds, or explicitly document that the published numbers were obtained under an identical protocol.
minor comments (7)
  1. [Throughout] There are typographical errors: 'nn.Cov3d' should be 'nn.Conv3d', 'Cantenate' should be 'Concatenate', 'trails' should be 'trials', and 'wad' should be 'was'; these should be corrected.
  2. [Sec. IV-A] The frequency screening grid is not specified (range, step size, and number of candidate ranges); please state it so that the selection procedure is reproducible.
  3. [Sec. II-A] The MST parameters a=5, b=50, c=74 are cited to [40], but the manuscript should state whether these were fixed a priori or tuned on data.
  4. [Table II] The column header 'Params / M' is ambiguous; specify whether the values are in millions and include the exact parameter counts used in the comparison.
  5. [Algorithm 1] The assignments in Algorithm 1 are not typeset cleanly and the notation ti=0 versus ti+=1 is hard to follow; please reformat the pseudocode.
  6. [Figures 5-9] Figures 5, 6, and 7 do not report error bars or the number of folds; adding this information would help assess the reliability of the exploratory findings.
  7. [Sec. IV-D] The sentence 'first, we used the 3D Feature data obtained by the MST method to train the corresponding model' should clarify that the temporal-importance evaluation is done on the already-trained model, as Algorithm 1 states.

Circularity Check

1 steps flagged · score 2.0 of 10

No equation-level circularity; the accuracy results are empirical, but the frequency-range selection in Sec. IV-A is self-referential because the range is chosen by network accuracy and the same 5-fold protocol is used for the reported results.

  1. fitted input called prediction [Section IV-A (Influence of Frequency Range), with Section III.C and Table II]
    "We decide the best frequency range based on the performance of the network. ... The final range of object MonJ is 0-52Hz, the frequency range of object MonC is 0-38Hz. ... In this paper, all the performance result is the average of 5 times test."

    The input frequency range F is selected by training the network on reduced datasets and taking the range with the best accuracy. The paper does not describe a separate validation partition for this selection, while Section III.C states that all reported performance results are the average of the same 5-fold test. If the same folds were used both to choose F and to report Table II, the final accuracy is the quantity that was maximized when choosing the input bandwidth. The subsequent statement that frequency is one of the important parameters affecting model performance then restates the selection criterion rather than providing an independent discovery, and the claimed accuracy gain over [5] is partly forced by the bandwidth choice.

full rationale

The central derivation chain is empirical rather than symbolic. MST is an external transform with a stated formula and parameters taken from reference [40]; the encoder is a newly designed architecture; and the accuracy numbers are measured, not derived from the method's equations. The model-size comparison is arithmetic, and the architecture comparison has independent content. The main self-referential element is the frequency-range screening in Section IV-A, where the input frequency band is chosen by network accuracy and then the same evaluation protocol is used for the reported final results without documenting a separate validation split. This is a correctness or leakage risk rather than a definitional equivalence, and the baseline [5] is prior work by the same group but its comparison is empirical rather than a load-bearing uniqueness argument. I therefore score 2: no equation-level circularity, but one mild self-referential model-selection step that should be treated as a methodological concern.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper's central claims rest on the applicability of MST to ECoG, the comparability of the private dataset to [5], and the validity of occlusion-based importance tests. The main free parameters are the MST window constants and the per-subject frequency ranges selected by model performance; the model's learned weights are not counted as free parameters because they are trained, not used to fit a derivation. No new entities are invented.

free parameters (2)
  • MST window parameters (a, b, c) = a=5, b=50, c=74
    Chosen by hand following the guideline in [40]; they determine the time-frequency resolution and affect all features fed to the network.
  • Per-subject frequency range = MonJ 0-52 Hz, MonC 0-38 Hz
    Selected in Section IV-A based on model accuracy across candidate ranges; this is a performance-fitted hyperparameter that directly sets the input size and the reported accuracy.
assumptions (4)
  • domain assumption Modified S Transform with window parameters a=5, b=50, c=74 provides a valid and sufficiently resolved time-frequency representation of ECoG signals.
    Invoked in Section II.A; if the transform or parameters misrepresent the signal, all downstream features and claims are affected.
  • domain assumption The 300 ms pre-stimulus background captures the monkey's baseline state, so subtracting its mean and variance removes non-stimulus confounds.
    Section III.B preprocessing; if baseline state is not stationary, the normalization may introduce artifacts.
  • domain assumption Occlusion-based importance testing, where segments of the MST feature space are zeroed and accuracy of a frozen model is measured, is a valid measure of when and where information is encoded.
    Section IV.D and Algorithm 1; this assumption is not independently validated and is confounded by the model's pooling and convolution receptive fields.
  • domain assumption The baseline results from [5] are directly comparable because the same dataset and task are used.
    Section III.E compares against [5]'s published accuracy without re-running baseline under identical splits and preprocessing; comparability is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explainable MST-ECoGNet Decode Visual Information from ECoG Signal." pith.science (2026). https://pith.science/paper/DMKNB5MH

@misc{pith2026241116165,
  author       = {Pith},
  title        = {Pith review of: Explainable MST-ECoGNet Decode Visual Information from ECoG Signal},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DMKNB5MH}},
  note         = {Machine review of arXiv:2411.16165}
}
read the original abstract

In the application of brain-computer interface (BCI), we not only need to accurately decode brain signals,but also need to consider the explainability of the decoding process, which is related to the reliability of the model. In the process of designing a decoder or processing brain signals, we need to explain the discovered phenomena in physical or physiological way. An explainable model not only makes the signal processing process clearer and improves reliability, but also allows us to better understand brain activities and facilitate further exploration of the brain. In this paper, we systematically analyze the multi-classification dataset of visual brain signals ECoG, using a simple and highly explainable method to explore the ways in which ECoG carry visual information, then based on these findings, we propose a model called MST-ECoGNet that combines traditional mathematics and deep learning. The main contributions of this paper are: 1) found that ECoG time-frequency domain information carries visual information, provides important features for visual classification tasks. The mathematical method of MST (Modified S Transform) can effectively extract temporal-frequency domain information; 2) The spatial domain of ECoG signals also carries visual information, the unique spatial features are also important features for classification tasks; 3) The real and imaginary information in the time-frequency domain are complementary. The effective combination of the two is more helpful for classification tasks than using amplitude information alone; 4) Finally, compared with previous work, our model is smaller and has higher performance: for the object MonJ, the model size is reduced to 10.82% of base model, the accuracy is improved by 6.63%; for the object MonC, the model size is reduced to 8.78%, the accuracy is improved by 16.63%.

Figures

Figures reproduced from arXiv: 2411.16165 by the authors.

Figure 1
Figure 1. Outline of Data Flow of MST-ECoGNet Bayes Classifier [36], as well as neural network methods [37][38][39]. Although these methods can complete the final classification task well and efficiently, they always pay less attention on the explainability of data processing. Based on the above findings and understanding, we fo￾cus on designing a model that is explainable and also has high performance. In the feature extract… view at source ↗
Figure 2
Figure 2. Outline of ECoG Record Each image will be used as visual stimuli, last 300 ms, ECoG signal will be measured via electrode array. time interval is consistent with the length of the active interval. We calculate the overall mean and variance of the background interval, then use each data of the active interval to subtract this mean and divide it by the variance for normalization. The formula is as follows: E‘acpn, tq … view at source ↗
Figure 4
Figure 4. Each filter has its own focused dimension, through comparing the respective performance results, we can clearly know which dimensional information carries more visual information. Experimental results refer to [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figures from the paper (5 more)
Figure 3
Figure 3. Figure 3: Influence from Frequency Range Results test on 2 subjects & 2 data part, final frequency range: MonJ 0-52Hz, MonC 0-38Hz. The experimental results clearly show the optimal value of the frequency range. The final range of object MonJ is 0- 52Hz, the frequency range of o…
Figure 6
Figure 6. Figure 6: Results of 2 Different Data Form Only test on MonJ dataset, Results show data form of Real-Imaginary would be better than Amplitude-Angle data form. complementary. Use a parallel structure to process real and imaginary data at the same time, concatenate the extracted f…
Figure 8
Figure 8. Figure 8: Temporal Importance Test Results show that along temporal axis, there is clear time lag about 50 msec in both test subjects [PITH_FULL_IMAGE:figures/full_fig_p006_8.png]
Figure 7
Figure 7. Figure 7: Parallel Encoder vs Single Encoder Parallel encoder using both real and imaginary data has better performance compare with single encoder which only one data real or imaginary part. The comparison of experimental results shows that using both data in parallel can effec…
Figure 9
Figure 9. Figure 9: Frequency Importance Test Results show that most information comes from low-frequency region, the comple￾mentary phenomenon exists between real and imaginary data. The experimental results clearly show that in the low￾frequency domain, the real and imaginary data show …

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Bi-Band ECoGNet for ECoG Decoding on Classification Task

    math.NA 2024-11 reject novelty 4.0 of 10

    Bi-Band ECoGNet uses dual-length learnable convolution kernels and a 2D electrode-grid encoder to replace a hand-crafted time-frequency transform, reporting +1.24% accuracy and 6x faster training over MST-ECoGNet on a...

Reference graph

Works this paper leans on

45 extracted references · 30 canonical work pages · cited by 1 Pith paper

  1. [5]

    Daniel L., K., Roberts, D., William M., S., Higgin, M., and Ventureyra, E., ``Alfentanil mediated activation of epileptiform activity in the electrocorticogram during resection of epileptogenic foci,'' Canadian Journal of Neurological Sciences / Journal Canadien des Sciences Neurologiques, vol. 24, no. 1, p. 29–36, 1997

    work page 1997

  2. [1]

    write newline

    " write newline "" initialize.prev.this.status FUNCTION begin.bib " write newline preamble empty 'skip preamble write newline if " thebibliography " longest.label * " " * write newline " [1] #1 " write newline " url@samestyle " write newline " " write newline " [2] #2 " write newline " =0pt " write newline " " ALTinterwordstretchfactor * " " * write newli...

  3. [2]

    11em plus .33em minus .07em 4000 4000 100 4000 4000 500 `\.=1000 = #1 \@IEEEnotcompsoconly \@IEEEcompsoconly #1 * [1] 0pt [0pt][0pt] #1 * \| ** #1 \@IEEEauthorblockNstyle \@IEEEcompsocnotconfonly \@IEEEcompsocconfonly \@IEEEauthorblockAstyle \@IEEEcompsocnotconfonly \@IEEEcompsocconfonly \@IEEEcompsocconfonly \@IEEEauthordefaulttextstyle \@IEEEcompsocnotc...

  4. [3]

    Wakuya, M., Inoue, T., Imoto, H., Maruta, Y., Nomura, S., Suzuki, M., and Yamakawa, T., ``Epileptic seizure–related changes in electrocorticogram, cortical temperature, and cerebral hemodynamics obtained via an implantable multimodal multichannel probe during preoperative monitoring: illustrative case,'' Journal of Neurosurgery: Case Lessons, vol. 3, no. ...

    2022

  5. [4]

    Kutlu, F. and K \"o se, C., ``Epileptic seizure detection from ecog signals acquired with experimental epilepsy,'' 2013 21st Signal Processing and Communications Applications Conference (SIU), pp. 1--4, 2013. [Online]. Available: https://api.semanticscholar.org/CorpusID:18315588

    work page 2013

  6. [6]

    Romanelli, P., Piangerelli, M., Ratel, D., Gaude, C., Costecalde, T., Puttilli, C., Picciafuoco, M., Benabid, A., and Torres, N., ``A novel neural prosthesis providing long-term electrocorticography recording and cortical stimulation for epilepsy and brain-computer interface,'' Journal of Neurosurgery, vol. 130, no. 4, pp. 1166 -- 1179, 2019. [Online]. Av...

    2019

  7. [7]

    Date, H., Kawasaki, K., Hasegawa, I., and Okatani, T., ``Deep learning for channel-agnostic brain decoding across multiple subjects,'' in 2020 8th International Winter Conference on Brain-Computer Interface (BCI), 2020, pp. 1--6

    2020

  8. [8]

    [Online]

    ------, ``Neuronal oscillatory activities in separate frequencies encode hierarchically distinct visual features,'' bioRxiv, 2020. [Online]. Available: https://www.biorxiv.org/content/early/2020/01/14/2020.01.13.902775

    work page 2020

Show all 45 references
  1. [9]

    2331--2336

    ------, ``Deep learning for natural image reconstruction from electrocorticography signals,'' in 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2019, pp. 2331--2336

    2019

  2. [10]

    Brodu, N., Lotte, F., and Lécuyer, A., ``Comparative study of band-power extraction techniques for motor imagery classification,'' in 2011 IEEE Symposium on Computational Intelligence, Cognitive Algorithms, Mind, and Brain (CCMB), 2011, pp. 1--6

    2011

  3. [11]

    Pfurtscheller, G., Neuper, C., Flotzinger, D., and Pregenzer, M., ``Eeg-based discrimination between imagination of right and left hand movement,'' Electroencephalography and Clinical Neurophysiology, vol. 103, no. 6, pp. 642--651, 1997. [Online]. Available: https://www.scienc...

    1997

  4. [12]

    3, 2010, pp

    Chong, L., Hai-bin, Z., Chun-sheng, L., and Hong, W., ``Classification of ecog signals for motor imagery tasks,'' in 2010 2nd International Conference on Signal Processing Systems, vol. 3, 2010, pp. V3--185--V3--188

    2010

  5. [13]

    Barachant, A., Bonnet, S., Congedo, M., and Jutten, C., ``Classification of covariance matrices using a riemannian-based kernel for bci applications,'' Neurocomputing, vol. 112, pp. 172--178, 2013, advances in artificial neural networks, machine learning, and computational int...

    2013

  6. [14]

    Jayaram, V., Alamgir, M., Altun, Y., Scholkopf, B., and Grosse-Wentrup, M., ``Transfer learning in brain-computer interfaces,'' IEEE Computational Intelligence Magazine, vol. 11, no. 1, pp. 20--31, 2016

    2016

  7. [15]

    Kaper, M., Meinicke, P., Grossekathoefer, U., Lingner, T., and Ritter, H., ``Bci competition 2003-data set iib: support vector machines for the p300 speller paradigm,'' IEEE Transactions on Biomedical Engineering, vol. 51, no. 6, pp. 1073--1076, 2004

    2003

  8. [16]

    and Bengio, S., ``Hmm and iohmm modeling of eeg rhythms for asynchronous bci systems,'' 2003

    Chiappa, S. and Bengio, S., ``Hmm and iohmm modeling of eeg rhythms for asynchronous bci systems,'' 2003. [Online]. Available: https://infoscience.epfl.ch/handle/20.500.14299/228388

    2003

  9. [17]

    Millan, J. and Mourino, J., ``Asynchronous bci and local neural classifiers: an overview of the adaptive brain interface project,'' IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 11, no. 2, pp. 159--161, 2003

    2003

  10. [18]

    Hammon, P. S. and de Sa, V. R., ``Preprocessing and meta-classification for brain-computer interfaces,'' IEEE Transactions on Biomedical Engineering, vol. 54, no. 3, pp. 518--525, 2007

    2007

  11. [19]

    W., Roberts, S

    Yoon, J. W., Roberts, S. J., Dyson, M., and Gan, J. Q., ``Adaptive classification for brain computer interface systems using sequential monte carlo sampling,'' Neural Networks, vol. 22, no. 9, pp. 1286--1294, 2009, brain-Machine Interface. [Online]. Available: https://www.scie...

    2009

  12. [20]

    Li, Y., Koike, Y., and Sugiyama, M., ``A framework of adaptive brain computer interfaces,'' in 2009 2nd International Conference on Biomedical Engineering and Informatics, 2009, pp. 1--5

    2009

  13. [21]

    Jiang, T., Jiang, T., Wang, T., Mei, S., Liu, Q., Li, Y., Wang, X., Prabhu, S., Sha, Z., and Ince, N. F., ``Characterization and decoding the spatial patterns of hand extension/flexion using high-density ecog,'' IEEE Transactions on Neural Systems and Rehabilitation Engineerin...

    2017

  14. [22]

    44--49, 2018

    Deng, X., Li, D., Mi, J., Gao, F., Chen, Q., Wang, J., and Liu, R., ``Motor imagery ecog signal classification using sparse representation with elastic net constraint,'' 2018 IEEE 7th Data Driven Control and Learning Systems Conference (DDCLS), pp. 44--49, 2018. [Online]. Avai...

    2018

  15. [23]

    0430--0435, 2018

    Du, A., Yang, S., Liu, W., and Huang, H., ``Decoding ecog signal with deep learning model based on lstm,'' TENCON 2018 - 2018 IEEE Region 10 Conference, pp. 0430--0435, 2018. [Online]. Available: https://api.semanticscholar.org/CorpusID:67874163

    2018

  16. [24]

    Martin M

    Śliwowski M. Martin M. Souloumiac A. Blanchart P. Aksenova T., ``Decoding ecog signal into 3d hand translation using deep learning.'' Journal of neural engineering, 19(2), 10.1088/1741-2552/ac5d69., 2022. [Online]. Available: https://doi.org/10.1088/1741-2552/ac5d69

    2022 doi

  17. [25]

    J., Solon, A

    Lawhern, V. J., Solon, A. J., Waytowich, N. R., Gordon, S. M., Hung, C. P., and Lance, B., ``Eegnet: a compact convolutional neural network for eeg-based brain–computer interfaces,'' Journal of Neural Engineering, vol. 15, 2016. [Online]. Available: https://api.semanticscholar...

    2016

  18. [26]

    W., and Beaufays, F., ``Long short-term memory recurrent neural network architectures for large scale acoustic modeling,'' in INTERSPEECH, 2014, pp

    Sak, H., Senior, A. W., and Beaufays, F., ``Long short-term memory recurrent neural network architectures for large scale acoustic modeling,'' in INTERSPEECH, 2014, pp. 338--342

    2014

  19. [27]

    4720--4727

    Jain, R., Jaiman, P., and Baths, V., ``Feature engineering for an efficient motor related ecog bci system,'' in 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC), 2023, pp. 4720--4727

    2023

  20. [28]

    2, 2010, pp

    Zhao, H.-b., Yu, C.-y., Liu, C., and Wang, H., ``Ecog-based brain-computer interface using relative wavelet energy and probabilistic neural network,'' in 2010 3rd International Conference on Biomedical Engineering and Informatics, vol. 2, 2010, pp. 873--877

    2010

  21. [29]

    CSAE '21

    Shi, H., Yu, P., and Li, H., ``The finger flexion related feature extraction method based on wavelet time-frequency analysis in ecog signals,'' in Proceedings of the 5th International Conference on Computer Science and Application Engineering, ser. CSAE '21. 1em plus 0.5em min...

    2021

  22. [30]

    [Online]

    Devlaminck, D., Waegeman, W., Bauwens, B., Wyns, B., Santens, P., and Otte, G., ``From circular ordinal regression to multilabel classification,'' 2010. [Online]. Available: https://api.semanticscholar.org/CorpusID:16074092

    2010

  23. [31]

    Yi, W., Qiu, S., Qi, H., Zhang, L., Wan, B., and Dong, M., ``Eeg feature comparison and classification of simple and compound limb motor imagery,'' Journal of NeuroEngineering and Rehabilitation, vol. 10, no. 1, oct 2013. [Online]. Available: https://doi.org/10.1186/1743-0003-10-106

    2013 doi

  24. [32]

    and Ismail Gursoy , M., ``Eeg signal classification using pca, ica, lda and support vector machines,'' Expert Systems with Applications, vol

    Subasi, A. and Ismail Gursoy , M., ``Eeg signal classification using pca, ica, lda and support vector machines,'' Expert Systems with Applications, vol. 37, no. 12, pp. 8659--8666, 2010. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0957417410005695

    2010

  25. [33]

    and Kak, A., ``Pca versus lda,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol

    Martinez, A. and Kak, A., ``Pca versus lda,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 2, pp. 228--233, 2001

    2001

  26. [34]

    and Dehuri, S., ``Multiclass classification of eeg signal for epilepsy detection using dwt based svd and fuzzy knn classifier,'' ntelligent Decision Technologies, no

    Singh, N. and Dehuri, S., ``Multiclass classification of eeg signal for epilepsy detection using dwt based svd and fuzzy knn classifier,'' ntelligent Decision Technologies, no. 2, pp. 239-252, 2020, vol. 14, 2020

    2020

  27. [35]

    You, Y., Chen, W., Li, M., Zhang, T., Jiang, Y., and Zheng, X., ``Automatic focal and non-focal eeg detection using entropy-based features from flexible analytic wavelet transform,'' Biomedical Signal Processing and Control, vol. 57, p. 101761, 2020. [Online]. Available: https...

    2020

  28. [36]

    G., Zaitseva, E

    Rabcan, J., Levashenko, V. G., Zaitseva, E. N., Kvassay, M., and Subbotin, S., ``Application of fuzzy decision tree for signal classification,'' IEEE Transactions on Industrial Informatics, vol. 15, pp. 5425--5434, 2019. [Online]. Available: https://api.semanticscholar.org/Cor...

    2019

  29. [37]

    Polat, K. and Güneş, S., ``A novel data reduction method: Distance based data reduction and its application to classification of epileptiform eeg signals,'' Applied Mathematics and Computation, vol. 200, no. 1, pp. 10--27, 2008. [Online]. Available: https://www.sciencedirect.c...

    2008

  30. [38]

    Bon, N. N. and Dai, L. V., ``Fault identification, classification, and location on transmission lines using combined machine learning methods,'' International Journal of Engineering and Technology Innovation, vol. 12, no. 2, p. 91–109, Feb. 2022. [Online]. Available: https://o...

    2022

  31. [39]

    and Chou, C.-A., ``Adaptive seizure onset detection framework using a hybrid pca–csp approach,'' IEEE Journal of Biomedical and Health Informatics, vol

    Khanmohammadi, S. and Chou, C.-A., ``Adaptive seizure onset detection framework using a hybrid pca–csp approach,'' IEEE Journal of Biomedical and Health Informatics, vol. 22, no. 1, pp. 154--160, 2018

    2018

  32. [40]

    and Hsieh, M.-H., ``Classification of mental task from eeg data using neural networks based on particle swarm optimization,'' Neurocomputing, vol

    Lin, C.-J. and Hsieh, M.-H., ``Classification of mental task from eeg data using neural networks based on particle swarm optimization,'' Neurocomputing, vol. 72, no. 4, pp. 1121--1130, 2009, brain Inspired Cognitive Systems (BICS 2006) / Interplay Between Natural and Artificia...

    2009

  33. [41]

    and Harabagiu, S

    Maldonado, R. and Harabagiu, S. M., ``Active deep learning for the identification of concepts and relations in electroencephalography reports,'' Journal of Biomedical Informatics, vol. 98, p. 103265, 2019. [Online]. Available: https://www.sciencedirect.com/science/article/pii/...

    2019

  34. [42]

    Jiao, B., Rui, Y., Gao, M., Fei, H., and Yu, Q., ``A modified s transform with adjustable window function,'' in 2019 IEEE International Conference on Signal, Information and Data Processing (ICSIDP), 2019, pp. 1--5

    2019

  35. [43]

    K., and Che Wan Fadzal, C

    Zabidi, A., Mansor, W., Lee, Y. K., and Che Wan Fadzal, C. W. N. F., ``Short-time fourier transform analysis of eeg signal generated during imagined writing,'' in 2012 International Conference on System Engineering and Technology (ICSET), 2012, pp. 1--4

    2012

  36. [44]

    715--718

    Pukhova, V., Gorelova, E., Ferrini, G., and Burnasheva, S., ``Time-frequency representation of signals by wavelet transform,'' in 2017 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering (EIConRus), 2017, pp. 715--718

    2017

  37. [45]

    Thangaraj, K., Muruganandham, J., Selvaumar, S., and Jagan, R., ``Analysis of harmonics using s-transform,'' in 2016 International Conference on Emerging Trends in Engineering, Technology and Science (ICETETS), 2016, pp. 1--5

    2016

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.