GAF-FusionNet: Multimodal ECG Analysis via Gramian Angular Fields and Split Attention

Feng Liu; Jiahao Qin

REVIEW 4 major objections 6 minor 27 references

GAF-FusionNet: Multimodal ECG Analysis via Gramian Angular Fields and Split Attention

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read GAF-FusionNet fuses ECG waveforms with Gramian Angular Field images and split attention, and reports accuracies of 94.5%, 96.9%, and 99.6% on ECG200, ECG5000, and MIT-BIH Arrhythmia.

desk verdict The MIT-BIH result is uninterpretable without the split; a solid incremental architecture is wasted on a likely-leaky benchmark. read the letter →

arxiv 2501.01960 v1 pith:QVPHHL5T submitted 2024-12-07 cs.CV cs.AIcs.GRcs.LG

Jiahao Qin , Feng Liu This is my paper · ORCID

classification cs.CVcs.AIcs.GRcs.LG

keywords ECGclassificationGramianAngularFieldSplitattentionMultimodalfusionArrhythmiadetectionTime-seriesimagingDeeplearning

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

GAF-FusionNet is an attempt to show that ECG classification improves when the raw heartbeat waveform and an image encoding of the same signal are learned jointly rather than treated as one modality. The paper maps each ECG segment to a Gramian Angular Field image, runs a 1D CNN plus BiLSTM on the waveform and a 2D CNN on the image, and merges the two branches with a dual-layer cross-channel split attention module. On ECG200, ECG5000, and the MIT-BIH Arrhythmia Database, the reported accuracies are 94.5%, 96.9%, and 99.6%, respectively, beating the listed baselines on every dataset and metric. The intended takeaway is that adaptive cross-modal fusion of waveform and image features is a productive direction for ECG classification.

What carries the argument

The central mechanism is the dual-layer cross-channel split attention module, a two-stage attention block between the temporal and spatial branches of the network. The temporal branch is a 1D CNN feeding a BiLSTM; the spatial branch is a 2D CNN applied to the Gramian Angular Field matrix, whose entries are $\cos(\phi_j + \phi_k)$ for angular encodings $\phi_j$ and $\phi_k$ of rescaled ECG samples. In the first attention layer, each branch self-attends over its own features; in the second, each branch attends to the other branch's projected features, and the two attended representations are added, normalized, concatenated, and passed to an MLP classifier. The module's role is to let the network decide per sample how much weight to give the waveform versus the image view, which the ablations identify as the main source of accuracy gain.

What would settle it

Re-run GAF-FusionNet on MIT-BIH with a truly patient-disjoint split—train on one set of subjects, test on a different set—and compare accuracy to the reported 99.6% beat-level figure; a large drop would show that same-patient beats, not learned cardiac patterns, carried the result.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a dual representation of ECG—raw time series plus a Gramian Angular Field image—combined through learned cross-modal attention beats every baseline it is compared against. The reported margins are 2.0 percentage points over the best baseline on ECG200, 1.2 points on ECG5000, and 0.8 points on the MIT-BIH Arrhythmia Database, where the model reaches 99.6% accuracy with a macro F1 of 99.5%. The ablation study attributes the gain to the dual-layer split attention module: removing it drops MIT-BIH accuracy to 97.8%, removing cross-channel interaction drops it to 98.1%, and either single-modality branch stays below 97.5%. The intended lesson is that learned, context-dependent weighting of waveform and image features matters more than simply concatenating them.

Load-bearing premise

The reported MIT-BIH accuracy presupposes that the 87,554 training beats and 21,892 test beats come from different patients; if the same patient's beats appear in both sets, the model can memorize individual heartbeats rather than learn generalizable arrhythmia patterns.

Editorial extensions

If this is right

Adding the GAF image branch and split-attention fusion should improve over waveform-only ECG models: the ablation shows the full model is 2.6 points above the time-series-only variant on MIT-BIH (99.6% vs 97.0%).
If the MIT-BIH result is patient-disjoint, 99.6% on 15 heartbeat classes would be a new benchmark among the listed baselines, whose best is Multi-Scale CNN at 97.8%.
The dual-layer attention module is the main source of gain: removing it costs 1.8 points, while removing only cross-channel attention costs 1.5 points.
The fusion recipe transfers across dataset sizes and recording lengths, from 200-sample ECG200 to the 109,446-beat MIT-BIH collection.

Reading between the lines

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

Editorial inference: the attention module is modality-agnostic, so the same GAF-plus-split-attention design should transfer to EEG, electromyography, or other one-dimensional biosignals without architectural change.
Editorial inference: because the cross-channel attention computes pairwise branch interactions, the design extends naturally to three or more input views—for instance, multiple ECG leads or an added spectrogram—by chaining additional cross-attention steps.
Editorial inference: the paper reports only aggregate metrics; a per-class confusion matrix on MIT-BIH, especially for rare arrhythmia classes, would clarify whether the 99.5% macro F1 is earned evenly or dominated by common beat types.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Quick take: the 99.6% MIT-BIH result is the whole ballgame, and the paper never tells you how the split was made. With 48 recordings and an 80/20 beat-level split, the same patient almost certainly shows up on both sides, so the headline is likely identity leakage. The stress-test note lands.

What's actually new: the dual-layer cross-channel split attention is a real although modest combination—intra-modality self-attention followed by cross-attention, with residual connections and layer norm. The ablation table is honest and shows each component earns its keep. The UCR fixed splits for ECG200 and ECG5000 are standard, so those numbers are less suspect; if the MIT-BIH split were patient-independent, the architecture could be a useful benchmark.

Soft spots beyond the split: no code is actually available despite the abstract promising it; no error bars or repeated runs; implementation details conflict (ResNet34 pretrained on ImageNet as backbone vs. a small 2D CNN in the figure, for GAF images whose size makes ImageNet pretraining odd); the learning-rate schedule in Eq. (21) is not the cosine annealing described in Section 4.1; the baselines appear to be reimplemented without citations to their reported numbers. The claim that GAF-to-ECG is 'largely unexplored' is inaccurate—there's prior work, some cited in their own paper. Minor: the conclusion drifts into 'psychiatric disorders,' which suggests careless editing.

Who it's for: a cautionary example of beat-level leakage, and maybe a starting point if the authors fix the split. As is, the central claim is unsupported. If I were the editor, I'd desk reject and invite a resubmission with a proper inter-patient split, error bars, and code.

Referee Report

4 major / 6 minor

Summary. The paper proposes GAF-FusionNet, a multimodal ECG classifier that processes raw ECG time series and Gramian Angular Field (GAF) images in two parallel branches, fusing them with a dual-layer cross-channel split attention module. The method is evaluated on ECG200, ECG5000, and the MIT-BIH Arrhythmia Database, and Table 2 reports accuracy of 94.5%, 96.9%, and 99.6% respectively, with the claim that GAF-FusionNet consistently outperforms all compared baselines. The paper also presents an ablation study on MIT-BIH. The central claim is empirical: the proposed fusion architecture sets a new state of the art on all three datasets.

Significance. If the reported results hold, GAF-FusionNet would be a useful contribution to multimodal ECG classification, showing that combining time-series and image-based representations with learned cross-modal attention can improve accuracy over single-modality models. The GAF formulation in Section 3.2 and the attention fusion equations in Section 3.3 are clearly presented, and the ablation study is a reasonable attempt to isolate the contribution of each component. However, the significance is currently undercut by missing experimental protocol details. The paper contains no theoretical derivations, no machine-checked proofs, and no released code or data split at the time of review, so the empirical claims must carry the entire contribution. The main result on MIT-BIH depends on a train/test split whose patient independence is not stated, and one of the standard dataset splits appears to be reversed. These issues are fixable but are load-bearing for the paper's central claim.

major comments (4)

[§4.1, Table 1; §4.2, Table 2] The MIT-BIH train/test protocol is not specified. Table 1 reports an 87,554/21,892 beat-level split, but the text never states that recordings from the same patient are confined to one side of the split, nor does it mention the standard inter-patient protocol for this database. With 48 recordings from roughly 47 subjects, a random beat-level 80/20 split makes it nearly certain that beats from every subject appear in both training and test sets, so the 99.6% accuracy in Table 2 may reflect memorization of patient-identity-specific waveform patterns rather than generalization. Please report the exact patient-to-split assignment and rerun all MIT-BIH results under an inter-patient protocol (e.g., training on a subset of patients and testing on held-out patients), stating the number of patients in each split.
[§4.1, Table 1] The ECG5000 split appears to be reversed. The UCR ECG5000 archive uses 500 training samples and 4,500 test samples, but Table 1 lists a 4,500/500 train/test split. If the authors trained on 4,500 samples and tested on 500, the reported 96.9% accuracy is not comparable with results obtained under the standard benchmark split. Please confirm the actual split and correct the table, or explain the discrepancy.
[§4.1, Implementation Details; §3.3] The implementation description is internally inconsistent. Section 3.3 defines a temporal branch using a 1D CNN followed by BiLSTM and a spatial branch using a 2D CNN, but the implementation paragraph says 'We use Resnet34, pre-trained by ImageNet, as backnone of the feature extraction layer' without specifying which branch this replaces or how a 1D ECG signal is adapted for a 2D ImageNet-pretrained network. Additionally, Eq. (21) defines a square-root decay learning-rate schedule while the text states a cosine annealing schedule. These details are essential for reproducibility and must be resolved.
[§4.2, Table 2] All reported results are single runs without error bars, confidence intervals, or statistical significance tests. The ECG200 test set contains only 100 samples, so the claimed 2.0-point accuracy improvement over the best baseline corresponds to two additional correct predictions. To support the claim that GAF-FusionNet 'consistently outperforms all baseline methods,' the authors should report results over multiple random seeds (or a paired test over the same test folds) and provide variance or significance information.

minor comments (6)

[Abstract] The code link is given as 'will soon be available,' but no code is currently provided; please make the code and data split publicly available at the time of publication so the experiments can be reproduced.
[Figure 1] Figure 1 is too coarse to verify the details of the dual-layer attention module; please annotate the tensor shapes and show explicitly where 'Split 1' through 'Split r' and 'Global pooling' correspond to the operations in Eqs. (12)–(17).
[§5, Conclusion] The concluding sentence mentions 'aiding in the understanding and treatment of psychiatric disorders'; this appears unrelated to the ECG classification task and should be corrected (probably to cardiovascular conditions).
[References] Several in-text citation names do not match the reference list entries (e.g., 'Wei et al.' vs. Guo et al. [7], 'Satria et al.' vs. Mandala et al. [15], 'Michal et al.' vs. Heldeweg et al. [9], 'Madeline et al.' vs. Kent et al. [11]). Please align all citations and reference entries.
[§3.1, Eq. (3)] The segmentation formula uses an overlap parameter o, but it is not stated how the window length w and overlap o are chosen for each dataset, nor whether the final window is padded when the signal ends; please specify these choices.
[§4.1, Implementation Details] The text says training was performed on 'an NVIDIA RTX 4090 GPU with 128GB memory'; the RTX 4090 has 24GB of memory, so this hardware description is inaccurate and should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity detected: the paper is an empirical benchmark study whose reported accuracies are produced by training, not by redefinition of inputs.

full rationale

GAF-FusionNet is an empirical benchmark paper, not a derivation paper. The GAF transform is taken from the independent prior work of Wang and Oates, the split-attention fusion is a new architectural contribution, and the reported accuracies in Table 2 are outcomes of training and evaluation rather than quantities constructed from fitted parameters or from the labels themselves. The 'consistent outperformance' claim depends on the baseline comparisons, but there is no evidence in the manuscript that any baseline number is a renamed version of the proposed model's own output, nor does the paper fit any parameter to a subset of data and then present a closely related quantity as a prediction. The self-citations in the reference list (e.g., Qin, Zong, and Liu [17]; Liu [14]) are contextual and are not load-bearing for the central experimental claim. The absence of an explicit statement about the inter-patient versus intra-patient MIT-BIH split is a serious experimental-protocol concern that bears on generalization validity, but it is not a circularity: a leaked split would inflate accuracy through memorization, not through the reported equations reducing to their own inputs. No equation in Section 3 is equivalent to another by construction, and no uniqueness theorem or prior-work premise is invoked to force the architecture. Therefore, under the hard rules, no circular step can be exhibited, and the appropriate score is 0.

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

The model's central claims rest on the correctness of the MIT-BIH train/test split, on unspecified hyperparameters, and on the transfer of ImageNet pretraining to GAF images. None of these are externally justified or documented in the paper.

free parameters (4)

Window length w = not reported
The GAF image size and the temporal branch input depend on w through Equation (3), but the experiments never state its value.
Overlap o = not reported
The segmentation step in Equation (3) uses overlap o to determine the number of segments, but the value is not given.
Network dimensions and layer counts = not reported
The number of CNN layers, filter sizes, and latent dimensions d_t and d_s are not specified, so the model architecture is incomplete.
Learning rate schedule = contradictory
Equation (21) specifies an inverse-square-root decay while the text says cosine annealing is used; the actual schedule is a hand-chosen, unreported parameter.

assumptions (4)

domain assumption MIT-BIH train/test split is beat-wise independent and does not leak patient information
The paper never states that recordings from the same patient are kept out of the training set. Without patient separation, the 99.6% accuracy claim is invalid. This is load-bearing.
domain assumption ImageNet-pretrained ResNet34 transfers to GAF images
Section 4.1 says ResNet34 pretrained on ImageNet is used as the feature extraction backbone, but no evidence is given that this helps on single-channel GAF images built from 96 or 140 time points.
domain assumption Gramian Angular Fields preserve temporal correlations relevant to ECG classification
The paper relies on the standard GAF claim from Wang and Oates [23] that temporal dependencies are encoded in the image, but no analysis or ablation supports this specifically for ECG.
domain assumption Baseline methods are evaluated under the same protocol as GAF-FusionNet
Table 2 compares against baselines, but the paper does not state whether all methods used identical preprocessing, segmentation, and train/test splits. Unequal protocols would invalidate the comparison.

how reviews work

0 comments

Cite this review

Pith. "Pith review of GAF-FusionNet: Multimodal ECG Analysis via Gramian Angular Fields and Split Attention." pith.science (2026). https://pith.science/paper/QVPHHL5T

@misc{pith2026250101960,
  author       = {Pith},
  title        = {Pith review of: GAF-FusionNet: Multimodal ECG Analysis via Gramian Angular Fields and Split Attention},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QVPHHL5T}},
  note         = {Machine review of arXiv:2501.01960}
}

read the original abstract

Electrocardiogram (ECG) analysis plays a crucial role in diagnosing cardiovascular diseases, but accurate interpretation of these complex signals remains challenging. This paper introduces a novel multimodal framework(GAF-FusionNet) for ECG classification that integrates time-series analysis with image-based representation using Gramian Angular Fields (GAF). Our approach employs a dual-layer cross-channel split attention module to adaptively fuse temporal and spatial features, enabling nuanced integration of complementary information. We evaluate GAF-FusionNet on three diverse ECG datasets: ECG200, ECG5000, and the MIT-BIH Arrhythmia Database. Results demonstrate significant improvements over state-of-the-art methods, with our model achieving 94.5\%, 96.9\%, and 99.6\% accuracy on the respective datasets. Our code will soon be available at https://github.com/Cross-Innovation-Lab/GAF-FusionNet.git.

Figures

Figures reproduced from arXiv: 2501.01960 by the authors.

**Figure 1.** Overview of GAF-FusionNet. where H(X) is the impulse response of the Butterworth filter, and ∗ denotes convolution. 2. Normalization: We normalize the filtered signal to zero mean and unit variance: Xnorm = Xf iltered − µ(Xf iltered) σ(Xf iltered) (2) where µ(·) and σ(·) denote mean and standard deviation, respectively. 3. Segmentation: We segment the normalized signal into fixed-length windows of size w with an ov… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 14 canonical work pages

[1]

Journal of Neural Engineering16(3), 031001 (2019)

Craik, A., He, Y., Contreras-Vidal, J.L.: Deep learning for electroencephalogram (EEG) classification tasks: A review. Journal of Neural Engineering16(3), 031001 (2019). https://doi.org/10.1088/1741-2552/ab0ab5

doi:10.1088/1741-2552/ab0ab5 2019
[2]

Journal of Biomedical Informatics 141, 104338 (2023)

Cremonesi, F., Planat, V., Kalokyri, V., Kondylakis, H., Sanavia, T., Miguel Mateos Resinas, V., Singh, B., Uribe, S.: The need for multi- modal health data modeling: A practical approach for a federated-learning healthcare platform. Journal of Biomedical Informatics 141, 104338 (2023). https://doi.org/https://doi.org/10.1016/j.jbi.2023.104338, https://ww...

arXiv 2023
[3]

Dau, H.A., Keogh, E., Kaveh, K., Yeh, C.C.M., Yan, Z., Shaghayegh, G., Ann, R.C., Yanping, Hu, B., Begum, N., Anthony, B., Abdullah, M., Gustavo, B., Hexagon-ML: The ucr time series classification archive (October 2018)

work page 2018
[4]

Nursing Open 11, e2225 (2024)

Emmett, A., Kent, B., James, A., March-McDonald, J.: Experiences of health pro- fessionals towards using mobile electrocardiogram (ecg) technology: A qualitative systematic review. Nursing Open 11, e2225 (2024). https://doi.org/10.1111/ jocn.16434

work page 2024
[5]

Scientific Reports 7(1), 10543 (2017)

Garcia, G., Moreira, G., Menotti, D., Luz, E.: Inter-patient ecg heartbeat clas- sification with temporal vcg optimized by pso. Scientific Reports 7(1), 10543 (2017). https://doi.org/10.1038/s41598-017-09837-3, https://doi.org/10. 1038/s41598-017-09837-3

work page doi:10.1038/s41598-017-09837-3 2017
[6]

Journal of Neuroscience Methods 410, 110223 (2024)

Gunda, N.K., Khalaf, M.I., Bhatnagar, S., Quraishi, A., Gudala, L., Venkata, A.K.P., Alghayadh, F.Y., Alsubai, S., Bhatnagar, V.: Lightweight attention mech- anisms for eeg emotion recognition for brain computer interface. Journal of Neuroscience Methods 410, 110223 (2024). https://doi.org/https://doi.org/ 10.1016/j.jneumeth.2024.110223, https://www.scien...

work page arXiv 2024
[7]

IEEE Access 7, 134236–134244 (2019)

Guo,W.,Ge,W.,Cui,L.,Li,H.,Kong,L.:Aninterpretablediseaseonsetpredictive model using crossover attention mechanism from electronic health records. IEEE Access 7, 134236–134244 (2019)

work page 2019
[8]

Nature Medicine25(1), 65–69 (2019)

Hannun, A.Y., Rajpurkar, P., Haghpanahi, M., Tison, G.H., Bourn, C., Turakhia, M.P., Ng, A.Y.: Cardiologist-level arrhythmia detection and classification in am- bulatory electrocardiograms using a deep neural network. Nature Medicine25(1), 65–69 (2019)

work page 2019

Show all 27 references

[9]

Critical Care20, 1–9 (2016)

Heldeweg, M.L.A., Liu, N., Koh, Z.X., Fook-Chong, S., Lye, W.K., Harms, M., Ong, M.E.H.: A novel cardiovascular risk stratification model incorporating ecg and heart rate variability for patients presenting to the emergency department with chest pain. Critical Care20, 1–9 (2016)

2016
[10]

Karim, F., Majumdar, S., Darabi, H., Harford, S.: Multivariate LSTM-FCNs for timeseriesclassification.In:NeuralNetworks.vol.116,pp.237–245.Elsevier(2019)

2019
[11]

Journal of Electrocardiology80, 24–33 (Sep GAF-FusionNet 13 2023)

Kent, M., Vasconcelos, L., Ansari, S., Ghanbari, H., Nenadic, I.: Fourier space approach for convolutional neural network (CNN) electrocardiogram (ECG) clas- sification: A proof-of-concept study. Journal of Electrocardiology80, 24–33 (Sep GAF-FusionNet 13 2023). https://doi.or...

2023 doi
[12]

arXiv preprint arXiv:1412.6980 (2014)

Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

2014 arXiv
[13]

In: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Process- ing (ICASSP)

Liu, C., Wan, Z., Cheng, S., Zhang, M., Arcucci, R.: ETP: Learning Trans- ferable ECG Representations via ECG-Text Pre-Training. In: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Process- ing (ICASSP). pp. 8230–8234. IEEE, Seoul, Korea, Repub...

2024
[14]

CAAI Artificial Intelligence Research 3, 9150040 (2024)

Liu, F.: Artificial intelligence in emotion quantification : A prospective overview. CAAI Artificial Intelligence Research 3, 9150040 (2024). https:// doi.org/10.26599/AIR.2024.9150040, https://www.sciopen.com/article/10. 26599/AIR.2024.9150040

2024
[15]

PLOS ONE 19(4), e0297551 (Apr 2024)

Mandala, S., Rizal, A., Adiwijaya, Nurmaini, S., Amini, S.S., Sudarisman, G.A., Hau, Y.W., Abdullah, A.H.: An improved method to detect arrhythmia using ensemble learning-based model in multi lead electrocardiogram (ECG). PLOS ONE 19(4), e0297551 (Apr 2024). https://doi.org/10...

2024 doi
[16]

IEEE engineering in medicine and biology magazine20(3), 45–50 (2001)

Moody, G.B., Mark, R.G.: The impact of the mit-bih arrhythmia database. IEEE engineering in medicine and biology magazine20(3), 45–50 (2001)

2001
[17]

Applied Sciences14(17) (2024)

Qin, J., Zong, L., Liu, F.: Exploring inner speech recognition via cross-perception approach in eeg and fmri. Applied Sciences14(17) (2024). https://doi.org/10. 3390/app14177720, https://www.mdpi.com/2076-3417/14/17/7720

2024
[18]

Nature communications 11(1), 1760 (2020)

Ribeiro, A.H., Ribeiro, M.H., Paixão, G.M., Oliveira, D.M., Gomes, P.R., Canazart, J.A., Ferreira, M.P., Andersson, C.R., Macfarlane, P.W., Meira Jr, W., et al.: Automatic diagnosis of the 12-lead ecg using a deep neural network. Nature communications 11(1), 1760 (2020)

2020
[19]

Information Fusion 102, 102040 (2024)

Shaik, T., Tao, X., Li, L., Xie, H., Velásquez, J.D.: A survey of multi- modal information fusion for smart healthcare: Mapping the journey from data to wisdom. Information Fusion 102, 102040 (2024). https://doi.org/https: //doi.org/10.1016/j.inffus.2023.102040, https://www.sc...

2024
[20]

In: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R

Vaswani,A.,Shazeer,N.,Parmar,N.,Uszkoreit,J.,Jones,L.,Gomez,A.N.,Kaiser, L.u., Polosukhin, I.: Attention is all you need. In: Guyon, I., Luxburg, U.V., Bengio, S., Wallach, H., Fergus, R., Vishwanathan, S., Garnett, R. (eds.) Advances in Neural Information Processing Systems. ...

2017
[21]

IEEE Journal of Biomedical and Health Informatics24(9), 2461–2472 (2020)

Wang, R., Fan, J., Li, Y.: Deep multi-scale fusion neural network for multi-class arrhythmia detection. IEEE Journal of Biomedical and Health Informatics24(9), 2461–2472 (2020). https://doi.org/10.1109/JBHI.2020.2981526

2020
[22]

Com- puters in Biology and Medicine 155, 106641 (2023)

Wang, Z., Stavrakis, S., Yao, B.: Hierarchical deep learning with generative adversarial network for automatic cardiac diagnosis from ecg signals. Com- puters in Biology and Medicine 155, 106641 (2023). https://doi.org/https: //doi.org/10.1016/j.compbiomed.2023.106641, https:/...

2023
[23]

In: Workshops at the Twenty-Ninth AAAI Conference on Artificial Intelligence (2015),https://api

Wang, Z., Oates, T.: Encoding time series as images for visual inspection and classification using tiled convolutional neural networks. In: Workshops at the Twenty-Ninth AAAI Conference on Artificial Intelligence (2015),https://api. semanticscholar.org/CorpusID:16409971 14 J. ...

2015
[24]

JAMA Internal Medicine (Jul 2024)

Yagi, R., Mori, Y., Goto, S., Iwami, T., Inoue, K.: Routine Electrocardiogram Screening and Cardiovascular Disease Events in Adults. JAMA Internal Medicine (Jul 2024). https://doi.org/10.1001/jamainternmed.2024.2270, https://doi. org/10.1001/jamainternmed.2024.2270

2024
[25]

In: Proceedings of the CHI Conference on Human Factors in Comput- ing Systems

Yildirim, N., Richardson, H., Wetscherek, M.T., Bajwa, J., Jacob, J., Pinnock, M.A., Harris, S., Coelho De Castro, D., Bannur, S., Hyland, S., Ghosh, P., Ran- jit, M., Bouzid, K., Schwaighofer, A., Pérez-García, F., Sharma, H., Oktay, O., Lungren, M., Alvarez-Valle, J., Nori, ...

2024
[26]

Applied Soft Computing 165, 112056 (2024)

Zeng, W., Shan, L., Yuan, C., Du, S.: Advancing cardiac diagnostics: Ex- ceptional accuracy in abnormal ecg signal classification with cascading deep learning and explainability analysis. Applied Soft Computing 165, 112056 (2024). https://doi.org/https://doi.org/10.1016/j.asoc...

2024
[27]

In: Proceed- ings of the AAAI conference on artificial intelligence

Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., Zhang, W.: Informer: Beyond efficient transformer for long sequence time-series forecasting. In: Proceed- ings of the AAAI conference on artificial intelligence. vol. 35, pp. 11106–11115 (2021)

2021

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

Journal of Neural Engineering16(3), 031001 (2019)

Craik, A., He, Y., Contreras-Vidal, J.L.: Deep learning for electroencephalogram (EEG) classification tasks: A review. Journal of Neural Engineering16(3), 031001 (2019). https://doi.org/10.1088/1741-2552/ab0ab5

doi:10.1088/1741-2552/ab0ab5 2019

[2] [2]

Journal of Biomedical Informatics 141, 104338 (2023)

Cremonesi, F., Planat, V., Kalokyri, V., Kondylakis, H., Sanavia, T., Miguel Mateos Resinas, V., Singh, B., Uribe, S.: The need for multi- modal health data modeling: A practical approach for a federated-learning healthcare platform. Journal of Biomedical Informatics 141, 104338 (2023). https://doi.org/https://doi.org/10.1016/j.jbi.2023.104338, https://ww...

arXiv 2023

[3] [3]

Dau, H.A., Keogh, E., Kaveh, K., Yeh, C.C.M., Yan, Z., Shaghayegh, G., Ann, R.C., Yanping, Hu, B., Begum, N., Anthony, B., Abdullah, M., Gustavo, B., Hexagon-ML: The ucr time series classification archive (October 2018)

work page 2018

[4] [4]

Nursing Open 11, e2225 (2024)

Emmett, A., Kent, B., James, A., March-McDonald, J.: Experiences of health pro- fessionals towards using mobile electrocardiogram (ecg) technology: A qualitative systematic review. Nursing Open 11, e2225 (2024). https://doi.org/10.1111/ jocn.16434

work page 2024

[5] [5]

Scientific Reports 7(1), 10543 (2017)

Garcia, G., Moreira, G., Menotti, D., Luz, E.: Inter-patient ecg heartbeat clas- sification with temporal vcg optimized by pso. Scientific Reports 7(1), 10543 (2017). https://doi.org/10.1038/s41598-017-09837-3, https://doi.org/10. 1038/s41598-017-09837-3

work page doi:10.1038/s41598-017-09837-3 2017

[6] [6]

Journal of Neuroscience Methods 410, 110223 (2024)

Gunda, N.K., Khalaf, M.I., Bhatnagar, S., Quraishi, A., Gudala, L., Venkata, A.K.P., Alghayadh, F.Y., Alsubai, S., Bhatnagar, V.: Lightweight attention mech- anisms for eeg emotion recognition for brain computer interface. Journal of Neuroscience Methods 410, 110223 (2024). https://doi.org/https://doi.org/ 10.1016/j.jneumeth.2024.110223, https://www.scien...

work page arXiv 2024

[7] [7]

IEEE Access 7, 134236–134244 (2019)

Guo,W.,Ge,W.,Cui,L.,Li,H.,Kong,L.:Aninterpretablediseaseonsetpredictive model using crossover attention mechanism from electronic health records. IEEE Access 7, 134236–134244 (2019)

work page 2019

[8] [8]

Nature Medicine25(1), 65–69 (2019)

Hannun, A.Y., Rajpurkar, P., Haghpanahi, M., Tison, G.H., Bourn, C., Turakhia, M.P., Ng, A.Y.: Cardiologist-level arrhythmia detection and classification in am- bulatory electrocardiograms using a deep neural network. Nature Medicine25(1), 65–69 (2019)

work page 2019