Pith. sign in

REVIEW 5 major objections 6 minor 1 cited by

Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning

T0 review · 5 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read A dual-view model that reads irregular medical time series as both sequences and images outperforms seven comparison methods on three clinical datasets.

desk verdict A promising joint sequence-image framework for irregular clinical time series with strong empirical results, but the image construction pipeline is underspecified and two of the 'self-supervised' losses lean on labels. read the letter →

arxiv 2502.06134 v1 pith:O234CACN submitted 2025-02-10 cs.CV cs.AI

Liuqing Chen , Shuhong Xiao , Shixian Ding , Shanhai Hu , Lingyun Sun This is my paper · ORCID
classification cs.CVcs.AI
keywords irregularmedicaltimeseriesself-supervisedlearningsequenceandimagemodelingjointrepresentationimputationcontrastiveclusteringlossclinicalclassification
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 claims that irregular medical time series with heavy missing values are better classified by a model that reads the data twice: once as a sequence and once as an image. The sequence branch imputes missing entries with a generator–discriminator pair, while the image branch turns the series into six visual encodings (line graph, frequency spectrum, Gramian angular fields, Markov transition fields, recurrence plot) and processes them with a pretrained Swin Transformer. Three self-supervised losses — an inter-sequence contrastive loss, a margin-based sequence-image contrastive loss, and a clustering loss — align the two views into one joint representation. On the PAM, P12, and P19 clinical datasets, the fused model beats seven comparison methods, with the largest margin on the most heavily missing dataset (P19, 94.9% missing). The practical claim is that joint sequence-image learning is a robust recipe for clinical classification when data are sparse.

What carries the argument

The key machinery is a two-branch architecture with three alignment losses. The sequence branch is a generator–discriminator pair: a bidirectional RNN generator with a time-decay term imputes missing values, and the discriminator tries to tell real from imputed entries. Reconstruction is trained with an NT-Xent loss rather than MSE, which the paper argues prevents the generator from collapsing under severe missingness. The image branch converts each series into six RGB images — line graph, frequency spectrum, Gramian angular summation/difference fields, Markov transition field, and recurrence plot — and encodes them with a pretrained Swin Transformer. A joint projection maps sequence and image representations into a shared space, where a margin-based sequence-image contrastive loss and a K-means clustering loss pull the two views together and separate classes. The fused feature is classified with cross-entropy loss on PAM and focal loss on P12 and P19.

What would settle it

Re-run the image-only and fused ablations with a stated rule for missing entries (zero-fill or drop) using images built from the raw sparse series; if the fused model's F1 gain over the best single view falls below the reported 2.6 points, the fusion advantage over the image branch is falsified.

Watch

Extended reading notes

Core claim

The central discovery, on the paper's own terms, is that the joint sequence-image representation trained with the three self-supervised losses outperforms all seven comparison methods on all three datasets: on PAM it raises accuracy, precision, recall, and F1 by 2.3–3.1 points over the second-best method, and on P19 it raises AUPRC by 5.8 points and AUROC by 2.3 points. The paper further reports that under simulated missingness on PAM, the fused model degrades less than the baselines, staying above 80% on all metrics even when 50% of sensors are masked. These results are presented as evidence that the two views carry complementary information and that the SSL losses are what fuse them effectively.

Load-bearing premise

The image branch assumes that six image transforms built from series with up to 94.9% missing values remain informative to a pretrained Swin Transformer, and the paper never states how missing entries are handled during image construction or whether the images come from the raw sparse series or from the imputed sequence.

Editorial extensions

If this is right

  • On the PAM, P12, and P19 benchmarks under the 8:1:1 split, the fused model becomes the reference point that future methods must beat.
  • Each of the three SSL losses contributes independently: removing the contrastive step or the clustering step lowers PAM F1 by about one point relative to the full model.
  • Under leave-sensors-out missingness at 50%, the method keeps all metrics above 80% and outscores the second-best method by 6.1% accuracy, indicating the representation is robust to whole sensors disappearing.
  • The clinical Turing test on P19 suggests that the imputed signals are not distinguishable from linear-interpolation-imputed signals by experienced ICU clinicians, which supports the clinical plausibility of the imputed values.

Reading between the lines

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

  • The paper never states whether the six images are built from the raw sparse series or from the imputed sequence X'; if the latter, the image branch is not an independent view and the reported fusion gain may largely inherit the imputation quality.
  • Because the image encoder is a fixed ImageNet-pretrained Swin Transformer, the method's success may depend on visual features that transfer well from natural images to physiological plots; this transferability is untested on non-clinical irregular series.
  • The robustness experiments are only run on PAM at a base 60% missing rate; whether the fusion advantage persists at P12/P19's 88–95% baseline missingness remains an open extension.
  • A direct ablation that varies the number of image transforms (from one to six) would show whether the gain comes from the fusion itself or from the sheer variety of encodings.
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

5 major / 6 minor

Summary. The paper proposes a joint learning framework for irregular medical time series that combines a sequence branch (a generator-discriminator imputation model with a BiRNN) and an image branch (six time-series-to-image transforms processed by a pretrained Swin Transformer). Three additional losses are introduced: an inter-sequence NT-Xent reconstruction loss, a sequence-image contrastive loss with a margin, and a clustering loss on the fused joint representations. The authors report state-of-the-art classification results on PAM, P12, and P19, robustness gains under leave-sensors-out and leave-samples-out missingness on PAM, an ablation study, and a clinical Turing test. The paper currently appears as an arXiv v1 that refers to appendices A-D, which are not present in the manuscript text provided, leaving several load-bearing details unspecified.

Significance. If the reported results are reproducible and the three losses are correctly characterized, the paper would make a strong empirical contribution: it demonstrates consistent gains over seven baselines on three clinical datasets, with five-run statistics and a robustness analysis under two missingness scenarios. The code link and the clinical Turing test are additional positive elements. However, the current manuscript leaves essential implementation details in absent appendices, and one of the three 'self-supervised' losses requires labels, so the significance can only be assessed after those gaps are closed. The core idea of fusing sequence and image representations for irregular time series is plausible and worth pursuing, but the paper as written does not yet provide a fully checkable account of its own method.

major comments (5)
  1. [Imaging Time Series / Table 1] The image construction is underspecified in a load-bearing way. The text states that six image transforms are used and defers details to Appendix A, but Appendix A is absent from the arXiv v1. With missing rates up to 94.9% (P19), the Gramian Angular Field, Markov Transition Field, and recurrence plot transforms require dense series to compute pairwise quantities, yet the paper never states whether the images are built from the raw sparse X, from a zero-filled/interpolated version, or from the imputed sequence X' produced by the generator G. If images are built from X', then the image branch is not independent of the sequence branch, the image-only ablation (Table 4, F1 95.4) is not well-defined without G, and because the final objective includes the classification loss Lclf, label information can flow into X' and then into the images, making the claimed self-supervised image signal partially label-leaking. The authors must specify the exact image construction pipeline, including missing-value handling, and state clearly whether G is updated by Lclf.
  2. [Eq. (9)] The sequence-image contrastive loss is described as 'self-supervised,' but the definition uses P(i), the set of all sample indices in the same category as sample i. This requires ground-truth labels, making it a supervised contrastive loss, not a self-supervised loss. This is not merely a terminological issue: it means that one of the three claimed SSL strategies uses labels during representation learning, and if those labels are the same classification labels used in Lclf, the representation may benefit from task supervision beyond the final classifier. Please relabel the loss and clarify whether the labels used to form P(i) are the training labels from the classification task.
  3. [Eq. (12)] Equation (12) defines the cluster center update as C_k = argmin_{u in S_k} sum_{u' in S_k} ||u - u'||^2, which selects the medoid of the cluster, not the K-means mean. The text repeatedly calls the procedure K-means and Eq. (10) is a standard K-means assignment step, so the algorithm as written is inconsistent. If the medoid update is intentional, the name should be changed; if K-means is intended, Eq. (12) should be the arithmetic mean of the members of S_k. This matters because the clustering loss in Eq. (11) is defined with respect to these centers.
  4. [Eqs. (10)-(12)] The clustering loss computes cluster assignments and centers from the same fused features u that are being optimized by L_cluster. This self-referential loop can lead to trivial or degenerate solutions (for example, all points assigned to one cluster, or centers collapsing to a single point) unless safeguards are specified. The paper does not state how k is chosen, whether cluster centers are reinitialized, whether empty clusters are handled, or what convergence threshold tau_c is used. The final classification is evaluated on held-out test data, so the central claim is not circular, but the clustering ablation's reported gain is not trustworthy without a demonstration that the clustering procedure avoids collapse.
  5. [Appendix availability / Implementation and Training] Several parts essential for reproducibility are said to be in appendices, but Appendix A (image transform details), Appendix B (hyperparameters and learning rates), Appendix C (baseline implementations), and Appendix D (additional experimental results) are not included in the arXiv v1 text provided. The Implementation and Training paragraph and the comparison section refer to these absent appendices for details such as learning rates and baseline settings. As a result, the exact configuration used to produce Table 2 cannot be reconstructed from the paper alone. The authors should include these appendices in the manuscript or in a supplementary document.
minor comments (6)
  1. [Contributions bullet] The third contribution bullet contains a grammatical error: 'We also simulates two classic types of missingness' should be 'We also simulate two classic types of missingness.'
  2. [Eq. (3)] The notation x_t- is used for 'the closest observation prior to the current missing value,' but the indexing is not formally defined; please clarify how x_t- is selected for each missing position.
  3. [Eq. (4)] The function 'Drop' is not defined; if it denotes dropout, please state the dropout rate or say that it is fixed in the appendix.
  4. [Eq. (6)] The NT-Xent formulation says 'Given 2B pairs (zi,zj)' but does not explicitly define which pairs are positive; in Eqs. (7) the intended pairs are clear from context, but the definition should be written out.
  5. [Clinical Turing tests] The description of the 'real measured samples' is confusing: 'half imputed using linear interpolation as real measured samples' seems to mean that linearly interpolated signals are labeled as real, but this should be stated explicitly, and the sample size of 60 is quite small for a Turing-test claim.
  6. [Table 1 / Introduction] The introduction cites an 80% missing rate for PhysioNet2012, while Table 1 reports 88.4% for P12 and 94.9% for P19; please reconcile these numbers or clarify that they refer to different datasets or preprocessing settings.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central empirical claim is evaluated on held-out test data against external baselines; the SSL losses are training objectives rather than predictions derived from their own outputs.

full rationale

The paper's strongest claim is that its joint sequence–image framework outperforms seven state-of-the-art baselines on three public clinical datasets. This claim rests on Table 2, which reports metrics computed on an 8:1:1 held-out split, and on robustness experiments that mask additional test-set observations. No parameter is fitted to a subset and then renamed as a prediction of a closely related quantity; the final classification loss Lclf is supervised by ground-truth labels, and the reported numbers are test-set metrics, not training objectives. The three self-supervised losses are training objectives. In particular, the clustering loss (Eqs. 10–12) derives its cluster targets from the same fused features that are being optimized, which is a standard self-training/EM-style mechanism (like DeepCluster) rather than a circular derivation of the paper's conclusions; the model's downstream performance is still assessed on data and labels not used to form those cluster targets. The sequence–image contrastive loss (Eq. 9) uses category labels as margins, so it is supervised in part, but that is a labeling issue, not circularity. There is also no load-bearing self-citation: the authors do not cite themselves for the key architectural choices, and the prior works cited for imputation, image transforms, and Swin Transformer are external. The main caveat is underspecification: the paper does not state whether the six image transforms are computed from the raw sparse series or from the imputed X', and Appendix A is absent from the arXiv v1. If the images were built from X', the 'image-only' ablation would not be fully independent of the sequence branch. However, this is a reproducibility and experimental-design concern, not a step where a claimed result reduces to its input by construction. Under the rules requiring a quotable reduction (Eq. X = Eq. Y, or fitted input called prediction), no such circular step is present. Therefore the appropriate finding is no significant circularity.

Assumptions & free parameters 7 free parameters · 5 assumptions · 0 invented entities

The framework relies on several hand-tuned hyperparameters (alpha, beta1, beta2, tau, m, k, tau_c), none of whose values appear in the provided text, and on a fixed ImageNet-pretrained Swin Transformer that is not trained on medical images. The central classification result is not machine-checked and no dataset artifacts are shipped.

free parameters (7)
  • alpha (reconstruction loss weight) = unknown
    Weights Lrec in the generator objective; chosen by hand, value not reported in the provided text.
  • beta1 (contrastive loss weight) = unknown
    Weights Lcont in the final objective; chosen by hand, value not reported.
  • beta2 (clustering loss weight) = unknown
    Weights Lcluster in the final objective; chosen by hand, value not reported.
  • tau (temperature) = unknown
    Used in NT-Xent and contrastive/clustering losses; chosen by hand, value not reported.
  • m (margin) = unknown
    Margin on negative pairs in Eq. 9; chosen by hand, value not reported.
  • k (number of clusters) = unknown
    K-means cluster count in the clustering loss; chosen by hand, value not reported.
  • tau_c (clustering convergence threshold) = unknown
    Threshold for cluster update in Eq. 12; chosen by hand, value not reported.
assumptions (5)
  • domain assumption Randomly masking test data simulates the two real-world missingness types
    Used in leave-sensors-out and leave-samples-out experiments; assumes the masking distribution matches clinical missingness.
  • domain assumption Pretrained Swin Transformer features on ImageNet transfer to medical time series images
    The image encoder is fixed from ImageNet-21K pretraining; no fine-tuning on medical images is described beyond the joint projection.
  • domain assumption The adversarial generator-discriminator training improves imputation without collapse
    Standard GAN training assumptions; the paper reports no evidence of training stability for the adversarial objective.
  • ad hoc to paper Cluster assignments computed from the features themselves provide a valid training target
    Eq. 10-12 derive cluster centers from the fused features u, then pull u toward those centers, a self-referential target.
  • domain assumption NT-Xent between original and imputed sequences is computable despite missing entries
    Lrec uses NT-Xent(X, X'); the original X contains missing values and the paper does not state how the similarity is masked.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning." pith.science (2026). https://pith.science/paper/O234CACN

@misc{pith2026250206134,
  author       = {Pith},
  title        = {Pith review of: Integrating Sequence and Image Modeling in Irregular Medical Time Series Through Self-Supervised Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O234CACN}},
  note         = {Machine review of arXiv:2502.06134}
}
read the original abstract

Medical time series are often irregular and face significant missingness, posing challenges for data analysis and clinical decision-making. Existing methods typically adopt a single modeling perspective, either treating series data as sequences or transforming them into image representations for further classification. In this paper, we propose a joint learning framework that incorporates both sequence and image representations. We also design three self-supervised learning strategies to facilitate the fusion of sequence and image representations, capturing a more generalizable joint representation. The results indicate that our approach outperforms seven other state-of-the-art models in three representative real-world clinical datasets. We further validate our approach by simulating two major types of real-world missingness through leave-sensors-out and leave-samples-out techniques. The results demonstrate that our approach is more robust and significantly surpasses other baselines in terms of classification performance.

Figures

Figures reproduced from arXiv: 2502.06134 by the authors.

Figure 1
Figure 1. The framework of our approach. datasets: PAM (Reiss and Stricker 2012), P12 (Goldberger et al. 2000), and P19 (Reyna et al. 2020) [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Performance under increased missingness: (a) leave-sensors-out and (b) leave-samples-out on the PAM dataset. Tests [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗

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. A Comprehensive Survey of Electronic Health Record Modeling: From Deep Learning Approaches to Large Language Models

    cs.LG 2025-07 reject novelty 4.0 of 10

    A survey that taxonomizes EHR modeling research into data-centric, architectural, learning-focused, multimodal, and LLM-based categories, with datasets and metrics.

Reference graph

Works this paper leans on

46 extracted references · 38 canonical work pages · cited by 1 Pith paper

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Ao, R.; and He, G. 2023. Image based deep learning in 12-lead ECG diagnosis. Frontiers in Artificial Intelligence, 5: 1087370

    work page 2023

  4. [4]

    M.; Hawryluk, I.; Prete Jr, C

    Brizzi, A.; Whittaker, C.; Servo, L. M.; Hawryluk, I.; Prete Jr, C. A.; de Souza, W. M.; Aguiar, R. S.; Araujo, L. J.; Bastos, L. S.; Blenkinsop, A.; et al. 2022. Spatial and temporal fluctuations in COVID-19 fatality rates in Brazilian hospitals. Nature medicine, 28(7): 1476--1485

    work page 2022

  5. [5]

    Chaudhary, K.; Vaid, A.; Duffy, \'A .; Paranjpe, I.; Jaladanki, S.; Paranjpe, M.; Johnson, K.; Gokhale, A.; Pattharanitima, P.; Chauhan, K.; et al. 2020. Utilization of deep learning for subphenotype identification in sepsis-associated acute kidney injury. Clinical Journal of the American Society of Nephrology, 15(11): 1557--1565

    work page 2020

  6. [6]

    Che, Z.; Purushotham, S.; Cho, K.; Sontag, D.; and Liu, Y. 2018. Recurrent neural networks for multivariate time series with missing values. Scientific reports, 8(1): 6085

    2018

  7. [7]

    Chen, T.; Kornblith, S.; Swersky, K.; Norouzi, M.; and Hinton, G. E. 2020. Big self-supervised models are strong semi-supervised learners. Advances in neural information processing systems, 33: 22243--22255

    work page 2020

  8. [8]

    Chen, Y.; Ren, K.; Wang, Y.; Fang, Y.; Sun, W.; and Li, D. 2024. Contiformer: Continuous-time transformer for irregular time series modeling. Advances in Neural Information Processing Systems, 36

    work page 2024

Show all 46 references
  1. [9]

    T.; Schuetz, A.; Stewart, W

    Choi, E.; Bahadori, M. T.; Schuetz, A.; Stewart, W. F.; and Sun, J. 2016. Doctor ai: Predicting clinical events via recurrent neural networks. In Machine learning for healthcare conference, 301--318. PMLR

    2016

  2. [10]

    Chong, U.-P.; et al. 2011. Signal model-based fault detection and diagnosis for induction motors using features of vibration signal in two-dimension domain. Strojni s ki vestnik , 57(9): 655--666

    2011

  3. [11]

    R.; Li, J.; Zhang, X.; Hong, D.; Gupta, R

    Chowdhury, R. R.; Li, J.; Zhang, X.; Hong, D.; Gupta, R. K.; and Shang, J. 2023. Primenet: Pre-training for irregular multivariate time series. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 7184--7192

    2023

  4. [12]

    A.; Wu, P.; Karki, R.; Sood, M.; Godard, P.; Ahmad, A.; Vrooman, H.; Hofmann-Apitius, M.; and Fr \"o hlich, H

    de Jong, J.; Emon, M. A.; Wu, P.; Karki, R.; Sood, M.; Godard, P.; Ahmad, A.; Vrooman, H.; Hofmann-Apitius, M.; and Fr \"o hlich, H. 2019. Deep learning for clustering of multivariate clinical patient trajectories with missing values. GigaScience, 8(11): giz134

    2019

  5. [13]

    Deng, Y.; Hua, M.; Yingjun, R.; Fanyue, Q.; and Di, P. 2023. An image characterisation method for AHU fault diagnosis based on residual neural networks. In Proceedings of Building Simulation 2023: 18th Conference of IBPSA, volume 18 of Building Simulation, 3827--3834. Shanghai...

    2023

  6. [14]

    Dey, R.; and Salem, F. M. 2017. Gate-variants of gated recurrent unit (GRU) neural networks. In 2017 IEEE 60th international midwest symposium on circuits and systems (MWSCAS), 1597--1600. IEEE

    2017

  7. [15]

    a r, M.; Sch \

    Gillette, K.; Gsell, M. A.; Nagel, C.; Bender, J.; Winkler, B.; Williams, S. E.; B \"a r, M.; Sch \"a ffter, T.; D \"o ssel, O.; Plank, G.; et al. 2023. MedalCare-XL: 16,900 healthy and pathological synthetic 12 lead ECGs from electrophysiological simulations. Scientific Data,...

    2023

  8. [16]

    L.; Amaral, L

    Goldberger, A. L.; Amaral, L. A.; Glass, L.; Hausdorff, J. M.; Ivanov, P. C.; Mark, R. G.; Mietus, J. E.; Moody, G. B.; Peng, C.-K.; and Stanley, H. E. 2000. PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals. circul...

    2000

  9. [17]

    Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; and Bengio, Y. 2020. Generative adversarial networks. Communications of the ACM, 63(11): 139--144

    2020

  10. [18]

    Han, X.; Zheng, H.; and Zhou, M. 2022. Card: Classification and regression diffusion models. Advances in Neural Information Processing Systems, 35: 18100--18115

    2022

  11. [19]

    C.; Ver Steeg, G.; and Galstyan, A

    Harutyunyan, H.; Khachatrian, H.; Kale, D. C.; Ver Steeg, G.; and Galstyan, A. 2019. Multitask learning and benchmarking with clinical time series data. Scientific data, 6(1): 96

    2019

  12. [20]

    Hatami, N.; Gavet, Y.; and Debayle, J. 2018. Classification of time-series images using deep convolutional neural networks. In Tenth international conference on machine vision (ICMV 2017), volume 10696, 242--249. SPIE

    2018

  13. [21]

    Horn, M.; Moor, M.; Bock, C.; Rieck, B.; and Borgwardt, K. 2020. Set functions for time series. In International Conference on Machine Learning, 4353--4363. PMLR

    2020

  14. [22]

    Huang, J.; Yang, B.; Yin, K.; and Xu, J. 2024. DNA-T: Deformable Neighborhood Attention Transformer for Irregular Medical Time Series. IEEE Journal of Biomedical and Health Informatics

    2024

  15. [23]

    H.; and Lukasiewicz, T

    Li, B.; Torr, P. H.; and Lukasiewicz, T. 2022. Clustering generative adversarial networks for story visualization. In Proceedings of the 30th ACM International Conference on Multimedia, 769--778

    2022

  16. [24]

    Li, Z.; Li, S.; and Yan, X. 2024. Time series as images: Vision transformer for irregularly sampled time series. Advances in Neural Information Processing Systems, 36

    2024

  17. [25]

    C.; Kale, D

    Lipton, Z. C.; Kale, D. C.; Wetzel, R.; et al. 2016. Modeling missing data in clinical time series with rnns. Machine Learning for Healthcare, 56(56): 253--270

    2016

  18. [26]

    Liu, Z.; Hu, H.; Lin, Y.; Yao, Z.; Xie, Z.; Wei, Y.; Ning, J.; Cao, Y.; Zhang, Z.; Dong, L.; Wei, F.; and Guo, B. 2022. Swin Transformer V2: Scaling Up Capacity and Resolution. In International Conference on Computer Vision and Pattern Recognition (CVPR)

    2022

  19. [27]

    Ma, Q.; Li, S.; and Cottrell, G. W. 2020. Adversarial joint-learning recurrent neural network for incomplete time series classification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(4): 1765--1776

    2020

  20. [28]

    Ma, Q.; Zheng, J.; Li, S.; and Cottrell, G. W. 2019. Learning representations for time series clustering. Advances in neural information processing systems, 32

    2019

  21. [29]

    P.; Sahu, D

    Maroor, J. P.; Sahu, D. N.; Nijhawan, G.; Karthik, A.; Shrivastav, A.; and Chakravarthi, M. K. 2024. Image-Based Time Series Forecasting: A Deep Convolutional Neural Network Approach. In 2024 4th International Conference on Innovative Practices in Technology and Management (IC...

    2024

  22. [30]

    Miao, X.; Wu, Y.; Wang, J.; Gao, Y.; Mao, X.; and Yin, J. 2021. Generative semi-supervised learning for multivariate time series imputation. In Proceedings of the AAAI conference on artificial intelligence, volume 35, 8983--8991

    2021

  23. [31]

    Neil, D.; Pfeiffer, M.; and Liu, S.-C. 2016. Phased lstm: Accelerating recurrent network training for long or event-based sequences. Advances in neural information processing systems, 29

    2016

  24. [32]

    Raghu, A.; Chandak, P.; Alam, R.; Guttag, J.; and Stultz, C. 2023. Sequential multi-dimensional self-supervised learning for clinical time series. In International Conference on Machine Learning, 28531--28548. PMLR

    2023

  25. [33]

    Reiss, A.; and Stricker, D. 2012. Introducing a new benchmarked dataset for activity monitoring. In 2012 16th international symposium on wearable computers, 108--109. IEEE

    2012

  26. [34]

    A.; Josef, C

    Reyna, M. A.; Josef, C. S.; Jeter, R.; Shashikumar, S. P.; Westover, M. B.; Nemati, S.; Clifford, G. D.; and Sharma, A. 2020. Early prediction of sepsis from clinical data: the PhysioNet/Computing in Cardiology Challenge 2019. Critical care medicine, 48(2): 210--217

    2020

  27. [35]

    J.; Wang, Z.; Oikonomou, E

    Sangha, V.; Khunte, A.; Holste, G.; Mortazavi, B. J.; Wang, Z.; Oikonomou, E. K.; and Khera, R. 2024. Biometric contrastive learning for data-efficient deep learning from electrocardiographic images. Journal of the American Medical Informatics Association, 31(4): 855--865

    2024

  28. [36]

    J.; Haimovich, A

    Sangha, V.; Mortazavi, B. J.; Haimovich, A. D.; Ribeiro, A. H.; Brandt, C. A.; Jacoby, D. L.; Schulz, W. L.; Krumholz, H. M.; Ribeiro, A. L. P.; and Khera, R. 2022. Automated multilabel diagnosis on electrocardiographic images and signals. Nature communications, 13(1): 1583

    2022

  29. [37]

    Semenoglou, A.-A.; Spiliotis, E.; and Assimakopoulos, V. 2023. Image-based time series forecasting: A deep convolutional neural network approach. Neural Networks, 157: 39--53

    2023

  30. [38]

    Sood, S.; Zeng, Z.; Cohen, N.; Balch, T.; and Veloso, M. 2021. Visual time series forecasting: an image-driven approach. In Proceedings of the Second ACM International Conference on AI in Finance, 1--9

    2021

  31. [39]

    Tripathy, R.; and Acharya, U. R. 2018. Use of features from RR-time series and EEG signals for automated classification of sleep stages in deep neural network framework. Biocybernetics and Biomedical Engineering, 38(4): 890--902

    2018

  32. [40]

    Wang, J.; Du, W.; Cao, W.; Zhang, K.; Wang, W.; Liang, Y.; and Wen, Q. 2024. Deep learning for multivariate time series imputation: A survey. arXiv preprint arXiv:2402.04059

    2024 arXiv

  33. [41]

    Wang, Z.; and Oates, T. 2015. Imaging time-series to improve classification and imputation. In Proceedings of the 24th International Conference on Artificial Intelligence, IJCAI'15, 3939–3945. AAAI Press. ISBN 9781577357384

    2015

  34. [42]

    B.; Wong, K

    Weerakody, P. B.; Wong, K. W.; and Wang, G. 2023. Policy gradient empowered LSTM with dynamic skips for irregular time series data. Applied Soft Computing, 142: 110314

    2023

  35. [43]

    Wu, H.; Hu, T.; Liu, Y.; Zhou, H.; Wang, J.; and Long, M. 2023. TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis. In International Conference on Learning Representations

    2023

  36. [44]

    Xu, S.; Xu, T.; Yang, Y.; and Chen, X. 2024. Learning metabolic dynamics from irregular observations by Bidirectional Time-Series State Transfer Network. mSystems, e00697--24

    2024

  37. [45]

    Zhang, J.; Zheng, S.; Cao, W.; Bian, J.; and Li, J. 2023. Warpformer: A multi-scale modeling approach for irregular clinical time series. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 3273--3285

    2023

  38. [46]

    Zhang, X.; Zeman, M.; Tsiligkaridis, T.; and Zitnik, M. 2022. Graph-Guided Network For Irregularly Sampled Multivariate Time Series. In International Conference on Learning Representations, ICLR

    2022

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.