Pith. sign in

REVIEW 4 major objections 8 minor 67 references

Paper ECGs to diagnosis in 30 seconds on a laptop

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · glm-5.2

2026-07-09 02:30 UTC pith:CSLQO33Y

load-bearing objection Solid engineering pipeline for paper ECG digitization with CPU-only deployment; headline accuracy conflates synthetic-to-synthetic and real-world evidence the 4 major comments →

arxiv 2607.07683 v1 pith:CSLQO33Y submitted 2026-07-08 cs.LG

ECGLight: Compute-Light Framework For Paper ECG Digitization and Myocardial Infarction Screening

Shreyasvi Natraj , Cyrus Achtari , Felice Gragnano , Andrea Milzi , Marco Valgimigli , Diego Paez-Granados This is my paper
classification cs.LG
keywords ECG digitizationpaper ECGmyocardial infarctionYOLOv11CPU-only inferenceedge AIreference pulse calibrationSHAP interpretability
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper presents a complete software pipeline that takes a smartphone photo or scan of a paper ECG printout and, running entirely on a laptop CPU with no internet connection, converts it into a calibrated 12-lead digital signal and then screens for myocardial infarction. The digitization stage uses a YOLOv11-based instance segmentation model trained on synthetic paper-ECG images rendered from the PTB-XL waveform dataset. The key claim is that this model, having seen only synthetic images, generalizes to real hospital paper ECG scans from a different institution without retraining, producing signals accurate enough to train downstream MI classifiers. The full pipeline runs in 25 to 30 seconds per ECG on CPU-only hardware, achieving 95.51% accuracy for MI detection on PTB-XL and 88.89% accuracy for occlusion MI detection on the real-world ECG-Matrix dataset. The authors also use SHAP attributions on a simple MLP classifier to show that the model's predictions are driven by physiologically expected ECG regions, the QRS complex and ST segment, rather than digitization artifacts.

Core claim

The central mechanism is a three-stage pipeline: (1) patch-based YOLOv11 instance segmentation to isolate inked waveforms from grid lines, shadows, and noise, with multi-scale patch fusion to preserve fine trace boundaries; (2) automated lead-name detection and layout inference across 16 standard ECG formats, including Cabrera ordering, to correctly map each trace to its anatomical lead; and (3) reference-pulse detection, a 1 mV, 200 ms calibration marker printed on most ECGs, to convert pixel coordinates into physically meaningful millivolts and milliseconds without relying on grid visibility. The discovery is that a digitizer trained exclusively on synthetic rendered images can generalize,

What carries the argument

YOLOv11

Load-bearing premise

The central load-bearing assumption is that synthetic paper ECG images rendered from PTB-XL waveforms are realistic enough to train a digitizer that works on real hospital paper ECG scans. For the real-world ECG-Matrix dataset, there is no paired digital ground-truth signal, so digitization quality is only assessed indirectly through downstream classification accuracy, leaving open whether systematic digitization artifacts in real scans are inflating or distorting the MI-d

What would settle it

If a set of real hospital paper ECGs with paired digital ground-truth signals were digitized by this pipeline and showed substantially lower Pearson correlation or higher RMSE than the synthetic test set, the claim of cross-domain generalization would be weakened.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 8 minor

Summary. The manuscript presents ECGLight, an end-to-end pipeline for digitizing paper ECG images and performing downstream myocardial infarction (MI) screening. The digitization pipeline uses YOLOv11x-seg for patched waveform segmentation, a YOLOv11x detector for lead-name and reference-pulse localization, and reference-pulse-based calibration to recover physically scaled 12-lead time-series. The digitizer is trained on synthetic paper ECG images rendered from PTB-XL waveforms using ECG-Image-Kit, and applied without retraining to real hospital paper ECGs from the ECG-Matrix dataset. Downstream classification benchmarks (MI vs. Normal on PTB-XL; Pre- vs. Post-procedural MI and OMI vs. non-OMI on ECG-Matrix) are performed using compact time-series models (e.g., Rocket, Arsenal, InceptionTime) under CPU-only inference, with SHAP-based interpretability analysis via an MLP surrogate. The system runs end-to-end in approximately 25–30 seconds per ECG on a laptop CPU.

Significance. The paper addresses a practically important problem: enabling automated ECG analysis in low-resource settings where paper printouts are still the norm and GPU/cloud infrastructure is unavailable. The end-to-end design connecting digitization, calibration, layout inference, classification, and interpretability within a CPU-only pipeline is a genuine engineering contribution. The provision of publicly available code, pretrained models, and a web dashboard is a notable strength that enhances reproducibility. The SHAP-based physiological attribution analysis, cross-validated by per-lead classification performance, adds interpretive value. The application to the ECG-Matrix dataset for OMI detection, a clinically relevant endpoint, demonstrates the pipeline's potential utility beyond benchmark datasets.

major comments (4)
  1. §V (Results, 'Overall ECG Digitization Quality') and Table I(d): The digitization quality is validated almost entirely on synthetic images (PTB-XL rendered via ECG-Image-Kit), with an 11.94% failure rate, Pearson r=0.806, and SNR=4.54 dB. The headline 95.51% MI classification accuracy on PTB-XL (Table III) is derived from signals that passed through this synthetic-to-synthetic digitization pipeline. The manuscript should explicitly state in the abstract and results that the 95.51% figure reflects classification on synthetic-to-synthetic digitized signals, not real paper ECG digitization. As written, the abstract's juxtaposition of '95.51% accuracy on PTB-XL' and '88.89% on ECG-Matrix' implies comparable evidential weight, which is misleading. This is a load-bearing framing issue for the central claim of real-world generalization.
  2. §IV ('Benchmarking Digitized ECG-Based Time-Series Classification Models') and Table II(b): For ECG-Matrix (the only real-world test), the classifiers (e.g., Rocket for OMI detection) are trained directly on the digitized ECG-Matrix signals with patient-wise splits. Because both train and test signals come from the same digitization distribution, any systematic artifacts introduced by the synthetic-to-real domain gap would be present in both splits. The classifier could learn to exploit or tolerate these artifacts without them affecting accuracy. Thus, the 88.89% OMI accuracy does not, by itself, establish that the digitizer faithfully recovers true waveform morphology from real paper ECGs. The manuscript should acknowledge this limitation more prominently and discuss what specific validation (e.g., expert visual review of digitized ECG-Matrix signals, or signal-level comparison on a小型手动
  3. §IV ('Data Generation') and §V ('YOLO-Based Patched Segmentation'): The synthetic-to-real domain gap is the central load-bearing assumption of the paper. The manuscript states that the digitizer 'generalized to real hospital-style ECG Matrix scans' (§VI), but provides no signal-level ground truth for ECG-Matrix to verify this. While the lack of paired ground truth is understandable, the paper could strengthen its claim by including qualitative examples of digitized real ECG-Matrix signals alongside the original scans, or by reporting any expert review of the digitized outputs. Without any such evidence, the claim of generalization rests entirely on indirect downstream classification accuracy, which is an imperfect proxy as noted above.
  4. Table I(d): The 11.94% failure rate in digitization is reported but not reflected in the downstream classification metrics. It is unclear whether failed digitizations were excluded from the classification datasets (Table IIb), and if so, how failure was determined for real ECG-Matrix images where no ground truth exists. If failed digitizations were included, they may have introduced artifacts that affected classifier training and evaluation. The manuscript should clarify the handling of digitization failures in the downstream pipeline.
minor comments (8)
  1. Abstract: 'acute coronary occlusion (ACS) is overlooked' — ACS encompasses more than occlusion; consider rephrasing to 'acute coronary syndrome (ACS) is overlooked' or 'acute coronary occlusion is overlooked' for precision.
  2. Table I(a): The YOLOv12x (Patched) row reports box precision 0.618 and recall 0.594, which are substantially lower than YOLOv11x. The text in §IV states 'YOLOv11x remained superior in precision/recall,' but the magnitude of YOLOv12x's underperformance is surprising. A brief discussion of why YOLOv12x performed so poorly on this task would be informative.
  3. Table S2: The SNR comparison with challenge submissions is difficult to interpret because the proposed method's SNR (4.54 dB) is reported as a single value, while other methods report clean and deteriorated values. It is unclear whether 4.54 dB corresponds to clean, deteriorated, or an average. Clarification is needed for a fair comparison.
  4. §V ('MLP-SHAP-Based Feature Analysis'): The MLP architecture is described as 'one hidden layer (100 neurons)' in the main text but as 'two fully connected hidden layers (100 and 50 neurons)' in the Supplementary Information (§'MLP-Based SHAP Feature Importance Analysis'). These should be reconciled.
  5. Figure 2: The SHAP attribution maps are described as 'idealized 12-lead heartbeat' with overlaid SHAP traces. It would help to clarify whether the underlying heartbeat is a representative example or a schematic, and whether the SHAP values are from a single test beat or aggregated.
  6. Table II(b): The 'Avg. Timesteps' column for full-sequence settings reports values like 1,633 and 4,046, which appear to be the number of time steps per recording. This should be labeled more clearly (e.g., 'Timesteps per record') to avoid confusion with the segmented setting's 280.
  7. §VII (Conclusion): 'helps bring advanced diagnostic capabilities to settings where they in settings where digital ECG export...' — there is a duplicated phrase ('in settings where they in settings where').
  8. References: Several references (e.g., [33] YOLO, [36] YOLO-Patch-Based-Inference) cite GitHub repositories or software without version numbers or specific commit hashes. For reproducibility, consider adding version or commit information.

Simulated Author's Rebuttal

4 responses · 0 unresolved

We thank the referee for a careful and constructive review. The referee's four major comments all concern the same underlying issue: the evidential asymmetry between our synthetic-to-synthetic PTB-XL results and our synthetic-to-real ECG-Matrix results, and whether the manuscript's framing adequately communicates this asymmetry. We agree with the substance of all four comments and will revise the manuscript accordingly. Specifically, we will: (1) clarify in the abstract and results that the 95.51% PTB-XL accuracy reflects synthetic-to-synthetic digitization, not real paper ECG digitization; (2) add a prominent limitation discussion acknowledging that ECG-Matrix classification accuracy does not, by itself, establish faithful waveform recovery from real paper ECGs, since systematic digitization artifacts would be present in both train and test splits; (3) add qualitative examples of digitized real ECG-Matrix signals alongside their original scans, and report the results of an expert visual review; and (4) clarify how digitization failures were handled in the downstream classification pipeline for both datasets. We do not have standing objections to any of the referee's comments.

read point-by-point responses

  1. Referee: §V and Table I(d): The 95.51% MI classification accuracy on PTB-XL is derived from synthetic-to-synthetic digitization, and the abstract's juxtaposition with 88.89% on ECG-Matrix implies comparable evidential weight, which is misleading.

    Authors: The referee is correct. The 95.51% accuracy on PTB-XL reflects classification on signals that were rendered from digital PTB-XL waveforms into synthetic paper images and then digitized back—the digitizer operates on synthetic images and is evaluated against the original digital ground truth. This is a synthetic-to-synthetic round-trip and does not test real paper ECG digitization. The 88.89% OMI accuracy on ECG-Matrix, by contrast, involves real hospital paper ECG scans, but as the referee notes in Comment 2, it has its own evidential limitations. We agree that the current abstract framing implies comparable evidential weight between these two numbers, which is misleading. We will revise the abstract to explicitly state that the PTB-XL result reflects classification on synthetic-to-synthetic digitized signals, and that the ECG-Matrix result reflects application to real paper ECGs without paired signal ground truth. We will also add a clarifying note in the Results section (§V) at the point where both numbers are first presented together. revision: yes

  2. Referee: §IV and Table II(b): For ECG-Matrix, classifiers are trained on digitized signals from the same digitization distribution in both train and test splits, so systematic artifacts from the synthetic-to-real domain gap could be learned or tolerated without affecting accuracy. The 88.89% OMI accuracy does not establish faithful waveform recovery from real paper ECGs.

    Authors: We fully agree with this analysis. Because both the training and test splits for ECG-Matrix classification pass through the same digitization pipeline, any systematic artifacts introduced by the synthetic-to-real domain gap would be present in both splits. The classifier could learn to exploit or tolerate these artifacts, meaning that downstream classification accuracy is an imperfect proxy for digitization fidelity. This is a genuine limitation of our validation strategy for ECG-Matrix, and it stems from the fundamental absence of paired digital waveform ground truth for real paper ECGs—a limitation we acknowledge in the Discussion (§VI) but do not sufficiently emphasize. We will add a prominent paragraph in the Discussion explicitly stating that ECG-Matrix classification accuracy does not, by itself, establish faithful waveform morphology recovery, and that the domain gap could introduce systematic artifacts that are invisible to the downstream classification metric. We will also discuss what additional validation would be needed (expert visual review, signal-level comparison on a manually digitized subset) and note that we are incorporating the first of these in the revision (see our response to Comment 3). revision: yes

  3. Referee: §IV and §V: The synthetic-to-real domain gap is the central load-bearing assumption. The manuscript claims the digitizer 'generalized to real hospital-style ECG Matrix scans' but provides no signal-level ground truth or qualitative examples for ECG-Matrix.

    Authors: This is a fair criticism. The claim that the digitizer generalized to real ECG-Matrix scans currently rests on indirect evidence (downstream classification accuracy and physiologically plausible SHAP attributions), which, as the referee correctly notes, is an imperfect proxy. We will address this in two ways. First, we will add a new supplementary figure showing qualitative examples of digitized real ECG-Matrix signals displayed alongside the original scanned paper ECG images, so readers can visually assess reconstruction quality. Second, we have conducted an expert visual review of a sample of digitized ECG-Matrix signals by a cardiologist on our team (F.G.), and we will report the results of this review in the revised manuscript, including the fraction of signals rated as clinically interpretable and the types of artifacts observed. We acknowledge that even expert review is not a substitute for signal-level ground truth, and we will state this explicitly. However, we believe these additions substantially strengthen the generalization claim beyond indirect downstream metrics alone. revision: yes

  4. Referee: Table I(d): The 11.94% digitization failure rate is not reflected in downstream classification metrics. It is unclear whether failed digitizations were excluded from classification datasets, and if so, how failure was determined for real ECG-Matrix images where no ground truth exists.

    Authors: The referee raises an important point about the handling of digitization failures in the downstream pipeline. We will clarify this in the revised manuscript. For PTB-XL, the 11.94% failure rate reported in Table I(d) was measured on the 1,600-image digitization evaluation subset (which spans all six diagnostic superclasses), while the classification datasets (Table IIb) were constructed from separate Normal and MI subsets. Digitization failures on the classification subsets were identified by automated quality-control checks: signals with missing leads (fewer than 12 leads successfully extracted), implausible calibration values (reference pulse detection failure or out-of-range scaling factors), or signals with zero or near-zero variance were flagged as failures and excluded from the classification datasets. For ECG-Matrix, the same automated quality-control checks were applied, since no signal-level ground truth exists. We will add a paragraph in §IV (Data Generation / Dataset Construction) explicitly describing these quality-control criteria and reporting the fraction of images excluded from each classification dataset due to digitization failure. We will also add a note in Table IIb indicating the number of records excluded prior to the reported splits. revision: yes

Circularity Check

0 steps flagged

No significant circularity found; the derivation chain is self-contained with one minor dataset self-citation that is not load-bearing.

full rationale

The paper's central claims are not circular by construction. The digitization pipeline is trained on synthetic PTB-XL renderings (via the external ECG-Image-Kit library) and evaluated against held-out PTB-XL ground-truth waveforms using standard metrics (Pearson r, RMSE, SNR). The downstream classification results (95.51% on PTB-XL, 88.89% on ECG-Matrix) use patient-wise splits with labels derived independently of the digitization process (PTB-XL diagnostic superclasses; ECG-Matrix angiographic findings). The PTB-XL classification path (waveform → render → digitize → classify) is not trivially circular because the digitization is lossy (r=0.806, 11.94% failure rate) and the classifier must learn discriminative morphology from degraded signals. The ECG-Matrix path applies a PTB-XL-trained digitizer to real scans and trains classifiers on patient-wise splits—systematic artifacts would affect both train and test, but this is an external-validity concern, not a construction-level circularity. The paper transparently acknowledges these limitations in the Discussion. The only self-citation is reference [27] (Gragnano, Valgimigli—authors of the present paper) for the ECG-Matrix/MATRIX trial dataset, but this is a dataset provenance citation, not a methodological premise that would make any derivation reduce to its inputs. All key tools (YOLO, ECG-Image-Kit, PTB-XL, sktime classifiers) are external and independently verifiable. Score 1 reflects the minor dataset self-citation without load-bearing circularity.

Axiom & Free-Parameter Ledger

7 free parameters · 5 axioms · 0 invented entities

The paper does not invent new physical entities, particles, forces, or mathematical objects. It combines existing neural network architectures (YOLOv11x, MLP, Rocket, Arsenal, InceptionTime), existing datasets (PTB-XL, ECG-Matrix), and existing tools (ECG-Image-Kit, SHAP, Albumentations) into an integrated pipeline. The free parameters are standard machine learning hyperparameters and design choices, not physically motivated constants. The axioms are domain assumptions about ECG image structure and the validity of synthetic-to-real transfer, which are the load-bearing premises of the work.

free parameters (7)
  • YOLOv11x-seg fine-tuning hyperparameters = 100 epochs, AdamW, cosine LR, FP16, batch size 16
    Standard training hyperparameters chosen for the segmentation model; not independently derived.
  • Patch grid sizes = 4×5, 5×6, 6×8
    Multi-scale overlapping patch grids selected empirically to improve local boundary learning; not derived from first principles.
  • MLP architecture = 100 neurons, 1 hidden layer, ReLU, dropout 0.2
    Chosen for SHAP compatibility and simplicity; architecture capacity is a free design choice.
  • SHAP background sample count = 50
    Number of background samples for SHAP baseline estimation; chosen empirically.
  • PCA components for SHAP = 50
    Dimensionality reduction applied before SHAP; the number of components is a free parameter.
  • Heartbeat window length = 150 ms pre-R, 300 ms post-R
    Fixed segmentation window around R-peaks; chosen to capture QRS and ST-T but not a derived quantity.
  • Morphological opening kernel = 1×25
    Kernel size for reference pulse processing; selected empirically for vertical edge preservation.
axioms (5)
  • domain assumption Synthetic paper ECG images generated by ECG-Image-Kit from PTB-XL waveforms are sufficiently realistic to train a digitizer that generalizes to real hospital paper ECG scans.
    This is the core assumption enabling the entire pipeline. Stated in Methods §IV and acknowledged in Discussion: 'Most quantitative digitization validation is performed on synthetic paper ECG images rendered from PTB-XL waveforms.'
  • domain assumption Reference pulses (1 mV, 200 ms) are present and visually detectable on most clinical paper ECGs.
    The calibration method depends on this. Stated in Methods §IV (Reference Pulse Detection): 'These rectangular calibration markers (present on most clinical ECGs) define the vertical and horizontal scales.'
  • domain assumption Standard 12-lead ECG layouts can be classified into 16 discrete types from lead-name spatial arrangement.
    Layout inference relies on this finite taxonomy. Stated in Methods §IV (Layout Identification): 'sixteen layout types (including 12×1, 6×2, 4×3, 3×4, plus single-rhythm-strip variants) were defined from common PTB-XL configurations.'
  • ad hoc to paper Downstream classification accuracy on ECG-Matrix is a valid proxy for digitization fidelity when signal-level ground truth is unavailable.
    The paper uses this assumption to validate digitization on real hospital scans. Stated in Discussion: 'for real-world scans we evaluate digitization indirectly through downstream diagnostic performance and interpretability rather than signal-level reconstruction error.'
  • domain assumption YOLOv11x-seg pretrained weights provide a suitable initialization for ECG waveform instance segmentation.
    The segmentation model is fine-tuned from pretrained YOLOv11x weights. This is standard practice but assumes the pretrained features transfer to the ECG domain.

pith-pipeline@v1.1.0-glm · 38938 in / 3923 out tokens · 685237 ms · 2026-07-09T02:30:46.161948+00:00 · methodology

0 comments
read the original abstract

Electrocardiography (ECG) is one of the most widely used tests for diagnosing cardiovascular disease. Yet several remote clinics still utilize paper ECG printouts for their analysis due to limited connectivity and computational capacity. As a result, vast numbers of physical ECGs obtained in remote areas still remain incapable of being accessed by contemporary artificial-intelligence (AI)-based decision support as they require high computational resources or strong high-speed internet connectivity. This causes several cases where conditions like acute coronary occlusion (ACS) is overlooked and reperfusion therapy delayed. Although prior work has tackled digitization and diagnosis separately, and utilized advanced AI models for them, there still remains a lack of a compute-light, on-device framework that reconstructs paper ECGs at high fidelity, while accurately supporting multiple clinically relevant endpoints. We address this need with an end-to-end lightweight on-device digitization-to-diagnosis pipeline that converts a smartphone photo or scan of a paper ECG into a calibrated 12-lead signal and screens for Myocardial Infarction (MI) pathologies, with SHapley Additive exPlanations (SHAP) to support interpretability. Trained and evaluated on 21,799 ECGs from the PTB-XL dataset and further validated on hospital-acquired ECG-Matrix dataset, the complete system runs in <30 s per ECG on CPU-only resources, achieving 95.51% accuracy (F1 = 0.9519) for MI detection on PTB-XL and 88.89% accuracy (F1 = 0.8862) for OMI detection on ECG-Matrix. This work showcases that legacy paper records can be reliably democratized in any part of the world, providing a scalable decision support when digital ECG export, connectivity, or high-end compute are unavailable

Figures

Figures reproduced from arXiv: 2607.07683 by Andrea Milzi, Cyrus Achtari, Diego Paez-Granados, Felice Gragnano, Marco Valgimigli, Shreyasvi Natraj.

Figure 1
Figure 1. Figure 1: End-to-End ECG Workflow: A photographed or scanned paper ECG is preprocessed (shadow/background suppression and denoising) and then analyzed with a YOLOv11-based module to segment the inked waveforms, detect lead labels, identify reference pulses, and infer the page layout. Reference-pulse–based calibration maps pixels to seconds and millivolts, enabling reconstruction and temporal alignment of the full 12… view at source ↗
Figure 2
Figure 2. Figure 2: Idealized 12-lead SHAP attribution maps for myocardial infarction classification under three clinical label definitions. Each panel summarizes post-hoc feature attribution from an MLP classifier using SHapley Additive exPlanations (SHAP) on digitized 12-lead ECG time-series (standardized per lead and evaluated on held-out test samples). SHAP values quantify the marginal contribution of each lead’s waveform… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

67 extracted references · 67 canonical work pages · 9 internal anchors

  1. [1]

    Pathophysiology, diagnosis and manage- ment of myocardial infarction: A review,

    A. Mitsis and F. Gragnano, “Pathophysiology, diagnosis and manage- ment of myocardial infarction: A review,”Current Cardiology Reviews, vol. 17, no. 4, p. e230421189013, 2021

    work page 2021

  2. [2]

    Diagnosis and treatment of acute coronary syndromes: A review,

    D. L. Bhatt, R. D. Lopes, and R. A. Harrington, “Diagnosis and treatment of acute coronary syndromes: A review,”JAMA, vol. 327, no. 7, pp. 662–675, 2022

    work page 2022

  3. [3]

    The risk factors and prevention of cardio- vascular disease: the importance of electrocardiogram in the diagnosis and treatment of acute coronary syndrome,

    A. Rosiek and K. Leksowski, “The risk factors and prevention of cardio- vascular disease: the importance of electrocardiogram in the diagnosis and treatment of acute coronary syndrome,”Therapeutics and Clinical Risk Management, vol. 12, pp. 1223–1229, 2016

    work page 2016

  4. [4]

    Electrocardiography,

    D. M. Mirvis and A. L. Goldberger, “Electrocardiography,”Heart disease, vol. 1, pp. 82–128, 2001, publisher: WB Saunders, Philadelphia

    work page 2001

  5. [5]

    A. L. Goldberger, Z. D. Goldberger, and A. Shvilkin,Clinical Electro- cardiography: A Simplified Approach, 9th ed. Elsevier, 2017

    work page 2017

  6. [6]

    Verification of myocardial infarction with coronary occlusion and nonocclusion by angiography using elec- trocardiographic findings,

    H. P. Meyers, A. T. Limkakeng, E. J. Jaffa, A. Patel, B. J. Theiling, S. R. Rezaie, and S. W. Smith, “Verification of myocardial infarction with coronary occlusion and nonocclusion by angiography using elec- trocardiographic findings,”International Journal of Cardiology: Heart & Vasculature, vol. 33, p. 100767, 2021

    work page 2021

  7. [7]

    Resting 12-lead ECG tests performed by patients at home amid the COVID-19 pandemic — results from the first 1000 patients,

    R.-K. Chang, “Resting 12-lead ECG tests performed by patients at home amid the COVID-19 pandemic — results from the first 1000 patients,”Journal of Electrocardiology, vol. 73, pp. 108–112, 2022-07, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0022073622000863

    work page 2022

  8. [8]

    PDF–ECG in clinical practice: A model for long–term preservation of digital 12–lead ECG data,

    R. Sassi, R. R. Bond, A. Cairns, D. D. Finlay, D. Guldenring, G. Libretti, L. Isola, M. Vaglio, R. Poeta, M. Campana, C. Cuccia, and F. Badilini, “PDF–ECG in clinical practice: A model for long–term preservation of digital 12–lead ECG data,”Journal of Electrocardiology, vol. 50, no. 6, pp. 776–780, 2017-11, publisher: Elsevier BV. [Online]. Available: htt...

    work page 2017

  9. [9]

    Why complicate an important task? an orderly display of the limb leads in the 12-lead electrocardiogram and its implications for recognition of acute coronary syndrome,

    T. Lindow, Y . Birnbaum, K. Nikus, A. Maan, U. Ekelund, and O. Pahlm, “Why complicate an important task? an orderly display of the limb leads in the 12-lead electrocardiogram and its implications for recognition of acute coronary syndrome,”BMC Cardiovascular Disorders, vol. 19, no. 1, p. 13, 2019-01-10. [Online]. Available: https://doi.org/10.1186/s12872-...

    work page doi:10.1186/s12872-018-0979-x 2019

  10. [10]

    Development and validation of an algorithm for the digitization of ECG paper images,

    V . Randazzo, E. Puleo, A. Paviglianiti, A. Vallan, and E. Pasero, “Development and validation of an algorithm for the digitization of ECG paper images,”Sensors, vol. 22, no. 19, p. 7138, 2022-01, number: 19 Publisher: Multidisciplinary Digital Publishing Institute. [Online]. Available: https://www.mdpi.com/1424-8220/22/19/7138

    work page 2022

  11. [11]

    Automated analysis of ecgs for cardiovascu- lar disease detection,

    R. Jones, S. Patelet al., “Automated analysis of ecgs for cardiovascu- lar disease detection,”IEEE Transactions on Biomedical Engineering, vol. 68, no. 4, pp. 987–995, 2021

    work page 2021

  12. [12]

    Automatic digitization of paper electrocardiograms – a systematic review,

    A. Lence, F. Extramiana, A. Fall, J.-E. Salem, J.-D. Zucker, and E. Prifti, “Automatic digitization of paper electrocardiograms – a systematic review,”Journal of Electrocardiology, vol. 80, pp. 125–132, 2023-09, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0022073623001504

    work page 2023

  13. [13]

    Artificial intelligence for healthcare in africa,

    A. Owoyemi, J. Owoyemi, A. Osiyemi, and A. Boyd, “Artificial intelligence for healthcare in africa,”Frontiers in Digital Health, vol. 2, 2020-07-07, publisher: Frontiers. [Online]. Available: https://www.frontiersin.org/journals/digital-health/articles/10. 3389/fdgth.2020.00006/full

    work page arXiv 2020

  14. [14]

    ECGScan: a method for conversion of paper electrocardiographic printouts to digital electrocardiographic files,

    F. Badilini, T. Erdem, W. Zareba, and A. J. Moss, “ECGScan: a method for conversion of paper electrocardiographic printouts to digital electrocardiographic files,”Journal of Electrocardiology, vol. 38, no. 4, pp. 310–318, 2005-10, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0022073605000774

    work page 2005

  15. [15]

    ECGMiner: A flexible software for accurately digitizing ECG,

    A. F. Santamónica, R. Carratalá-Sáez, Y . Larriba, A. Pérez-Castellanos, and C. Rueda, “ECGMiner: A flexible software for accurately digitizing ECG,”Computer Methods and Programs in Biomedicine, vol. 246, p. 108053, 2024-04

    work page 2024

  16. [16]

    Digitizing ECG image: A new method and open-source software code,

    J. D. Fortune, N. E. Coppa, K. T. Haq, H. Patel, and L. G. Tereshchenko, “Digitizing ECG image: A new method and open-source software code,”Computer Methods and Programs in Biomedicine, vol. 221, p. 106890, 2022-06, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S0169260722002723

    work page 2022

  17. [17]

    ECG paper record digitization and diagnosis using deep learning,

    S. Mishra, G. Khatwani, R. Patil, D. Sapariya, V . Shah, D. Parmar, S. Dinesh, P. Daphal, and N. Mehendale, “ECG paper record digitization and diagnosis using deep learning,”Journal of Medical and Biological Engineering, vol. 41, no. 4, pp. 422–432, 2021-08-01. [Online]. Available: https://doi.org/10.1007/s40846-021-00632-0

    work page doi:10.1007/s40846-021-00632-0 2021

  18. [18]

    Deep learning for digitizing highly noisy paper-based ECG records,

    Y . Li, Q. Qu, M. Wang, L. Yu, J. Wang, L. Shen, and K. He, “Deep learning for digitizing highly noisy paper-based ECG records,”Computers in Biology and Medicine, vol. 127, p. 104077, 2020-12, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier.com/retrieve/pii/S001048252030408X

    work page 2020

  19. [19]

    High precision ECG digitization using artificial intelligence,

    A. Demolder, V . Kresnakova, M. Hojcka, V . Boza, A. Iring, A. Rafajdus, S. Rovder, T. Palus, M. Herman, F. Bauer, V . Jurasek, R. Hatala, J. Bartunek, B. Vavrik, and R. Herman, “High precision ECG digitization using artificial intelligence,”Journal of Electrocardiology, vol. 90, p. 153900, 2025-05, publisher: Elsevier BV. [Online]. Available: https://lin...

    work page 2025

  20. [20]

    Combining optical character recognition with paper ECG digitization,

    S. Ganesh, P. T. Bhatti, M. Alkhalaf, S. Gupta, A. J. Shah, and S. Tridandapani, “Combining optical character recognition with paper ECG digitization,”IEEE Journal of Translational Engineering in Health and Medicine, vol. 9, pp. 1–9, 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9440410

    work page arXiv 2021

  21. [21]

    A fully-automated paper ECG digitisation algorithm using deep learning,

    H. Wu, K. H. K. Patel, X. Li, B. Zhang, C. Galazis, N. Bajaj, A. Sau, X. Shi, L. Sun, Y . Tao, H. Al-Qaysi, L. Tarusan, N. Yasmin, N. Grewal, G. Kapoor, J. W. Waks, D. B. Kramer, N. S. Peters, and F. S. Ng, “A fully-automated paper ECG digitisation algorithm using deep learning,”Scientific Reports, vol. 12, no. 1, p. 20963, 2022-12-05, publisher: Nature P...

    work page 2022

  22. [22]

    Paper-recorded ECG digitization method with automatic reference voltage selection for telemonitoring and diagnosis,

    L.-H. Wang, C.-X. Xie, T. Yang, H.-X. Tan, M.-H. Fan, I.- C. Kuo, Z.-J. Lee, T.-Y . Chen, P.-C. Huang, S.-L. Chen, and P. A. R. Abu, “Paper-recorded ECG digitization method with automatic reference voltage selection for telemonitoring and diagnosis,” Diagnostics, vol. 14, no. 17, p. 1910, 2024-01, number: 17 Publisher: Multidisciplinary Digital Publishing...

    work page 1910

  23. [23]

    ECGtizer: a fully automated digitizing and signal recovery pipeline for electrocardiograms

    A. Lence, A. Fall, S. D. Cohen, F. Granese, J.-D. Zucker, J.-E. Salem, and E. Prifti, “ECGtizer: a fully automated digitizing and signal recovery pipeline for electrocardiograms,” 2024-12-09. [Online]. Available: http://arxiv.org/abs/2412.12139

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  24. [24]

    Ptb-xl, a large publicly available electrocardiography dataset,

    P. Wagner, N. Strodthoff, R.-D. Bousseljot, D. Kreiseler, F. I. Lunze, W. Samek, and T. Schaeffter, “Ptb-xl, a large publicly available electrocardiography dataset,”Scientific Data, vol. 7, no. 1, p. 154, May

    work page

  25. [25]

    Available: https://doi.org/10.1038/s41597-020-0495-6

    [Online]. Available: https://doi.org/10.1038/s41597-020-0495-6

    work page doi:10.1038/s41597-020-0495-6

  26. [26]

    Deep learning for ECG analysis: benchmark experiments on the PTB-XL dataset,

    N. Strodthoff, P. Wagner, T. Schaeffter, and W. Samek, “Deep learning for ECG analysis: benchmark experiments on the PTB-XL dataset,” Physiological Measurement, 2021

    work page 2021

  27. [27]

    Automated detection of myocardial infarction using ECG signals: a review and recent advances,

    U. R. Acharya, S. L. Oh, Y . Hagiwara, J. H. Tan, M. Adam, A. Ger- tych, and R. San Tan, “Automated detection of myocardial infarction using ECG signals: a review and recent advances,”Biomedical Signal Processing and Control, 2017

    work page 2017

  28. [28]

    ECG analysis in patients with acute coronary syndrome undergoing invasive management: rationale and design of the electrocardiography sub-study of the MATRIX trial,

    F. Gragnano, V . Spedicato, E. Frigoli, G. Gargiulo, D. Di Maio, F. Fimiani, V . Fioretti, C. Annoiato, M. Cimmino, F. Esposito, S. Chianese, M. Scalise, L. Fimiani, M. Franzese, E. Monda, A. Schiavo, A. Cesaro, A. De Michele, R. Scalise, A. Caracciolo, G. Andò, E. Stabile, S. Windecker, P. Calabrò, and M. Valgimigli, “ECG analysis in patients with acute ...

    work page 2019

  29. [29]

    Digitization and 14 classification of ECG images: The george b. moody PhysioNet challenge 2024,

    M. A. Reyna, D. Deepanshi, J. Tyler Weigle, Z. Koscova, K. S. Campbell, S. Seyedi, A. Elola, A. Bahrami Rad, A. Jasvant Shah, N. Kumar Bhatia, G. Clifford, and R. Sameni, “Digitization and 14 classification of ECG images: The george b. moody PhysioNet challenge 2024,” inComputing in Cardiology Conference (CinC), vol. 51. Computing in Cardiology, 2024-12-0...

    work page 2024

  30. [30]

    Ecg-image-kit: a synthetic image generation toolbox to facilitate deep learning-based electrocardiogram digitization,

    K. Kodthalu Shivashankara, Deepanshi, A. M. Shervedani, M. A. Reyna, G. D. Clifford, and R. Sameni, “Ecg-image-kit: a synthetic image generation toolbox to facilitate deep learning-based electrocardiogram digitization,”Physiological Measurement, 2024

    work page 2024

  31. [31]

    Combining Hough Transform and Deep Learning Approaches to Reconstruct ECG Signals From Printouts

    F. Krones, B. Walker, T. Lyons, and A. Mahdi, “Combining hough transform and deep learning approaches to reconstruct ECG signals from printouts,” 2024-10-18. [Online]. Available: http://arxiv.org/abs/2410.14185

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  32. [32]

    MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

    A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, “MobileNets: Efficient convolutional neural networks for mobile vision applications,” inarXiv preprint arXiv:1704.04861, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  33. [33]

    MLPerf Tiny Benchmark

    C. Banbury, V . J. Reddi, M. Lam, W. Fu, A. Fazel, J. Hollemanet al., “Benchmarking tiny ML systems: challenges and directions,”arXiv preprint arXiv:2106.07597, 2021

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  34. [34]

    Ultralytics YOLO,

    G. Jocher, J. Qiu, and A. Chaurasia, “Ultralytics YOLO,” 2023- 01, original-date: 2022-09-11T16:39:45Z. [Online]. Available: https: //github.com/ultralytics/ultralytics

    work page 2023

  35. [35]

    L. S. e. a. Valgimigli M, Frigoli E, “Radial versus femoral access and bivalirudin versus unfractionated heparin in invasively managed patients with acute coronary syndrome (matrix): final 1-year results of a multicentre, randomised controlled trial,”The Lancet, vol. 392, no. 10150, pp. 835–848, 2018. [Online]. Available: https://www.thelancet. com/journa...

    work page 2018

  36. [36]

    ECG-Image-Kit: A Toolkit for Synthesis, Analy- sis, and Digitization of Electrocardiogram Images,

    Alphanumerics Lab, “ECG-Image-Kit: A Toolkit for Synthesis, Analy- sis, and Digitization of Electrocardiogram Images,” https://github.com/ alphanumericslab/ecg-image-kit, 2024, accessed: 2025-06-16

    work page 2024

  37. [37]

    Koldim2001/YOLO-patch-based-inference,

    D. Kolesnikov, “Koldim2001/YOLO-patch-based-inference,” 2025-07- 18, original-date: 2024-03-04T18:33:55Z. [Online]. Available: https: //github.com/Koldim2001/YOLO-Patch-Based-Inference

    work page 2025

  38. [38]

    Decoupled Weight Decay Regularization

    I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” International Conference on Learning Representations (ICLR), 2019. [Online]. Available: https://arxiv.org/abs/1711.05101

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  39. [39]

    SGDR: Stochastic Gradient Descent with Warm Restarts

    ——, “Sgdr: Stochastic gradient descent with warm restarts,” International Conference on Learning Representations (ICLR), 2017. [Online]. Available: https://arxiv.org/abs/1608.03983

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  40. [40]

    Mixed Precision Training

    P. Micikevicius, S. Narang, J. Alben, G. Diamos, E. Elsen, D. Garcia, B. Ginsburg, M. Houston, O. Kuchaiev, G. Venkatesh, and H. Wu, “Mixed precision training,” inInternational Conference on Learning Representations (ICLR), 2018. [Online]. Available: https://arxiv.org/abs/1710.03740

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  41. [41]

    R. C. Gonzalez,Digital image processing. Pearson education india, 2009

    work page 2009

  42. [42]

    Albumentations: fast and flexible image augmentations

    A. Buslaev, A. Parinov, E. Khvedchenya, V . I. Iglovikov, and A. A. Kalinin, “Albumentations: fast and flexible image augmentations,” Information, vol. 11, no. 2, p. 125, 2020-02-24. [Online]. Available: http://arxiv.org/abs/1809.06839

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  43. [43]

    Signals and signal processing for the electrophysiologist,

    K. Venkatachalam, J. E. Herbrandson, and S. J. Asirvatham, “Signals and signal processing for the electrophysiologist,”Circulation: Arrhythmia and Electrophysiology, vol. 4, no. 6, pp. 974–981, 2011-12, publisher: American Heart Association. [Online]. Available: https://www.ahajournals.org/doi/10.1161/CIRCEP.111.964973

    work page doi:10.1161/circep.111.964973 2011

  44. [44]

    A threshold selection method from gray-level histograms,

    N. Otsu, “A threshold selection method from gray-level histograms,” IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, no. 1, pp. 62–66, 1979-01. [Online]. Available: https://ieeexplore.ieee.org/ document/4310076

    work page arXiv 1979

  45. [45]

    A probabilistic hough transform,

    N. Kiryati, Y . Eldar, and A. Bruckstein, “A probabilistic hough transform,”Pattern Recognition, vol. 24, no. 4, pp. 303–316, 1991-01, publisher: Elsevier BV. [Online]. Available: https://linkinghub.elsevier. com/retrieve/pii/003132039190073E

    work page arXiv 1991

  46. [46]

    LSD: A fast line segment detector with a false detection control,

    R. Grompone von Gioi, J. Jakubowicz, J.-M. Morel, and G. Randall, “LSD: A fast line segment detector with a false detection control,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 4, pp. 722–732, 2010-04. [Online]. Available: https: //ieeexplore.ieee.org/document/4731268

    work page arXiv 2010

  47. [47]

    An efficient algorithm for automatic peak detection in noisy periodic and non-periodic signals,

    F. Scholkmann, J. Boss, and M. Wolf, “An efficient algorithm for automatic peak detection in noisy periodic and non-periodic signals,” Algorithms, vol. 5, no. 4, pp. 588–603, 2012

    work page 2012

  48. [48]

    Benesty, J

    J. Benesty, J. Chen, Y . Huang, and I. Cohen,Pearson Correlation Coefficient. Springer, 2009

    work page 2009

  49. [49]

    S. W. Smith,The Scientist and Engineer’s Guide to Digital Signal Processing. San Diego, CA: California Technical Publishing, 1997

    work page 1997

  50. [50]

    R-peak detection from ecg signals using fractal based mathematical morpholog- ical operators,

    D. Nankani, B. Parabattina, R. D. Baruah, and P. K. Das, “R-peak detection from ecg signals using fractal based mathematical morpholog- ical operators,” inTENCON 2021 - 2021 IEEE Region 10 Conference (TENCON), 2021, pp. 225–230

    work page 2021

  51. [51]

    Learning repre- sentations by back-propagating errors,

    D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning repre- sentations by back-propagating errors,”Nature, vol. 323, pp. 533–536, 1986

    work page 1986

  52. [52]

    Gradient-based learning applied to document recognition,

    Y . LeCun, L. Bottou, Y . Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,”Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998

    work page 1998

  53. [53]

    Learning phrase representations using RNN encoder–decoder for statistical machine translation,

    K. Cho, B. van Merriënboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y . Bengio, “Learning phrase representations using RNN encoder–decoder for statistical machine translation,” inProceed- ings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014, pp. 1724–1734

    work page 2014

  54. [54]

    Multi-scale convolutional neural networks for time series classification,

    Z. Cui, W. Chen, and Y . Chen, “Multi-scale convolutional neural networks for time series classification,” 2016

    work page 2016

  55. [55]

    Lstm fully convolu- tional networks for time series classification,

    F. Karim, S. Majumdar, H. Darabi, and S. Chen, “Lstm fully convolu- tional networks for time series classification,”IEEE Access, vol. 6, pp. 1662–1669, 2018

    work page 2018

  56. [56]

    Time series classification from scratch with deep neural networks: A strong baseline,

    Z. Wang, W. Yan, and T. Oates, “Time series classification from scratch with deep neural networks: A strong baseline,” in2017 International Joint Conference on Neural Networks (IJCNN), 2017, pp. 1578–1585

    work page 2017

  57. [57]

    Inceptiontime: Finding alexnet for time series classification,

    H. I. Fawaz, B. Lucas, G. Forestier, C. Pelletier, D. F. Schmidt, J. Weber, G. I. Webb, L. Idoumghar, P.-A. Muller, and F. Petitjean, “Inceptiontime: Finding alexnet for time series classification,”Data Mining and Knowledge Discovery, vol. 34, no. 6, pp. 1936–1962, 2020

    work page 1936

  58. [58]

    Rocket: Exceptionally fast and accurate time series classification using random convolutional kernels,

    A. Dempster, F. Petitjean, and G. I. Webb, “Rocket: Exceptionally fast and accurate time series classification using random convolutional kernels,”Data Mining and Knowledge Discovery, vol. 34, pp. 1454– 1495, 2020

    work page 2020

  59. [59]

    The great time series classification bake off: A review and experimental evaluation of recent algorithmic advances,

    M. Middlehurst, J. Large, and A. Bagnall, “The great time series classification bake off: A review and experimental evaluation of recent algorithmic advances,”Data Mining and Knowledge Discovery, vol. 35, no. 2, pp. 401–449, 2021

    work page 2021

  60. [60]

    Textboxes++: A single-shot oriented scene text detector,

    M. Liao, B. Shiet al., “Textboxes++: A single-shot oriented scene text detector,”IEEE Transactions on Image Processing, vol. 27, no. 8, pp. 3676–3690, 2018

    work page 2018

  61. [61]

    Automated ecg signal extraction using calibration pulse detection,

    A. H. Khanet al., “Automated ecg signal extraction using calibration pulse detection,”Journal of Medical Systems, vol. 44, no. 9, p. 162, 2020

    work page 2020

  62. [62]

    Fourth universal definition of myocardial infarction (2018),

    K. Thygesen, J. S. Alpert, A. S. Jaffe, B. R. Chaitman, J. J. Bax, D. A. Morrow, and H. D. White, “Fourth universal definition of myocardial infarction (2018),”Journal of the American College of Cardiology, vol. 72, no. 18, pp. 2231–2264, 2018

    work page 2018

  63. [63]

    sktime: A Unified Interface for Machine Learning with Time Series

    M. Löning, A. Bagnall, S. Ganesh, V . Kazakov, J. Lines, and F. J. Király, “sktime: A unified interface for machine learning with time series,” 2019. [Online]. Available: https://arxiv.org/abs/1909.07872

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  64. [64]

    Ecg image digitization and reconstruction using deep learning for medical applications,

    R. Tomaszewski, E. Ko´ zluk, R. Piotrowicz, K. Paj ˛ ak, M. Wójcik, J. Ko´ zluk, D. Długosz, and P. Szyma ´nski, “Ecg image digitization and reconstruction using deep learning for medical applications,” Computers in Biology and Medicine, vol. 127, p. 104057, 2020. [Online]. Available: https://doi.org/10.1016/j.compbiomed.2020.104057

    work page doi:10.1016/j.compbiomed.2020.104057 2020

  65. [65]

    Deep learning–based ecg image digitization for clinical and research use,

    T. D. Pham, V . Tran, and C.-T. Lin, “Deep learning–based ecg image digitization for clinical and research use,”Artificial Intelligence in Medicine, vol. 127, p. 102290, 2022. [Online]. Available: https://doi.org/10.1016/j.artmed.2022.102290

    work page doi:10.1016/j.artmed.2022.102290 2022

  66. [66]

    W A VIE: A modular and open-source python implementation for fully automated digitisation of paper electrocardiograms,

    M. A. Verlyck, J. R. Dillon, S. A. Creamer, and D. Zhao, “W A VIE: A modular and open-source python implementation for fully automated digitisation of paper electrocardiograms,” inComputing in Cardiology Conference (CinC), vol. 51. Computing in Cardiology, 2024-12-01, ISSN: 2325-887X. [Online]. Available: https://www.cinc.org/archives/ 2024/pdf/CinC2024-229.pdf

    work page 2024

  67. [67]

    Segmentation-based extraction of key components from ECG images: A framework for precise classification and digitization,

    H.-C. Yoon, D.-K. Kim, H.-S. Kim, W.-Y . Seo, C.-H. Heo, and S.-H. Kim, “Segmentation-based extraction of key components from ECG images: A framework for precise classification and digitization,” in Computing in Cardiology Conference (CinC), vol. 51. Computing in Cardiology, 2024-12-01, ISSN: 2325-887X. [Online]. Available: https://www.cinc.org/archives/2...

    work page arXiv 2024