ecg heartbeat categorization dataset Search Results


kaggle electrocardiography dataset  (Kaggle Inc)
86
Kaggle Inc kaggle electrocardiography dataset
Architecture of the SimCardioNet model for <t>ECG</t> classification, integrating pre-processing, CNN-based feature extraction, contrastive learning through SimCLR, and classification using ResNet blocks.
Kaggle Electrocardiography Dataset, supplied by Kaggle Inc, used in various techniques. Bioz Stars score: 86/100, based on 1 PubMed citations. ZERO BIAS - scores, article reviews, protocol conditions and more
https://www.bioz.com/product/ecg+heartbeat+categorization+dataset/pmc12936081-82-21-21?v=Kaggle+Inc
Average 86 stars, based on 1 article reviews
kaggle electrocardiography dataset - by Bioz Stars, 2026-06
86/100 stars
  Buy from Supplier

Image Search Results


Architecture of the SimCardioNet model for ECG classification, integrating pre-processing, CNN-based feature extraction, contrastive learning through SimCLR, and classification using ResNet blocks.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Architecture of the SimCardioNet model for ECG classification, integrating pre-processing, CNN-based feature extraction, contrastive learning through SimCLR, and classification using ResNet blocks.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Extraction

PCA visualization of the 128-dimensional projection head features, showing the separation of ECG classes along the first two principal components.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: PCA visualization of the 128-dimensional projection head features, showing the separation of ECG classes along the first two principal components.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

PCA visualization of self-supervised ECG feature representations from the Dataset III, illustrating class-wise distributions and overlap among normal and pathological conditions.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: PCA visualization of self-supervised ECG feature representations from the Dataset III, illustrating class-wise distributions and overlap among normal and pathological conditions.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

t-SNE visualization of the 128-dimensional projection head features, illustrating the clustering and separation of ECG classes in a 2D space.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: t-SNE visualization of the 128-dimensional projection head features, illustrating the clustering and separation of ECG classes in a 2D space.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

t-SNE projection of self-supervised ECG feature embeddings from the Dataset III, illustrating local clustering behavior and class-wise distribution of normal and pathological recordings.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: t-SNE projection of self-supervised ECG feature embeddings from the Dataset III, illustrating local clustering behavior and class-wise distribution of normal and pathological recordings.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

Conv1 activation maps for sample (Class: 0), showing the extraction of low-level features such as rhythm and shape from the ECG signal, with activation focused on specific regions of the waveform of Dataset I.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv1 activation maps for sample (Class: 0), showing the extraction of low-level features such as rhythm and shape from the ECG signal, with activation focused on specific regions of the waveform of Dataset I.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay, Extraction

Conv1 activation maps for sample (Class: 0), showing the extraction of low-level features such as rhythm and shape from the ECG signal, with activation focused on specific regions of the waveform of Dataset II.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv1 activation maps for sample (Class: 0), showing the extraction of low-level features such as rhythm and shape from the ECG signal, with activation focused on specific regions of the waveform of Dataset II.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay, Extraction

Conv2 activation maps for sample (Class: 0), highlighting intermediate-level features such as the P-QRS-T wave patterns and focusing on the shape of the ECG waveform Dataset I.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv2 activation maps for sample (Class: 0), highlighting intermediate-level features such as the P-QRS-T wave patterns and focusing on the shape of the ECG waveform Dataset I.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay

Conv2 activation maps for sample (Class: 0), highlighting intermediate-level features such as the P-QRS-T wave patterns and focusing on the shape of the ECG waveform Dataset II.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv2 activation maps for sample (Class: 0), highlighting intermediate-level features such as the P-QRS-T wave patterns and focusing on the shape of the ECG waveform Dataset II.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay

Conv3 activation maps for sample (Class: 0), highlighting deep features and advanced patterns in the ECG signal, such as arrhythmias and myocardial infarctions Dataset I.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv3 activation maps for sample (Class: 0), highlighting deep features and advanced patterns in the ECG signal, such as arrhythmias and myocardial infarctions Dataset I.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay

Conv3 activation maps for sample (Class: 0), highlighting deep features and advanced patterns in the ECG signal, such as arrhythmias and myocardial infarctions of Dataset II.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Conv3 activation maps for sample (Class: 0), highlighting deep features and advanced patterns in the ECG signal, such as arrhythmias and myocardial infarctions of Dataset II.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay

Visualization of conv1 activation maps for a normal ECG recording, showing channel-wise responses of the first convolutional layer to different temporal and morphological signal patterns.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Visualization of conv1 activation maps for a normal ECG recording, showing channel-wise responses of the first convolutional layer to different temporal and morphological signal patterns.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Activation Assay

Mean confusion matrix for cross-validation (CV) in multi-class ECG classification of Dataset I.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Mean confusion matrix for cross-validation (CV) in multi-class ECG classification of Dataset I.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Biomarker Discovery

Mean confusion matrix for cross-validation (CV) in multi-class ECG classification of Dataset II.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Mean confusion matrix for cross-validation (CV) in multi-class ECG classification of Dataset II.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Biomarker Discovery

Mean confusion matrix for cross-validation in multi-class ECG classification of Dataset III.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Mean confusion matrix for cross-validation in multi-class ECG classification of Dataset III.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: Biomarker Discovery

ECG waveform samples with true and predicted labels.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: ECG waveform samples with true and predicted labels.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset I.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset I.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset II.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset II.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques:

Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset III.

Journal: Scientific Reports

Article Title: A hybrid learning framework for automated multiclass electrocardiogram classification with SimCardioNet

doi: 10.1038/s41598-026-36932-1

Figure Lengend Snippet: Grad-CAM heatmaps highlighting regions driving the model’s predictions. Warm colors indicate higher contribution; overlays show attention focused on ECG waveforms rather than page artifacts od Dataset III.

Article Snippet: We evaluate SimCardioNet across three distinct ECG image datasets: (1) a 4-class Pakistani clinical ECG dataset (Dataset I), (2) an external Kaggle electrocardiography dataset for out-of-distribution validation (Dataset II), and (3) the large-scale PTB-XL benchmark (Dataset III) covering five diagnostic superclasses.

Techniques: