4-Class Arrhythmia Classification (ARR-4-EFF-SM)
Overview
Section titled “Overview”The following table provides the latest pre-trained model for 4-class arrhythmia classification. Below we also provide additional details including training configuration, performance metrics, and downloads.
| NAME | DATASET | FS | DURATION | # CLASSES | MODEL | PARAMS | FLOPS | METRIC |
|---|---|---|---|---|---|---|---|---|
| ARR-2-EFF-SM | Icentia11K, PTB-XL, LSAD | 100Hz | 5s | 2 | EfficientNetV2 | 18K | 1.2M | 99.5% F1 |
| ARR-4-EFF-SM | LSAD | 100Hz | 5s | 4 | EfficientNetV2 | 27K | 1.6M | 95.9% F1 |
The model is trained on 5-second, raw ECG frames sampled at 100 Hz.
- Sensor: ECG
- Location: Wrist
- Sampling Rate: 100 Hz
- Frame Size: 5 seconds
Class Mapping
Section titled “Class Mapping”The model is trained on raw ECG data and is able to discern normal sinus rhythm (SR) from sinus bradycardia (SBRAD), atrial fibrillation (AFIB), atrial flutter (AFL), supraventricular tachycardia (STACH), and general supraventricular tachycardia (GSVT). The class mapping is as follows:
| Base Class | Target Class | Label |
|---|---|---|
| 0-SR | 0 | Sinus Rhythm (SR) |
| 1-SBRAD | 1 | Sinus Bradycardia (SBRAD) |
| 7-AFIB, 8-AFL | 2 | AFIB/AFL (AFIB) |
| 2-STACH, 5-SVT | 3 | General supraventricular tachycardia (GSVT) |
Datasets
Section titled “Datasets”The model is trained on the following datasets:
Model Performance
Section titled “Model Performance”The confusion matrix for the model is depicted below.
Downloads
Section titled “Downloads”| Asset | Description |
|---|---|
| configuration.json | Configuration file |
| model.keras | Keras Model file |
| model.tflite | TFLite Model file |
| metrics.json | Metrics file |