Research map / Video / Video segmentation and tracking
CardiacNet: Learning to Reconstruct Abnormalities for Cardiac Disease Assessment from Echocardiogram Videos
Jiewen Yang, Yiqun Lin, Bin Pu, Jiarong GUO, Xiaowei Xu, Xiaomeng Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add the Consistency Deformation Codebook (CDC) with no CDD (ablation) | CardiacNet · CardiacNet-PAH | ACC (%): 46.15 -> 79.59 | Improved |
| add the CDD discriminator on top of CDC (ablation) | CardiacNet · CardiacNet-PAH | ACC (%): 79.59 -> 85.71 | Improved |
| add CDD without CDC (ablation) | CardiacNet · CardiacNet-PAH | ACC (%): 46.15 -> 53.84 | No clear effect |
| add learnable temporal positional encoding to CDC (ablation) | CDC module · CardiacNet-PAH (CDC only) | ACC (%): 75.51 -> 79.59 | Improved |
| add optimal transport distance loss to CDC (ablation) | CDC module · CardiacNet-PAH (CDC only) | ACC (%): 65.38 -> 79.59 | Improved |
| add both global and local discriminators instead of one (ablation) | CDD module · CardiacNet-PAH (CDC enabled) | ACC (%): 83.67 -> 85.71 | Improved |
| add global discriminator only (ablation) | CDD module · CardiacNet-PAH (CDC enabled) | ACC (%): 79.59 -> 83.67 | Improved |
| add local discriminator only (ablation) | CDD module · CardiacNet-PAH (CDC enabled) | ACC (%): 79.59 -> 81.63 | Improved |
| reconstruction-based CardiacNet vs best classification baseline HiFuse | CardiacNet · CardiacNet-PAH | AUC-ROC (%): 84.11 -> 89.32 | Improved |
| CardiacNet vs best baseline DeepGuide | CardiacNet · CardiacNet-ASD | AUC-ROC (%): 85.02 -> 91.24 | Improved |
| CardiacNet vs best baseline for EF regression | CardiacNet · EchoNet-Dynamic | EF MAE: 4.08 -> 3.83 | Improved |
| CardiacNet reconstruction quality vs Wolleb et al. | CardiacNet · EchoNet-Dynamic | FID: 13.25 vs 13.18 | No clear effect |
Closest papers by meaning
- EchoXFlow: A Beamspace Echocardiography Dataset for Cardiac Motion, Flow, and Function (NeurIPS 2026 · similarity 0.88)
- Self-supervised Learning of Echocardiographic Video Representations via Online Cluster Distillation (NeurIPS 2025 · similarity 0.86)
- CardioLens: Revealing the Clinical Reality Gap of MLLMs via Multi-Sequence Cardiac MRI Evaluations (NeurIPS 2026 · similarity 0.86)
- Bidirectional Recurrence for Cardiac Motion Tracking with Gaussian Process Latent Coding (NeurIPS 2024 · similarity 0.86)
- Q-PART: Quasi-Periodic Adaptive Regression with Test-time Training for Pediatric Left Ventricular Ejection Fraction Regression (CVPR 2025 · similarity 0.83)
- EchoONE: Segmenting Multiple Echocardiography Planes in One Model (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
