EchoNet-Dynamic
PulseAugur coverage of EchoNet-Dynamic — every cluster mentioning EchoNet-Dynamic across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New framework quantifies temporal explainability in medical AI video analysis
Researchers have developed a new quantitative framework to evaluate the temporal explainability of deep learning models used in echocardiographic video segmentation. This framework uses four metrics to assess temporal c…
-
New VIDS-Seg method improves AI safety in pediatric cardiac imaging
Researchers have developed VIDS-Seg, a new method for uncertainty quantification in medical image segmentation, specifically for pediatric cardiac ultrasound. This approach, built on the VIDS framework, uses amortized v…
-
New research highlights bias in echocardiographic AI models
Researchers have identified a "Conditioning-Availability Bias" in echocardiographic segmentation models, where auxiliary signals used during training are cleaner than those available at deployment. This shortcut learnin…
-
New research explores cardiovascular digital twins and calibration methods · 2 sources tracked
Two new arXiv papers explore the development and calibration of digital twins for cardiovascular health. The first paper reviews various modeling approaches, from physics-based to data-driven, highlighting the integrati…
-
AI models for heart health are spatially accurate but temporally blind
A new research paper published on arXiv investigates the attribution methods used to explain the decisions of deep learning models in echocardiography. The study found that while these models can accurately estimate lef…
-
New STLSF module enhances echocardiography segmentation accuracy
Researchers have developed a novel STLSF module to improve the accuracy of deep learning models in segmenting echocardiography images, which are often plagued by noise and ambiguous boundaries. This module utilizes loca…
-
EAGT research enhances echocardiogram segmentation with geometry-based augmentation
A new research paper introduces EAGT (Echocardiography Augmentation for Generalisability and Transferability), a method designed to improve the performance of deep learning models in segmenting echocardiograms. The stud…
-
New neural atlas method slashes ultrasound annotation burden
Researchers have developed a new method for creating neural atlases of ultrasound videos, which can significantly reduce the need for expert annotations. This approach trains a single canonical chart using generative la…