A deep multi-stream model for robust prediction of left ventricular ejection fraction in 2D echocardiography
Journal article, 2024
We propose a deep multi-stream model for left ventricular ejection fraction (LVEF) prediction in 2D echocardiographic (2DE) examinations. We use four standard 2DE views as model input, which are automatically selected from the full 2DE examination. The LVEF prediction model processes eight streams of data (images + optical flow) and consists of convolutional neural networks terminated with transformer layers. The model is made robust to missing, misclassified and duplicate views via pre-training, sampling strategies and parameter sharing. The model is trained and evaluated on an existing clinical dataset (12,648 unique examinations) with varying properties in terms of quality, examining physician, and ultrasound system. We report R2= 0.84 and mean absolute error = 4.0% points for the test set. When evaluated on two public benchmarks, the model performs on par or better than all previous attempts on fully automatic LVEF prediction. Code and trained models are available on a public project repository .
Author
Jennifer Alvén
University of Gothenburg
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Eva Hagberg
Sahlgrenska University Hospital
University of Gothenburg
David Hagerman Olzon
Chalmers, Electrical Engineering, Signal Processing and Biomedical Engineering
Sahlgrenska University Hospital
University of Gothenburg
Richard Petersen
Sahlgrenska University Hospital
University of Gothenburg
Ola Hjelmgren
Sahlgrenska University Hospital
University of Gothenburg
Scientific Reports
2045-2322 (ISSN) 20452322 (eISSN)
Vol. 14 1 2104Subject Categories
Cardiac and Cardiovascular Systems
DOI
10.1038/s41598-024-52480-y