EchoFM: Foundation Model for Generalizable Echocardiogram Analysis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kim, Sekeun, Jin, Pengfei, Song, Sifan, Chen, Cheng, Li, Yiwei, Ren, Hui, Li, Xiang, Liu, Tianming, Li, Quanzheng
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913668696899584
author Kim, Sekeun
Jin, Pengfei
Song, Sifan
Chen, Cheng
Li, Yiwei
Ren, Hui
Li, Xiang
Liu, Tianming
Li, Quanzheng
author_facet Kim, Sekeun
Jin, Pengfei
Song, Sifan
Chen, Cheng
Li, Yiwei
Ren, Hui
Li, Xiang
Liu, Tianming
Li, Quanzheng
contents Foundation models have recently gained significant attention because of their generalizability and adaptability across multiple tasks and data distributions. Although medical foundation models have emerged, solutions for cardiac imaging, especially echocardiography videos, are still unexplored. In this paper, we introduce EchoFM, a foundation model specifically designed to represent and analyze echocardiography videos. In EchoFM, we propose a self-supervised learning framework that captures both spatial and temporal variability patterns through a spatio-temporal consistent masking strategy and periodic-driven contrastive learning. This framework can effectively capture the spatio-temporal dynamics of echocardiography and learn the representative video features without any labels. We pre-train our model on an extensive dataset comprising over 290,000 echocardiography videos covering 26 scan views across different imaging modes, with up to 20 million frames of images. The pre-trained EchoFM can then be easily adapted and fine-tuned for a variety of downstream tasks, serving as a robust backbone model. Our evaluation was systemically designed for four downstream tasks after the echocardiography examination routine. Experiment results show that EchoFM surpasses state-of-the-art methods, including specialized echocardiography methods, self-supervised pre-training models, and general-purposed pre-trained foundation models, across all downstream tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23413
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EchoFM: Foundation Model for Generalizable Echocardiogram Analysis
Kim, Sekeun
Jin, Pengfei
Song, Sifan
Chen, Cheng
Li, Yiwei
Ren, Hui
Li, Xiang
Liu, Tianming
Li, Quanzheng
Computer Vision and Pattern Recognition
Foundation models have recently gained significant attention because of their generalizability and adaptability across multiple tasks and data distributions. Although medical foundation models have emerged, solutions for cardiac imaging, especially echocardiography videos, are still unexplored. In this paper, we introduce EchoFM, a foundation model specifically designed to represent and analyze echocardiography videos. In EchoFM, we propose a self-supervised learning framework that captures both spatial and temporal variability patterns through a spatio-temporal consistent masking strategy and periodic-driven contrastive learning. This framework can effectively capture the spatio-temporal dynamics of echocardiography and learn the representative video features without any labels. We pre-train our model on an extensive dataset comprising over 290,000 echocardiography videos covering 26 scan views across different imaging modes, with up to 20 million frames of images. The pre-trained EchoFM can then be easily adapted and fine-tuned for a variety of downstream tasks, serving as a robust backbone model. Our evaluation was systemically designed for four downstream tasks after the echocardiography examination routine. Experiment results show that EchoFM surpasses state-of-the-art methods, including specialized echocardiography methods, self-supervised pre-training models, and general-purposed pre-trained foundation models, across all downstream tasks.
title EchoFM: Foundation Model for Generalizable Echocardiogram Analysis
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.23413