SAID-Net: enhancing segment anything model with implicit decoding for echocardiography sequences segmentation

Yagang Wu

doi:10.1007/s11517-025-03419-6

2025-07-17

PRE

郑洲顺 (Zhoushun Zheng) *

DOI：10.1007/s11517-025-03419-6

原文链接

原文求助

摘要

En 中文

超声心动图序列分割在现代心脏病学中至关重要。尽管分割任意模型（Segment Anything Model, SAM）在通用分割中表现优异，但其直接应用于超声心动图面临挑战，原因在于复杂的心脏解剖结构和微妙的超声边界。我们引入SAID（使用隐式解码的分割任意模型），一种将隐式神经表征（Implicit Neural Representations, INR）与SAM相结合的新型框架，以提升准确性、适应性和鲁棒性。SAID采用基于Hiera的编码器进行多尺度特征提取，并使用掩码单元注意力解码器捕获精细细节，这对心脏轮廓描绘至关重要。正交化增强了特征多样性，而I^2 Net改进了对错位上下文特征的处理。在CAMUS和EchoNet-Dynamics数据集上的测试表明，SAID优于当前最先进的方法，在CAMUS上实现了93.2%的Dice相似系数（DSC）和5.02 mm的Hausdorff距离（HD95），在EchoNet-Dynamics上实现了92.3%的DSC和4.05 mm的HD95，证实了其在超声心动图序列分割中的有效性和鲁棒性。SAID-Net：推进SAM用于超声心动图序列分割

Keyword:

Echocardiography sequences segmentation

Segment Anything Model

Implicit neural representations

Orthogonal mining

期刊

Medical and Biological Engineering and Computing

IF:

2.6

论文数:

311

被引数:

7.7K

机构

School of Automation

学者数:

741

论文数: 290

被引数: 0

School of Computer Science and Engineering

学者数:

1.3K

论文数: 590

被引数: 2

Department of Cardiovascular Surgery

学者数:

374

论文数: 146

被引数: 1

School of Mathematics and Statistics

学者数:

906

论文数: 486

被引数: 0

查看更多机构

引用论文

Segment anything in medical images 分割医学图像中的任何内容

NATURE COMMUNICATIONS

IF15.7

2024-01-22

274

OAAI

Ma, Jun; He, Yuting; Li, Feifei; Han, Lin; You, Chenyu; Wang, Bo

U-Net: Convolutional Networks for Biomedical Image Segmentation U-net: 用于生物医学图像分割的卷积网络

Lecture Notes in Computer Science

IF0

2015-11-18

PREAI

Olaf Ronneberger; Philipp Fischer; Thomas Brox

Deep Learning for Segmentation Using an Open Large-Scale Dataset in 2D Echocardiography

IEEE TRANSACTIONS ON MEDICAL IMAGING

IF9.8

2019-09-01

362

OAAI

Leclerc, Sarah; Smistad, Erik; Pedrosa, Joao; Ostvik, Andreas; Cervenansky, Frederic; Espinosa, Florian; Espeland, Torvald; Berg, Erik Andreas Rye; Jodoin, Pierre-Marc; Grenier, Thomas; Lartizien, Carole; D'hooge, Jan; Lovstakken, Lasse; Bernard, Olivier

Segment anything model for medical images?分割医学图像的任何模型？

MEDICAL IMAGE ANALYSIS

IF11.8

2024-02-01

OAAI

Huang, Yuhao; Yang, Xin; Liu, Lian; Zhou, Han; Chang, Ao; Zhou, Xinrui; Chen, Rusi; Yu, Junxuan; Chen, Jiongquan; Chen, Chaoyu; Liu, Sijing; Chi, Haozhe; Hu, Xindi; Yue, Kejuan; Li, Lei; Grau, Vicente; Fan, Deng-Ping; Dong, Fajin; Ni, Dong

PraNet: Parallel Reverse Attention Network for Polyp Segmentation PraNet: 并行反向注意网络用于息肉分割

Lecture Notes in Computer Science

IF0

2020-09-29

PREAI

Deng-Ping Fan; Ge-Peng Ji; Tao Zhou; Geng Chen; Huazhu Fu; Jianbing Shen; Ling Shao

Fast and Fully Automatic Left Ventricular Segmentation and Tracking in Echocardiography Using Shape-Based B-Spline Explicit Active Surfaces

IEEE Transactions on Medical Imaging

IF0

2017-11-01

PREAI

Joao Pedrosa; Sandro Queiros; Olivier Bernard; Jan Engvall; Thor Edvardsen; Eike Nagel; Jan D'hooge

SonoSAM - Segment Anything on Ultrasound Images

Lecture Notes in Computer Science

IF0

2023-10-01

PREAI

Hariharan Ravishankar; Rohan Patil; Vikram Melapudi; Pavan Annangi

E-Res U-Net: An improved U-Net model for segmentation of muscle images

EXPERT SYSTEMS WITH APPLICATIONS

IF7.5

2021-12-01

PREAI

Zhou, Junsheng; Lu, Yiwen; Tao, Siyi; Cheng, Xuan; Huang, Chenxi

Contour tracking in echocardiographic sequences via sparse representation and dictionary learning

MEDICAL IMAGE ANALYSIS

IF11.8

2014-02-01

OAAI

Huang, Xiaojie; Dione, Donald P.; Compas, Colin B.; Papademetris, Xenophon; Lin, Ben A.; Bregasi, Alda; Sinusas, Albert J.; Staib, Lawrence H.; Duncan, James S.

Deep Learning for Cardiac Image Segmentation: A Review

Frontiers in Cardiovascular Medicine

IF0

2020-03-05

OAAI

Chen Chen; Chen Qin; Huaqi Qiu; Giacomo Tarroni; Jinming Duan; Wenjia Bai; Daniel Rueckert

查看更多内容