Effects of Important Parameters Variations on Computing Eigenspace-Based Minimum Variance Weights for Ultrasound Tissue Harmonic Imaging

2018-02-26

In recent years, the minimum variance (MV) beamforming has been widely studied due to its high resolution and contrast in B-mode Ultrasound imaging (USI). However, the performance of the MV beamformer is degraded at the presence of noise, as a result of the inaccurate covariance matrix estimation which leads to a low quality image. Second harmonic imaging (SHI) provides many advantages over the conventional pulse-echo USI, such as enhanced axial and lateral resolutions. However, the low signal-to-noise ratio (SNR) is a major problem in SHI. In this paper, Eigenspace-based minimum variance (EIBMV) beamformer has been employed for second harmonic USI. The Tissue Harmonic Imaging (THI) is achieved by Pulse Inversion (PI) technique. Using the EIBMV weights, instead of the MV ones, would lead to reduced sidelobes and improved contrast, without compromising the high resolution of the MV beamformer (even at the presence of a strong noise). In addition, we have investigated the effects of variations of the important parameters in computing EIBMV weights, i.e., K, L, and {\delta}, on the resolution and contrast obtained in SHI. The results are evaluated using numerical data (using point target and cyst phantoms), and the proper parameters of EIBMV are indicated for THI.

📄 PDF Abstract BibTeX arXiv:1802.09316

Code (0)

등록된 구현이 없습니다.

Communication-efficient distributed eigenspace estimation with arbitrary node failures

2022-05-31 · Vasileios Charisopoulos, Anil Damle

We develop an eigenspace estimation algorithm for distributed environments with arbitrary node failures, where a subset of computing nodes can return structurally valid but otherwise arbitrarily chosen responses. Notably…

Distributed Computing valid

Connecting Parameter Magnitudes and Hessian Eigenspaces at Scale using Sketched Methods

2025-04-20 · Andres Fernandez, Frank Schneider, Maren Mahsereci, Philipp Hennig

Recently, it has been observed that when training a deep neural net with SGD, the majority of the loss landscape's curvature quickly concentrates in a tiny *top* eigenspace of the loss Hessian, which remains largely stab…

Privacy-Preserving Transfer Learning for Community Detection using Locally Distributed Multiple Networks

2025-04-01 · Xiao Guo, Xuming He, Xiangyu Chang, Shujie Ma

This paper develops a new spectral clustering-based method called TransNet for transfer learning in community detection of network data. Our goal is to improve the clustering performance of the target network using auxil…

Clustering Community Detection Privacy Preserving Transfer Learning

On the Downstream Performance of Compressed Word Embeddings

2019-09-03 · NeurIPS 2019 12 · Avner May, Jian Zhang, Tri Dao, Christopher Ré

Compressing word embeddings is important for deploying NLP models in memory-constrained settings. However, understanding what makes compressed embeddings perform well on downstream tasks is challenging---existing measure…

Generalization Bounds Quantization Word Embeddings

FedPower: Privacy-Preserving Distributed Eigenspace Estimation

2021-03-01 · Xiao Guo, Xiang Li, Xiangyu Chang, Shusen Wang 외

Eigenspace estimation is fundamental in machine learning and statistics, which has found applications in PCA, dimension reduction, and clustering, among others. The modern machine learning community usually assumes that …

BIG-bench Machine Learning Dimensionality Reduction Federated Learning Privacy Preserving