Homodyned K-distribution: parameter estimation and uncertainty quantification using Bayesian neural networks

TL;DR

This paper introduces a Bayesian Neural Network approach to estimate Homodyned K-distribution parameters in ultrasound imaging, providing both parameter estimates and uncertainty quantification to improve tissue property analysis.

Contribution

It presents a novel application of Bayesian neural networks for reliable estimation and uncertainty quantification of HK-distribution parameters in ultrasound data.

Findings

BNN effectively estimates HK-distribution parameters.

Uncertainty quantification improves confidence in tissue property estimates.

Method enhances robustness in diverse scattering conditions.

Abstract

Quantitative ultrasound (QUS) allows estimating the intrinsic tissue properties. Speckle statistics are the QUS parameters that describe the first order statistics of ultrasound (US) envelope data. The parameters of Homodyned K-distribution (HK-distribution) are the speckle statistics that can model the envelope data in diverse scattering conditions. However, they require a large amount of data to be estimated reliably. Consequently, finding out the intrinsic uncertainty of the estimated parameters can help us to have a better understanding of the estimated parameters. In this paper, we propose a Bayesian Neural Network (BNN) to estimate the parameters of HK-distribution and quantify the uncertainty of the estimator.