基于半监督神经网络的弹性模量分布重建  

Reconstruction of elasticity modulus distribution base on semi-supervised neural network

作　　者：张潇[1] 彭博 王锐 魏星月 罗建文[2] ZHANG Xiao;PENG Bo;WANG Rui;WEI Xingyue;LUO Jianwen(School of Computer Science and Software Engineering,Southwest Petroleum University,Chengdu 610500,P.R.China;Department of Biomedical Engineering,School of Medicine,Tsinghua University,Beijing 100084,P.R.China)

机构地区：[1]西南石油大学计算机与软件学院,成都610500 [2]清华大学医学院生物医学工程系,北京100084

出　　处：《生物医学工程学杂志》2024年第2期262-271,共10页Journal of Biomedical Engineering

基　　金：国家自然科学基金(61871251);四川省自然科学基金(2022NSFSC0833);四川省南充市科技局(SXQHJH046,SXHZ019)。

摘　　要：在超声弹性成像中,准确重建组织弹性模量分布是一项重要挑战。现有的基于深度学习的全监督重建方法在训练中只使用了添加噪声的计算机仿真位移数据,不能完全模拟在体超声数据的复杂性和多样性。因此,本研究在训练中引入对在体超声射频信号追踪得到的位移数据(即真实位移数据),对模型进行半监督训练,旨在提高网络的预测准确度。实验结果显示,在仿体实验中,加入了真实位移数据的半监督模型的平均绝对误差和平均相对误差均在3%左右,而全监督模型的相应数据在5%左右。在处理真实位移数据时,半监督模型预测错误区域明显少于全监督模型。本文研究结果证实了所提方法的有效性和实用性,为在体超声数据在弹性模量分布重建的深度学习方法中的使用提供了新思路。Accurate reconstruction of tissue elasticity modulus distribution has always been an important challenge in ultrasound elastography.Considering that existing deep learning-based supervised reconstruction methods only use simulated displacement data with random noise in training,which cannot fully provide the complexity and diversity brought by in-vivo ultrasound data,this study introduces the use of displacement data obtained by tracking in-vivo ultrasound radio frequency signals(i.e.,real displacement data)during training,employing a semi-supervised approach to enhance the prediction accuracy of the model.Experimental results indicate that in phantom experiments,the semi-supervised model augmented with real displacement data provides more accurate predictions,with mean absolute errors and mean relative errors both around 3%,while the corresponding data for the fully supervised model are around 5%.When processing real displacement data,the area of prediction error of semi-supervised model was less than that of fully supervised model.The findings of this study confirm the effectiveness and practicality of the proposed approach,providing new insights for the application of deep learning methods in the reconstruction of elastic distribution from in-vivo ultrasound data.

关 键 词：半监督 反问题 深度学习 弹性模量分布重建 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] R319[自动化与计算机技术—控制科学与工程]