arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2401.14970eess.SPeess.IV
收藏

基于轮廓辅助反投影深度学习的微波淋巴水肿评估

Microwave lymphedema assessment using deep learning with contour assisted backprojection

Yuyi Chang, Nithin Sugavanam, Emre Ertin

AI总结:

本研究提出一种结合轮廓辅助反投影微波成像与U-Net深度网络的淋巴水肿早期检测方法,通过肢体轮廓优化成像聚焦,仿真实验证实其显著提升了肢体淋巴液积聚的位置与体积识别能力。

摘要:

我们提出了一种淋巴水肿患者淋巴液积聚的早期检测方法,该方法基于对肢体体积跨空气间隙的微波成像。所提出的算法利用被成像肢体表面的轮廓信息来局部近似波传播速度,以求解程函方程,从而实现伴随成像算子。这种改进的反投影算法可在淋巴液积聚出现的肢体表面附近生成聚焦图像。接下来,采用基于U-Net架构的深度神经网络来识别淋巴液的位置和范围。我们通过多种上臂和下臂轮廓的仿真研究,对比了所提出的轮廓辅助反投影成像与假设介质均匀的基线成像方法的聚焦性能。仿真实验的实证结果表明,所提出的成像方法显著提升了深度网络模型识别肢体中多余积液位置和体积的能力。

英文摘要:

We present a method for early detection of lymphatic fluid accumulation in lymphedema patients based on microwave imaging of the limb volume across an air gap. The proposed algorithm uses contour information of the imaged limb surface to approximate the wave propagation velocity locally to solve the eikonal equation for implementing the adjoint imaging operator. This modified backprojection algorithm results in focused imagery close to the limb surface where lymphatic fluid accumulation presents itself. Next, a deep neural network based on U-Net architecture is employed to identify the location and extent of the lymphatic fluid. Simulation studies with various upper and lower arm profiles compare the focusing performance of the proposed contour assisted backprojection imaging with the baseline imaging approach that assumes homogeneous media. The empirical results of the simulation experiments show that the proposed imaging method significantly improves the ability of the deepnet model to identify the location and the volume of the excess fluid in the limb.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑