Skip to main content
QUICK REVIEW

[论文解读] Deep learning-based attenuation correction in the image domain for myocardial perfusion SPECT imaging

Samaneh Mostafapour, Faeze Gholamiankhah|arXiv (Cornell University)|Feb 9, 2021
Medical Imaging Techniques and Applications参考文献 45被引用 6
一句话总结

本研究提出了一种基于深度学习的图像域衰减校正(AC)方法,用于心肌灌注SPECT(MPI-SPECT),采用ResNet和UNet架构,直接从非衰减校正的SPECT扫描预测基于CT的AC图像。与传统的Chang方法相比,该方法表现出更优的准确性,平均绝对误差(MAE)分别为-6.99±16.72和-4.41±11.80,结构相似性指数(SSIM)分别为0.99±0.04和0.98±0.05,显示出与参考CT-AC图像高度一致的临床表现。

ABSTRACT

Objective: In this work, we set out to investigate the accuracy of direct attenuation correction (AC) in the image domain for the myocardial perfusion SPECT imaging (MPI-SPECT) using two residual (ResNet) and UNet deep convolutional neural networks. Methods: The MPI-SPECT 99mTc-sestamibi images of 99 participants were retrospectively examined. UNet and ResNet networks were trained using SPECT non-attenuation corrected images as input and CT-based attenuation corrected SPECT images (CT-AC) as reference. The Chang AC approach, considering a uniform attenuation coefficient within the body contour, was also implemented. Quantitative and clinical evaluation of the proposed methods were performed considering SPECT CT-AC images of 19 subjects as reference using the mean absolute error (MAE), structural similarity index (SSIM) metrics, as well as relevant clinical indices such as perfusion deficit (TPD). Results: Overall, the deep learning solution exhibited good agreement with the CT-based AC, noticeably outperforming the Chang method. The ResNet and UNet models resulted in the ME (count) of ${-6.99\pm16.72}$ and ${-4.41\pm11.8}$ and SSIM of ${0.99\pm0.04}$ and ${0.98\pm0.05}$, respectively. While the Change approach led to ME and SSIM of ${25.52\pm33.98}$ and ${0.93\pm0.09}$, respectively. Similarly, the clinical evaluation revealed a mean TPD of ${12.78\pm9.22}$ and ${12.57\pm8.93}$ for the ResNet and UNet models, respectively, compared to ${12.84\pm8.63}$ obtained from the reference SPECT CT-AC images. On the other hand, the Chang approach led to a mean TPD of ${16.68\pm11.24}$. Conclusion: We evaluated two deep convolutional neural networks to estimate SPECT-AC images directly from the non-attenuation corrected images. The deep learning solutions exhibited the promising potential to generate reliable attenuation corrected SPECT images without the use of transmission scanning.

研究动机与目标

  • 开发一种基于深度学习的直接衰减校正方法,仅使用发射图像进行心肌灌注SPECT(MPI-SPECT)的衰减校正。
  • 通过直接从非衰减校正的SPECT图像学习衰减校正,消除对传输扫描的需求。
  • 评估深度学习模型(ResNet和UNet)与传统Chang AC方法及基于CT的参考图像的性能表现。
  • 通过灌注缺损(TPD)和图像质量指标(MAE、SSIM)等临床指标评估其临床相关性。
  • 证明深度学习可生成与金标准CT-AC图像相当的可靠AC图像,且无需额外的成像协议。

提出的方法

  • 在99例MPI-SPECT 99mTc-sestamibi研究中训练了两个深度卷积神经网络——ResNet和UNet。
  • 以非衰减校正的SPECT图像作为输入,以基于CT的衰减校正SPECT图像作为真实标签参考。
  • 在ResNet中应用残差连接,在UNet中应用跳跃连接,以改善特征学习和梯度传播。
  • 端到端训练模型,直接在图像域中预测衰减校正后的SPECT图像。
  • 使用平均绝对误差(MAE)、结构相似性指数(SSIM)以及临床灌注缺损(TPD)评估性能。
  • 使用相同数据集和参考图像,将结果与传统的Chang AC方法进行对比。

实验结果

研究问题

  • RQ1深度学习模型是否能够在无传输扫描的情况下,直接在图像域中学习到准确的衰减校正?
  • RQ2与传统的Chang方法相比,基于ResNet和UNet的AC在图像质量和定量指标方面表现如何?
  • RQ3基于深度学习的AC结果在多大程度上与金标准CT-AC图像的临床指标(如TPD)相匹配?
  • RQ4所提出的方法是否能在避免CT传输扫描的同时,保持高结构相似性和低误差?
  • RQ5该深度学习方法是否足够稳健,能够针对不同患者解剖结构生成临床可靠的衰减校正SPECT图像?

主要发现

  • ResNet模型实现了-6.99±16.72的平均误差(ME)和0.99±0.04的SSIM,表明其与CT-AC参考图像具有高度的结构保真度。
  • UNet模型实现了ME为-4.41±11.80和SSIM为0.98±0.05,表现出优异的图像质量和一致性。
  • Chang方法的误差显著更高(ME: 25.52±33.98),SSIM更低(0.93±0.09),表明其性能较差。
  • 临床评估显示,ResNet的平均TPD为12.78±9.22,UNet为12.57±8.93,与参考CT-AC的平均TPD(12.84±8.63)非常接近。
  • Chang方法的平均TPD为16.68±11.24,表明其对灌注缺损存在临床相关的过度估计。
  • 总体而言,深度学习模型在定量和临床指标上均优于传统Chang方法,显示出在MPI-SPECT中替代基于CT的AC的强大潜力。

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。