Deep residual-convolutional neural networks for event positioning in a monolithic annular PET scanner.
Deep residual-convolutional neural networks for event positioning in a monolithic annular PET scanner.
复制标题
DOI：
10.1088/1361-6560/ac0d0c
复制
发表时间：
2021-07-12
影响因子：
3.5
通讯作者：
Raylman RR
中科院分区：
文献类型：
--
作者：
Jaliparthi G;Martone PF;Stolin AV;Raylman RR
与传统的PET扫描仪相比，基于单片闪烁体的PET扫描仪可以潜在地产生更优越的性能特征(例如，高空间分辨率和检测灵敏度)。因此，我们开始了临床前PET系统的开发，该系统基于一个7.2厘米长的Lyso环，称为AnnPET。虽然该系统可以促进高质量图像的创建，但其独特的几何结构导致光学系统可能会使探测器中事件定位的估计复杂化。为了应对这一挑战，我们评估了深度残差卷积神经网络(DR-CNN)来估计湮灭光子相互作用的三维位置。AnnPET扫描仪的蒙特卡罗模拟被用来复制扫描仪的物理，包括光学。结果表明，十层DR-CNN最适合应用于AnnPET。已知事件位置之间的误差，以及由该网络估计的误差和使用常用的质心算法(COM)计算的误差被用来评估性能。基于十层DR-CNN的事件位置的平均绝对误差(MAE)沿x(轴向)、y(横轴)和z(相互作用深度)轴分别为0.54 mm、0.42 mm和0.45 mm。对于COM估计，x、y和z方向的平均误差分别为1.22 mm、1.04 mm和2.79 mm。用3D-FBP算法(5 mm源偏移量)重建网络估计数据得到的空间分辨率(半高全宽)为0.8 mm(径向)、0.7 mm(切向)和0.71 mm(轴向)。COM数据的重建得到了1.15 mm(径向)、0.96 mm(切向)和1.14 mm(轴向)的空间分辨率(FWHM)。这些发现表明，与标准分析方法相比，使用基于闪烁体整体环的PET扫描仪的十层DR-CNN具有产生优异性能的潜力。
PET scanners based on monolithic pieces of scintillator can potentially produce superior performance characteristics (high spatial resolution and detection sensitivity, for example) compared to conventional PET scanners. Consequently, we initiated development of a preclinical PET system based on a single 7.2 cm long annulus of LYSO, called AnnPET. While this system could facilitate creation of high-quality images, its unique geometry results in optics that can complicate estimation of event positioning in the detector. To address this challenge, we evaluated deep-residual convolutional neural networks (DR-CNN) to estimate the three-dimensional position of annihilation photon interactions. Monte Carlo simulations of the AnnPET scanner were used to replicate the physics, including optics, of the scanner. It was determined that a ten-layer-DR-CNN was most suited to application with AnnPET. The errors between known event positions, and those estimated by this network and those calculated with the commonly used center-of-mass algorithm (COM) were used to assess performance. The mean absolute errors (MAE) for the ten-layer-DR-CNN-based event positions were 0.54 mm, 0.42 mm and 0.45 mm along the x (axial)-, y (transaxial)- and z- (depth-of-interaction) axes, respectively. For COM estimates, the MAEs were 1.22 mm, 1.04 mm and 2.79 mm in the x-, y- and z-directions, respectively. Reconstruction of the network-estimated data with the 3D-FBP algorithm (5 mm source offset) yielded spatial resolutions (full-width-at-half-maximum (FWHM)) of 0.8 mm (radial), 0.7 mm (tangential) and 0.71 mm (axial). Reconstruction of the COM-derived data yielded spatial resolutions (FWHM) of 1.15 mm (radial), 0.96 mm (tangential) and 1.14 mm (axial). These findings demonstrated that use of a ten-layer-DR-CNN with a PET scanner based on a monolithic annulus of scintillator has the potential to produce excellent performance compared to standard analytical methods.
登录
查看更多内容
影响因子：
1.8
作者：
Barrett HH;Hunter WC;Miller BW;Moore SK;Chen Y;Furenlid LR
通讯作者：
Furenlid LR
影响因子：
1.6
作者：
ANGER, HO
通讯作者：
ANGER, HO
影响因子：
3.5
作者：
Borghi, Giacomo;Peet, Bart Jan;Schaart, Dennis R.
通讯作者：
Schaart, Dennis R.
影响因子：
3.5
作者：
Ito, Mikiko;Lee, Jae Sung;Hong, Seong Jong
通讯作者：
Hong, Seong Jong
影响因子：
1.8
作者：
Hunter WC;Barrett HH;Furenlid LR
通讯作者：
Furenlid LR