Reconstructing Quantitative Cerebral Perfusion Images Directly From Measured Sinogram Data Acquired Using C-arm Cone-Beam CT

TL;DR

This work tackles the challenge of obtaining quantitative cerebral perfusion images with C-arm CBCT by introducing TRAINER, a joint optimization framework that directly reconstructs perfusion parametric maps from measured sinogram data. TRAINER represents the perfusion images as a subject-specific conditional generative network and enforces consistency with a forward time-resolved CT model and a convolutional perfusion model, avoiding handcrafted regularization. Empirical results show TRAINER yields accurate CBF and MTT maps across tissue types even at low exposure and slow gantry speeds, outperforming traditional deconvolution-based baselines in both qualitative and quantitative metrics. This approach has practical potential to shorten door-to-puncture times and enable in-situ perfusion assessment in interventional settings.

Abstract

To shorten the door-to-puncture time for better treating patients with acute ischemic stroke, it is highly desired to obtain quantitative cerebral perfusion images using C-arm cone-beam computed tomography (CBCT) equipped in the interventional suite. However, limited by the slow gantry rotation speed, the temporal resolution and temporal sampling density of typical C-arm CBCT are much poorer than those of multi-detector-row CT in the diagnostic imaging suite. The current quantitative perfusion imaging includes two cascaded steps: time-resolved image reconstruction and perfusion parametric estimation. For time-resolved image reconstruction, the technical challenge imposed by poor temporal resolution and poor sampling density causes inaccurate quantification of the temporal variation of cerebral artery and tissue attenuation values. For perfusion parametric estimation, it remains a technical challenge to appropriately design the handcrafted regularization for better solving the associated deconvolution problem. These two challenges together prevent obtaining quantitatively accurate perfusion images using C-arm CBCT. The purpose of this work is to simultaneously address these two challenges by combining the two cascaded steps into a single joint optimization problem and reconstructing quantitative perfusion images directly from the measured sinogram data. In the developed direct cerebral perfusion parametric image reconstruction technique, TRAINER in short, the quantitative perfusion images have been represented as a subject-specific conditional generative model trained under the constraint of the time-resolved CT forward model, perfusion convolutional model, and the subject's own measured sinogram data. Results shown in this paper demonstrated that using TRAINER, quantitative cerebral perfusion images can be accurately obtained using C-arm CBCT in the interventional suite.

Figures (7)

• Figure 1: (a) The architecture of $\mathcal{G}$. $n_u[i]$, $n_d[i]$, and $n_s[i]$ denote the number of filters for upsampling, downsampling, and skip connections at the $i$-th depth respectively. Parameters $k_u[i]$, $k_d[i]$, and $k_s[i]$ denote the kernel sizes at the $i$-th depth respectively. (b) Conditional image generation. During each iteration, the network parameters $\mathbf{\Theta}$ determine the image $\mathcal{G}_\mathbf{\Theta}(\mathbf{Z})$, and the mapping $\mathcal{G}$ is represented as a deep convolutional neural network parameterized by $\mathbf{\Theta}$.
• Figure 2: The change of loss function value (Eq. \ref{['eq:unconstrained-optimization']}) with respect to each iteration.
• Figure 3: CBF images of numerical phantom generated from the ground truth, baseline methods, and TRAINER. Data were simulated at gantry rotation speed of $dt=8$ sec and four exposure levels. Display windows are shown as the color bar. All images are shown with a W/L: 60/30.
• Figure 4: MTT images of numerical phantom generated from the ground truth, baseline methods, and TRAINER. Data were simulated at gantry rotation speed of $dt=8$ sec and four exposure levels. Display windows are shown as the color bar. All images are shown with a W/L: 18/9.
• Figure 5: Comparison of time attenuation curves for different tissue types. The plots illustrate the time attenuation curves for healthy tissue (top), penumbra (middle), and ischemic core (bottom). The TACs generated by FBP (blue squares), TRT (black circles), eSMART (gray plus), and TRAINER (red stars) are compared with the ground truth (green circles).