Perfusion Imaging and Single Material Reconstruction in Polychromatic Photon Counting CT

TL;DR

The paper addresses dose-efficient perfusion CT by reconstructing dynamic iodine maps from polychromatic photon-counting CT data with known background, using VI-PRISM — a monotone variational-inequality based method. VI-PRISM operates on a polychromatic forward model $\mathbb{E}[y_{\omega,i}(x_{\text{iodine}})]$ and iterates via $x^{t+1} = P_{\mathcal{X}}(x^t - \alpha_t F(x^t))$ to recover the iodine map while enforcing nonnegativity and a TV constraint. Across varied photon budgets and view counts, VI-PRISM achieves consistently lower RMSE, reduced noise, and higher SNR than FBP, with iodine concentration errors often below $0.4\ \mathrm{mg/ml}$ even under 10x–100x dose reductions. These results suggest that VI-PRISM can enable robust, dose-efficient quantitative perfusion imaging in photon-counting CT, though further validation on physical phantoms and clinical data is needed and future work will focus on acceleration and multi-material extensions.

Abstract

Background: Perfusion computed tomography (CT) images the dynamics of a contrast agent through the body over time, and is one of the highest X-ray dose scans in medical imaging. Recently, a theoretically justified reconstruction algorithm based on a monotone variational inequality (VI) was proposed for single material polychromatic photon-counting CT, and showed promising early results at low-dose imaging. Purpose: We adapt this reconstruction algorithm for perfusion CT, to reconstruct the concentration map of the contrast agent while the static background tissue is assumed known; we call our method VI-PRISM (VI-based PeRfusion Imaging and Single Material reconstruction). We evaluate its potential for dose-reduced perfusion CT, using a digital phantom with water and iodine of varying concentration. Methods: Simulated iodine concentrations range from 0.05 to 2.5 mg/ml. The simulated X-ray source emits photons up to 100 keV, with average intensity ranging from $10^5$ down to $10^2$ photons per detector element. The number of tomographic projections was varied from 984 down to 8 to characterize the tradeoff in photon allocation between views and intensity. Results: We compare VI-PRISM against filtered back-projection (FBP), and find that VI-PRISM recovers iodine concentration with error below 0.4 mg/ml at all source intensity levels tested. Even with a dose reduction between 10x and 100x compared to FBP, VI-PRISM exhibits reconstruction quality on par with FBP. Conclusion: Across all photon budgets and angular sampling densities tested, VI-PRISM achieved consistently lower RMSE, reduced noise, and higher SNR compared to filtered back-projection. Even in extremely photon-limited and sparsely sampled regimes, VI-PRISM recovered iodine concentrations with errors below 0.4 mg/ml, showing that VI-PRISM can support accurate and dose-efficient perfusion imaging in photon-counting CT.

Figures (10)

• Figure 1: Visualization of the spatial distribution of (a) air, (b) water, and (c) iodine in the ground truth phantom, along with (d) the ground truth HU image.
• Figure 2: Noise and signal to noise ratio (SNR) as a function of total photon budget and number of projections. Each curve corresponds to a different number of projections for the given total budget. Solid lines correspond to VI-PRISM, and dotted lines correspond to FBP. Subfigure (a) reports noise and subfigure (b) reports SNR, both evaluated within the ring mask shown in green in the inset.
• Figure 3: RMSE versus number of views, evaluated within the ring mask inset shown in subfigure (h), for both VI-PRISM and FBP under different total photon budgets. The first row shows RMSE values for both VI-PRISM and FBP reconstructions, while the second row focuses on VI-PRISM. Error bars indicate the standard deviation across 9 random seeds. Star annotations indicate statistical significance after Bonferroni correction, denoting whether the RMSE with a smaller number of projections is statistically different from the RMSE with the same photon budget spread among the full set of $984$ projections.
• Figure 4: HU error for VI-PRISM reconstruction as a function of number of views, for each total photon budget. Error bars indicate the standard deviation across 9 random seeds. Evaluation is within each iodine insert, as visualized in (a); the title of each subfigure details the ground truth CT number (HU) in the corresponding insert.
• Figure 5: Iodine concentration error (mg/ml) for VI-PRISM reconstruction as a function of number of views, for each total photon budget. Error bars indicate the standard deviation across 9 random seeds. Evaluation is within each iodine insert, as visualized in (a); the title of each subfigure details the ground truth iodine concentration in the corresponding insert.