Implicitly Defined Material Decomposition Estimator and Learned Physics-Informed Neural Proxy for Photon Counting CT
- 1 May 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (5) , 2455-2467
Abstract
Photon counting detector-based CT (PCCT) systems provide spectral count measurements, enabling material decomposition (MD) for quantitative imaging. Maximum-likelihood estimation (MLE) for MD offers asymptotically unbiased and efficient (minimum variance) results but is usually solved iteratively, making the entire process computationally expensive and time-consuming. Conversely, representative empirical methods relying on calibration aim to construct a direct measurement-decomposition conversion, which can be fast but may suffer from bias or noise amplification. In this work, we show that the iterative MLE method implicitly defines the functional mapping from measurements to estimates, i.e., MD results, and the corresponding mean and noise yield analytical approximation forms from the Implicit Function Theorem. From this perspective, we demonstrate that it is possible to distill knowledge from the implicit function defined by the iterative MLE, i.e., finding the explicit proxy, by leveraging universal approximators such as neural networks and the derivative-aware Sobolev Training paradigm. We show that the proposed method, namely Proxy MD, is both computationally efficient (providing >200 times speedup) and approaches the performance of iterative MLE. Thus it outperforms conventional empirical methods, enabling high-quality real-time quantitative spectral imaging. Furthermore, we also demonstrate that the theoretical Jacobian analysis provides new perspectives in making iterative MD differentiable, enabling differentiable PCCT quantitative imaging and corresponding cross-domain end-to-end training and optimization. The code has been made available at: https://github.com/senwang320/ProxyMD_Demo.Keywords
Funding Information
- GE Healthcare
This publication has 34 references indexed in Scilit:
- Optimal weighting strategies for maximizing contrast‐to‐noise ratio in photon counting CT imagesMedical Physics, 2024
- The effects of intra‐detector Compton scatter on low‐frequency DQE for photon‐counting CT using edge‐on‐irradiated silicon detectorsMedical Physics, 2024
- Derivative-Informed Neural Operator: An efficient framework for high-dimensional parametric derivative learningJournal of Computational Physics, 2023
- Initial Clinical Images From a Second-Generation Prototype Silicon-Based Photon-Counting Computed Tomography SystemAcademic Radiology, 2023
- Empirical optimization of energy bin weights for compressing measurements with realistic photon counting x‐ray detectorsMedical Physics, 2023
- Estimating the accuracy of dual energy chest radiography for coronary calcium detection with lateral or anteroposterior orientationsMedical Physics, 2022
- Modified nnU-Net for the MICCAI KiTS21 Challenge2022
- Feasibility of unconstrained three-material decomposition: imaging an excised human heart using a prototype silicon photon-counting CT detectorEuropean Radiology, 2020
- Experimental investigation of neural network estimator and transfer learning techniques for K-edge spectral CT imagingMedical Physics, 2020
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989