Wang Sen, Yang Yirong, Stevens Grant M, Yin Zhye, Wang Adam S
IEEE Trans Med Imaging. 2025 Jan;44(1):530-542. doi: 10.1109/TMI.2024.3449817. Epub 2025 Jan 2.
Photon counting CT (PCCT) acquires spectral measurements and enables generation of material decomposition (MD) images that provide distinct advantages in various clinical situations. However, noise amplification is observed in MD images, and denoising is typically applied. Clean or high-quality references are rare in clinical scans, often making supervised learning (Noise2Clean) impractical. Noise2Noise is a self-supervised counterpart, using noisy images and corresponding noisy references with zero-mean, independent noise. PCCT counts transmitted photons separately, and raw measurements are assumed to follow a Poisson distribution in each energy bin, providing the possibility to create noise-independent pairs. The approach is to use binomial selection to split the counts into two low-dose scans with independent noise. We prove that the reconstructed spectral images inherit the noise independence from counts domain through noise propagation analysis and also validated it in numerical simulation and experimental phantom scans. The method offers the flexibility to split measurements into desired dose levels while ensuring the reconstructed images share identical underlying features, thereby strengthening the model's robustness for input dose levels and capability of preserving fine details. In both numerical simulation and experimental phantom scans, we demonstrated that Noise2Noise with binomial selection outperforms other common self-supervised learning methods based on different presumptive conditions.
光子计数CT（PCCT）可获取光谱测量数据，并能够生成物质分解（MD）图像，这些图像在各种临床情况下都具有明显优势。然而，在MD图像中会观察到噪声放大现象，因此通常会进行去噪处理。在临床扫描中，干净或高质量的参考图像很少见，这使得监督学习（Noise2Clean）通常不切实际。Noise2Noise是一种自监督方法，它使用具有零均值、独立噪声的噪声图像和相应的噪声参考图像。PCCT分别对透射光子进行计数，并且假定原始测量数据在每个能量区间内遵循泊松分布，这为创建与噪声无关的图像对提供了可能性。该方法是使用二项式选择将计数分成两个具有独立噪声的低剂量扫描。我们通过噪声传播分析证明，重建的光谱图像在计数域中继承了噪声独立性，并在数值模拟和实验体模扫描中对其进行了验证。该方法提供了将测量数据分割成所需剂量水平的灵活性，同时确保重建图像共享相同的潜在特征，从而增强了模型对输入剂量水平的鲁棒性以及保留精细细节的能力。在数值模拟和实验体模扫描中，我们都证明了基于二项式选择的Noise2Noise优于基于不同假设条件的其他常见自监督学习方法。
