Learning low‐dose CT degradation from unpaired data with flow‐based model

Abstract

Background There has been growing interest in low‐dose computed tomography (LDCT) for reducing the X‐ray radiation to patients. However, LDCT always suffers from complex noise in reconstructed images. Although deep learning‐based methods have shown their strong performance in LDCT denoising, most of them require a large number of paired training data of normal‐dose CT (NDCT) images and LDCT images, which are hard to acquire in the clinic. Lack of paired training data significantly undermines the practicability of supervised deep learning‐based methods. To alleviate this problem, unsupervised or weakly supervised deep learning‐based methods are required. Purpose We aimed to propose a method that achieves LDCT denoising without training pairs. Specifically, we first trained a neural network in a weakly supervised manner to simulate LDCT images from NDCT images. Then, simulated training pairs could be used for supervised deep denoising networks. Methods We proposed a weakly supervised method to learn the degradation of LDCT from unpaired LDCT and NDCT images. Concretely, LDCT and normal‐dose images were fed into one shared flow‐based model and projected to the latent space. Then, the degradation between low‐dose and normal‐dose images was modeled in the latent space. Finally, the model was trained by minimizing the negative log‐likelihood loss with no requirement of paired training data. After training, an NDCT image can be input to the trained flow‐based model to generate the corresponding LDCT image. The simulated image pairs of NDCT and LDCT can be further used to train supervised denoising neural networks for test. Results Our method achieved much better performance on LDCT image simulation compared with the most widely used image‐to‐image translation method, CycleGAN, according to the radial noise power spectrum. The simulated image pairs could be used for any supervised LDCT denoising neural networks. We validated the effectiveness of our generated image pairs on a classic convolutional neural network, REDCNN, and a novel transformer‐based model, TransCT. Our method achieved mean peak signal‐to‐noise ratio (PSNR) of 24.43dB, mean structural similarity (SSIM) of 0.785 on an abdomen CT dataset, mean PSNR of 33.88dB, mean SSIM of 0.797 on a chest CT dataset, which outperformed several traditional CT denoising methods, the same network trained by CycleGAN‐generated data, and a novel transfer learning method. Besides, our method was on par with the supervised networks in terms of visual effects. Conclusion We proposed a flow‐based method to learn LDCT degradation from only unpaired training data. It achieved impressive performance on LDCT synthesis. Next, we could train neural networks with the generated paired data for LDCT denoising. The denoising results are better than traditional and weakly supervised methods, comparable to supervised deep learning methods.

CT image denoising methods for image quality improvement and radiation dose reduction

“…Despite the limited or noisy labeling, the goal is to leverage this weaker supervision to learn meaningful patterns and representations from the data. 75,96 Different from weakly-supervised learning, although they both involve working with less-than-full supervision. The primary goal of semi-supervised learning is to leverage the available unlabeled data to improve the model's performance on tasks that require supervised learning.…”

“…The predominant DL models for CT denoising are GANs and CNNs. As shown in Figure 2a, out of 99 publications examined, 61 studies use the models based on CNN, 59–119 while 30 studies are based on GAN 120–149 . Additionally, two studies adopt Transformer‐based approaches 150,151 .…”

“…In weak supervision, the labels might be noisy, partial, or coarse, which makes the learning task more challenging. Despite the limited or noisy labeling, the goal is to leverage this weaker supervision to learn meaningful patterns and representations from the data 75,96 …”

Recent advances of Transformers in medical image analysis: A comprehensive review

“…Liu et al 188 proposed a weakly supervised method to learn the degradation of low‐dose CT from unpaired low‐dose and normal‐dose CT images. To be specific, low‐dose CT and normal‐dose images were fed into one shared flow‐based model and projected to the latent space.…”

Synthetic CT generation based on CBCT using improved vision transformer CycleGAN

“…Their DCLGAN 21 model further refines this concept by filtering strip artifacts, using inverse coordinate transformations to address ring artifacts, and generating corrected images by subtracting these artifacts from the original CT scans, striking a balance between artifact removal and detail preservation. Liu et al 22 introduced a weakly-supervised strategy that models the degradation between low-dose and normal-dose images in the latent space, focusing on generating high-quality low-dose images. Diffusion models, too, have garnered attention, with Li et al 23 frequency-guided diffusion model (FGDM) being a notable example.…”