Stage-by-Stage Wavelet Optimization Refinement Diffusion Model for Sparse-View CT Reconstruction

Multi-scale geometric transformer for sparse-view X-ray 3D foot reconstruction

BACKGROUND

Sparse-View X-ray 3D Foot Reconstruction aims to reconstruct the three-dimensional structure of the foot from sparse-view X-ray images, a challenging task due to data sparsity and limited viewpoints.

OBJECTIVE

This paper presents a novel method using a multi-scale geometric Transformer to enhance reconstruction accuracy and detail representation.

METHODS

Geometric position encoding technology and a window mechanism are introduced to divide X-ray images into local areas, finely capturing local features. A multi-scale Transformer module based on Neural Radiance Fields (NeRF) enhances the model's ability to express and capture details in complex structures. An adaptive weight learning strategy further optimizes the Transformer's feature extraction and long-range dependency modelling.

RESULTS

Experimental results demonstrate that the proposed method significantly improves the reconstruction accuracy and detail preservation of the foot structure under sparse-view X-ray conditions. The multi-scale geometric Transformer effectively captures local and global features, leading to more accurate and detailed 3D reconstructions.

CONCLUSIONS

The proposed method advances medical image reconstruction, significantly improving the accuracy and detail preservation of 3D foot reconstructions from sparse-view X-ray images.

DECT sparse reconstruction based on hybrid spectrum data generative diffusion model

PURPOSE

Dual-energy computed tomography (DECT) enables the differentiation of different materials. Additionally, DECT images consist of multiple scans of the same sample, revealing information similarity within the energy domain. To leverage this information similarity and address safety concerns related to excessive radiation exposure in DECT imaging, sparse view DECT imaging is proposed as a solution. However, this imaging method can impact image quality. Therefore, this paper presents a hybrid spectrum data generative diffusion reconstruction model (HSGDM) to improve imaging quality.

METHOD

To exploit the spectral similarity of DECT, we use interleaved angles for sparse scanning to obtain low- and high-energy CT images with complementary incomplete views. Furthermore, we organize low- and high-energy CT image views into multichannel forms for training and inference and promote information exchange between low-energy features and high-energy features, thus improving the reconstruction quality while reducing the radiation dose. In the HSGDM, we build two types of diffusion model constraint terms trained by the image space and wavelet space. The wavelet space diffusion model exploits mainly the orientation and scale features of artifacts. By integrating the image space diffusion model, we establish a hybrid constraint for the iterative reconstruction framework. Ultimately, we transform the iterative approach into a cohesive sampling process guided by the measurement data, which collaboratively produces high-quality and consistent reconstructions of sparse view DECT.

RESULTS

Compared with the comparison methods, this approach is competitive in terms of the precision of the CT values, the preservation of details, and the elimination of artifacts. In the reconstruction of 30 sparse views, with increases of 3.51 dB for the peak signal-to-noise ratio (PSNR), 0.03 for the structural similarity index measure (SSIM), and a reduction of 74.47 for the Fréchet inception distance (FID) score on the test dataset. In the ablation study, we determined the effectiveness of our proposed hybrid prior, consisting of the wavelet prior module and the image prior module, by comparing the visual effects and quantitative results of the methods using an image space model, a wavelet space model, and our hybrid model approach. Both qualitative and quantitative analyses of the results indicate that the proposed method performs well in sparse DECT reconstruction tasks.

CONCLUSION

We have developed a unified optimized mathematical model that integrates the image space and wavelet space prior knowledge into an iterative model. This model is more practical and interpretable than existing approaches are. The experimental results demonstrate the competitive performance of the proposed model.

Convergent-Diffusion Denoising Model for multi-scenario CT Image Reconstruction

A generic and versatile CT Image Reconstruction (CTIR) scheme can efficiently mitigate imaging noise resulting from inherent physical limitations, substantially bolstering the dependability of CT imaging diagnostics across a wider spectrum of patient cases. Current CTIR techniques often concentrate on distinct areas such as Low-Dose CT denoising (LDCTD), Sparse-View CT reconstruction (SVCTR), and Metal Artifact Reduction (MAR). Nevertheless, due to the intricate nature of multi-scenario CTIR, these techniques frequently narrow their focus to specific tasks, resulting in limited generalization capabilities for diverse scenarios. We propose a novel Convergent-Diffusion Denoising Model (CDDM) for multi-scenario CTIR, which utilizes a stepwise denoising process to converge toward an imaging-noise-free image with high generalization. CDDM uses a diffusion-based process based on a priori decay distribution to steadily correct imaging noise, thus avoiding the overfitting of individual samples. Within CDDM, a domain-correlated sampling network (DS-Net) provides an innovative sinogram-guided noise prediction scheme to leverage both image and sinogram (i.e., dual-domain) information. DS-Net analyzes the correlation of the dual-domain representations for sampling the noise distribution, introducing sinogram semantics to avoid secondary artifacts. Experimental results validate the practical applicability of our scheme across various CTIR scenarios, including LDCTD, MAR, and SVCTR, with the support of sinogram knowledge.

TD-STrans: Tri-domain sparse-view CT reconstruction based on sparse transformer

BACKGROUND AND OBJECTIVE

Sparse-view computed tomography (CT) speeds up scanning and reduces radiation exposure in medical diagnosis. However, when the projection views are severely under-sampled, deep learning-based reconstruction methods often suffer from over-smoothing of the reconstructed images due to the lack of high-frequency information. To address this issue, we introduce frequency domain information into the popular projection-image domain reconstruction, proposing a Tri-Domain sparse-view CT reconstruction model based on Sparse Transformer (TD-STrans).

METHODS

TD-STrans integrates three essential modules: the projection recovery module completes the sparse-view projection, the Fourier domain filling module mitigates artifacts and over-smoothing by filling in missing high-frequency details; the image refinement module further enhances and preserves image details. Additionally, a multi-domain joint loss function is designed to simultaneously enhance the reconstruction quality in the projection domain, image domain, and frequency domain, thereby further improving the preservation of image details.

RESULTS

The results of simulation experiments on the lymph node dataset and real experiments on the walnut dataset consistently demonstrate the effectiveness of TD-STrans in artifact removal, suppression of over-smoothing, and preservation of structural fidelity.

CONCLUSION

The reconstruction results of TD-STrans indicate that sparse transformer across multiple domains can alleviate over-smoothing and detail loss caused by reduced views, offering a novel solution for ultra-sparse-view CT imaging.

Deep Learning-Based Multi-View Projection Synthesis Approach for Improving the Quality of Sparse-View CBCT in Image-Guided Radiotherapy

While radiation hazards induced by cone-beam computed tomography (CBCT) in image-guided radiotherapy (IGRT) can be reduced by sparse-view sampling, the image quality is inevitably degraded. We propose a deep learning-based multi-view projection synthesis (DLMPS) approach to improve the quality of sparse-view low-dose CBCT images. In the proposed DLMPS approach, linear interpolation was first applied to sparse-view projections and the projections were rearranged into sinograms; these sinograms were processed with a sinogram restoration model and then rearranged back into projections. The sinogram restoration model was modified from the 2D U-Net by incorporating dynamic convolutional layers and residual learning techniques. The DLMPS approach was trained, validated, and tested on CBCT data from 163, 30, and 30 real patients respectively. Sparse-view projection datasets with 1/4 and 1/8 of the original sampling rate were simulated, and the corresponding full-view projection datasets were restored via the DLMPS approach. Tomographic images were reconstructed using the Feldkamp-Davis-Kress algorithm. Quantitative metrics including root-mean-square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and feature similarity (FSIM) were calculated in both the projection and image domains to evaluate the performance of the DLMPS approach. The DLMPS approach was compared with 11 state-of-the-art (SOTA) models, including CNN and Transformer architectures. For 1/4 sparse-view reconstruction task, the proposed DLMPS approach achieved averaged RMSE, PSNR, SSIM, and FSIM values of 0.0271, 45.93 dB, 0.9817, and 0.9587 in the projection domain, and 0.000885, 37.63 dB, 0.9074, and 0.9885 in the image domain, respectively. For 1/8 sparse-view reconstruction task, the DLMPS approach achieved averaged RMSE, PSNR, SSIM, and FSIM values of 0.0304, 44.85 dB, 0.9785, and 0.9524 in the projection domain, and 0.001057, 36.05 dB, 0.8786, and 0.9774 in the image domain, respectively. The DLMPS approach outperformed all the 11 SOTA models in both the projection and image domains for 1/4 and 1/8 sparse-view reconstruction tasks. The proposed DLMPS approach effectively improves the quality of sparse-view CBCT images in IGRT by accurately synthesizing missing projections, exhibiting potential in substantially reducing imaging dose to patients with minimal loss of image quality.

Linear diffusion noise boosted deep image prior for unsupervised sparse-view CT reconstruction

Objective.Deep learning has markedly enhanced the performance of sparse-view computed tomography reconstruction. However, the dependence of these methods on supervised training using high-quality paired datasets, and the necessity for retraining under varied physical acquisition conditions, constrain their generalizability across new imaging contexts and settings.Approach.To overcome these limitations, we propose an unsupervised approach grounded in the deep image prior framework. Our approach advances beyond the conventional single noise level input by incorporating multi-level linear diffusion noise, significantly mitigating the risk of overfitting. Furthermore, we embed non-local self-similarity as a deep implicit prior within a self-attention network structure, improving the model's capability to identify and utilize repetitive patterns throughout the image. Additionally, leveraging imaging physics, gradient backpropagation is performed between the image domain and projection data space to optimize network weights.Main Results.Evaluations with both simulated and clinical cases demonstrate our method's effective zero-shot adaptability across various projection views, highlighting its robustness and flexibility. Additionally, our approach effectively eliminates noise and streak artifacts while significantly restoring intricate image details.Significance. Our method aims to overcome the limitations in current supervised deep learning-based sparse-view CT reconstruction, offering improved generalizability and adaptability without the need for extensive paired training data.