Deep Variational Networks with Exponential Weighting for Learning Computed Tomography

Spatial and channel attention-based conditional Wasserstein GAN for direct and rapid image reconstruction in ultrasound computed tomography

Ultrasound computed tomography (USCT) is an emerging technology that offers a noninvasive and radiation-free imaging approach with high sensitivity, making it promising for the early detection and diagnosis of breast cancer. The speed-of-sound (SOS) parameter plays a crucial role in distinguishing between benign masses and breast cancer. However, traditional SOS reconstruction methods face challenges in achieving a balance between resolution and computational efficiency, which hinders their clinical applications due to high computational complexity and long reconstruction times. In this paper, we propose a novel and efficient approach for direct SOS image reconstruction based on an improved conditional generative adversarial network. The generator directly reconstructs SOS images from time-of-flight information, eliminating the need for intermediate steps. Residual spatial-channel attention blocks are integrated into the generator to adaptively determine the relevance of arrival time from the transducer pair corresponding to each pixel in the SOS image. An ablation study verified the effectiveness of this module. Qualitative and quantitative evaluation results on breast phantom datasets demonstrate that this method is capable of rapidly reconstructing high-quality SOS images, achieving better generation results and image quality. Therefore, we believe that the proposed algorithm represents a new direction in the research area of USCT SOS reconstruction.

Deep Learning Ultrasound Computed Tomography Under Sparse Sampling

Ultrasound computed tomography (USCT) is a fast-emerging imaging modality that is expected to help detect and characterize breast tumors by quantifying the distribution of the speed of sound (SOS) and acoustic attenuation in breast tissue. High-quality quantitative SOS reconstruction in USCT requires a large number of transducers, which incurs high system costs and slow computation. In contrast, sparsely distributed arrays are low-cost and fast but significantly degrade image quality. Thus, we propose a framework to achieve high-quality SOS reconstruction under sparse sampling based on a convolutional neural network (SRSS-Net) with faster computation. We first apply the bent-ray algorithm to sparsely sampled data and then apply the SRSS-Net to efficiently improve the image quality. Experimental results on synthetic and real datasets demonstrate that the proposed SRSS-Net provides reconstructions that are superior to those of state-of-the-art methods in terms of artifact suppression, structural preservation, quantitative restoration, and computational speed. As demonstrated in our experiments, the fine-tuning training strategy is suggested when applying SRSS-Net to real-world circumstances. The imaging and computational performance of SRSS-Net on the inhomogeneous breast phantom further demonstrates that SRSS-Net has great potential in real-time breast cancer detection.

Sam's Net: A Self-Augmented Multistage Deep-Learning Network for End-to-End Reconstruction of Limited Angle CT

Limited angle reconstruction is a typical ill-posed problem in computed tomography (CT). Given incomplete projection data, images reconstructed by conventional analytical algorithms and iterative methods suffer from severe structural distortions and artifacts. In this paper, we proposed a self-augmented multi-stage deep-learning network (Sam's Net) for end-to-end reconstruction of limited angle CT. With the merit of the alternating minimization technique, Sam's Net integrates multi-stage self-constraints into cross-domain optimization to provide additional constraints on the manifold of neural networks. In practice, a sinogram completion network (SCNet) and artifact suppression network (ASNet), together with domain transformation layers constitute the backbone for cross-domain optimization. An online self-augmentation module was designed following the manner defined by alternating minimization, which enables a self-augmented learning procedure and multi-stage inference manner. Besides, a substitution operation was applied as a hard constraint for the solution space based on the data fidelity and a learnable weighting layer was constructed for data consistency refinement. Sam's Net forms a new framework for ill-posed reconstruction problems. In the training phase, the self-augmented procedure guides the optimization into a tightened solution space with enriched diverse data distribution and enhanced data consistency. In the inference phase, multi-stage prediction can improve performance progressively. Extensive experiments with both simulated and practical projections under 90-degree and 120-degree fan-beam configurations validate that Sam's Net can significantly improve the reconstruction quality with high stability and robustness.

MAGIC: Manifold and Graph Integrative Convolutional Network for Low-Dose CT Reconstruction

Low-dose computed tomography (LDCT) scans, which can effectively alleviate the radiation problem, will degrade the imaging quality. In this paper, we propose a novel LDCT reconstruction network that unrolls the iterative scheme and performs in both image and manifold spaces. Because patch manifolds of medical images have low-dimensional structures, we can build graphs from the manifolds. Then, we simultaneously leverage the spatial convolution to extract the local pixel-level features from the images and incorporate the graph convolution to analyze the nonlocal topological features in manifold space. The experiments show that our proposed method outperforms both the quantitative and qualitative aspects of state-of-the-art methods. In addition, aided by a projection loss component, our proposed method also demonstrates superior performance for semi-supervised learning. The network can remove most noise while maintaining the details of only 10% (40 slices) of the training data labeled.

Data Extrapolation From Learned Prior Images for Truncation Correction in Computed Tomography

Data truncation is a common problem in computed tomography (CT). Truncation causes cupping artifacts inside the field-of-view (FOV) and anatomical structures missing outside the FOV. Deep learning has achieved impressive results in CT reconstruction from limited data. However, its robustness is still a concern for clinical applications. Although the image quality of learning-based compensation schemes may be inadequate for clinical diagnosis, they can provide prior information for more accurate extrapolation than conventional heuristic extrapolation methods. With extrapolated projection, a conventional image reconstruction algorithm can be applied to obtain a final reconstruction. In this work, a general plug-and-play (PnP) method for truncation correction is proposed based on this idea, where various deep learning methods and conventional reconstruction algorithms can be plugged in. Such a PnP method integrates data consistency for measured data and learned prior image information for truncated data. This shows to have better robustness and interpretability than deep learning only. To demonstrate the efficacy of the proposed PnP method, two state-of-the-art deep learning methods, FBPConvNet and Pix2pixGAN, are investigated for truncation correction in cone-beam CT in noise-free and noisy cases. Their robustness is evaluated by showing false negative and false positive lesion cases. With our proposed PnP method, false lesion structures are corrected for both deep learning methods. For FBPConvNet, the root-mean-square error (RMSE) inside the FOV can be improved from 92HU to around 30HU by PnP in the noisy case. Pix2pixGAN solely achieves better image quality than FBPConvNet solely for truncation correction in general. PnP further improves the RMSE inside the FOV from 42HU to around 27HU for Pix2pixGAN. The efficacy of PnP is also demonstrated on real clinical head data.

Speed-of-sound imaging using diverging waves

PURPOSE

Due to its safe, low-cost, portable, and real-time nature, ultrasound is a prominent imaging method in computer-assisted interventions. However, typical B-mode ultrasound images have limited contrast and tissue differentiation capability for several clinical applications.

METHODS

Recent introduction of imaging speed-of-sound (SoS) in soft tissues using conventional ultrasound systems and transducers has great potential in clinical translation providing additional imaging contrast, e.g., in intervention planning, navigation, and guidance applications. However, current pulse-echo SoS imaging methods relying on plane wave (PW) sequences are highly prone to aberration effects, therefore suboptimal in image quality. In this paper we propose using diverging waves (DW) for SoS imaging and study this comparatively to PW.

RESULTS

We demonstrate wavefront aberration and its effects on the key step of displacement tracking in the SoS reconstruction pipeline, comparatively between PW and DW on a synthetic example. We then present the parameterization sensitivity of both approaches on a set of simulated phantoms. Analyzing SoS imaging performance comparatively indicates that using DW instead of PW, the reconstruction accuracy improves by over 20% in root-mean-square-error (RMSE) and by 42% in contrast-to-noise ratio (CNR). We then demonstrate SoS reconstructions with actual US acquisitions of a breast phantom. With our proposed DW, CNR for a high contrast tumor-representative inclusion is improved by 42%, while for a low contrast cyst-representative inclusion a 2.8-fold improvement is achieved.

CONCLUSION

SoS imaging, so far only studied using a plane wave transmission scheme, can be made more reliable and accurate using DW. The high imaging contrast of DW-based SoS imaging will thus facilitate the clinical translation of the method and utilization in computer-assisted interventions such as ultrasound-guided biopsies, where B-Mode contrast is often to low to detect potential lesions.

Training Variational Networks With Multidomain Simulations: Speed-of-Sound Image Reconstruction

Speed-of-sound (SoS) has been shown as a potential biomarker for breast cancer imaging, successfully differentiating malignant tumors from benign ones. SoS images can be reconstructed from time-of-flight measurements from ultrasound images acquired using conventional handheld ultrasound transducers. Variational networks (VNs) have recently been shown to be a potential learning-based approach for optimizing inverse problems in image reconstruction. Despite earlier promising results, these methods, however, do not generalize well from simulated to acquired data, due to the domain shift. In this work, we present for the first time a VN solution for a pulse-echo SoS image reconstruction problem using diverging waves with conventional transducers and single-sided tissue access. This is made possible by incorporating simulations with varying complexity into training. We use loop unrolling of gradient descent with momentum, with an exponentially weighted loss of outputs at each unrolled iteration in order to regularize the training. We learn norms as activation functions regularized to have smooth forms for robustness to input distribution variations. We evaluate reconstruction quality on the ray-based and full-wave simulations as well as on the tissue-mimicking phantom data, in comparison with a classical iterative [limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS)] optimization of this image reconstruction problem. We show that the proposed regularization techniques combined with multisource domain training yield substantial improvements in the domain adaptation capabilities of VN, reducing the median root mean squared error (RMSE) by 54% on a wave-based simulation data set compared to the baseline VN. We also show that on data acquired from a tissue-mimicking breast phantom, the proposed VN provides improved reconstruction in 12 ms.