Positron emission tomography (PET) is an advanced medical imaging technique widely used in various clinical applications, such as tumor detection and neurologic disorders. Reducing the radiotracer dose is desirable in PET imaging because it decreases the patient’s radiation exposure. However, reducing the dose can also increase noise, affecting the image quality. Therefore, an advanced image reconstruction algorithm based on low-dose PET data is needed to improve the quality of the reconstructed image. In this study, we propose the use of a direct PET image reconstruction network (DPIR-Net) using an improved Wasserstein generative adversarial network (WGAN) framework to enhance image quality. This study provides two main findings. First, our network uses sinogram data as input and outputs high-quality PET images direct, resulting in shorter reconstruction times compared with traditional model-based reconstruction networks. Second, we combine perceptual loss, mean square error, and the Wasserstein distance as the loss function, which effectively solves the problem of excessive smoothness and loss of detailed information in traditional network image reconstruction. We performed a comparative study using maximum-likelihood expectation-maximization (MLEM) with a post-Gaussian filter, a total variation (TV)-norm regularization, a nonlocal means (NLM) denoising method, a neural network denoising method, a traditional deep learning PET reconstruction network, and our proposed DPIR-Net method and evaluated the proposed method using both human and mouse data. The mouse data were obtained from a small-animal PET prototype system developed by our laboratory. The quantitative and qualitative results show that our proposed method outperformed the conventional methods.

In this paper, a single-computed tomography (CT) image super-resolution (SR) reconstruction scheme is proposed. This SR reconstruction scheme is based on sparse representation theory and dictionary learning of low- and high-resolution image patch pairs to improve the poor quality of low-resolution CT images obtained in clinical practice using low-dose CT technology. The proposed strategy is based on the idea that image patches can be well represented by sparse coding of elements from an overcomplete dictionary. To obtain similarity of the sparse representations, two dictionaries of low- and high-resolution image patches are jointly trained. Then, sparse representation coefficients extracted from the low-resolution input patches are used to reconstruct the high-resolution output. Sparse representation is used such that the trained dictionary pair can reduce computational costs. Combined with several appropriate iteration operations, the reconstructed high-resolution image can attain better image quality. The effectiveness of the proposed method is demonstrated using both clinical CT data and simulation image data. Image quality evaluation indexes (root mean squared error (RMSE) and peak signal-to-noise ratio (PSNR)) indicate that the proposed method can effectively improve the resolution of a single CT image.

Although lowering X-ray radiation helps reduce the risk of cancer in patients, low-dose computed tomography (LDCT) technology usually leads to poor image quality, such as amplified mottle noise and streak artifacts, which severely impact the diagnostic results. To improve diagnostic performance, we propose an algorithm based on a cycle-consistent generative adversarial network (CycleGAN) to suppress noise and reduce artifacts. In addition, we include attention mechanisms in the proposed network to expand the receptive field and capture richer contextual dependencies. Unlike traditional methods that manually match similar local blocks, our proposed method can autonomously learn the relationship between local features and their global dependencies. Specifically, two different types of attention modules (criss-cross self-attention (CCSA) and channel attention (CA)) are adopted to enhance feature interdependencies in the spatial and channel dimensions separately. Because of the self-attention mechanism, noise and artifacts can be restored using cues from all the local feature locations. Moreover, the CA mechanism adaptively reassigns the weights of each feature map. Extensive experiments show that our proposed method achieves better metrics and visual effects than state-of-the-art methods.

Lead halide perovskites have recently emerged as promising X/γ-ray scintillators. However, the small Stokes shift of exciton luminescence in perovskite scintillators creates problems for the light extraction efficiency and severely impedes their applications in hard X/γ-ray detection. Dopants have been used to shift the emission wavelength, but the radioluminescence lifetime has also been unwantedly extended. Herein, we demonstrate the intrinsic strain in 2D perovskite crystals as a general phenomenon, which could be utilized as self-wavelength shifting to reduce the self-absorption effect without sacrificing the radiation response speed. Furthermore, we successfully demonstrated the first imaging reconstruction by perovskites for application of positron emission tomography. The coincidence time resolution for the optimized perovskite single crystals (4 × 4 × 0.8 mm³) reached 119 ± 3 ps. This work provides a new paradigm for suppressing the self-absorption effect in scintillators and may facilitate the application of perovskite scintillators in practical hard X/γ-ray detections.

In this work, a small animal PET scanner named SIAT aPET was developed using dual-ended readout depth encoding detectors to simultaneously achieve high spatial resolution and high sensitivity. The scanner consists of four detector rings with 12 detector modules per ring; the ring diameter is 111 mm and the axial field of view (FOV) is 105.6 mm. The images are reconstructed using an ordered subset expectation maximization (OSEM) algorithm. The spatial resolution of the scanner was measured by using a ²² Na point source at the center axial FOV with different radial offsets. The sensitivity of the scanner was measured at center axis of the scanner with different axial positions. The count rate performance of the system was evaluated by scanning mouse-sized and rat-sized phantoms. An ultra-micro hot-rods phantom and two mice injected with ¹⁸ F-NaF and ¹⁸ F-FDG were scanned on the scanner. An average depth of interaction (DOI) resolution of 1.96 mm, energy resolution of 19.1% and timing resolution of 1.20 ns were obtained for the detector. Average spatial resolutions of 0.82 mm and 1.16 mm were obtained up to a distance of 30 mm radially from the center of the FOV when reconstructing a point source in 1% and 10% warm backgrounds, respectively, using OSEM reconstruction with 16 subsets and 10 iterations. Sensitivities of 16.0% and 11.9% were achieved at center of the scanner for energy windows of 250–750 keV and 350–750 keV respectively. Peak noise equivalent count rates (NECRs) of 324 kcps and 144 kcps were obtained at an activity of 26.4 MBq for the mouse-sized and rat-sized phantoms. Rods of 1.0 mm diameter can be visually resolved from the image of the ultra-micro hot-rods phantom. The capability of the scanner was demonstrated by high quality in-vivo mouse images.

The suppression of streak artifacts in computed tomography with a limited-angle configuration is challenging. Conventional analytical algorithms, such as filtered backprojection (FBP), are not successful due to incomplete projection data. Moreover, model-based iterative total variation algorithms effectively reduce small streaks but do not work well at eliminating large streaks. In contrast, FBP mapping networks and deep-learning-based postprocessing networks are outstanding at removing large streak artifacts; however, these methods perform processing in separate domains, and the advantages of multiple deep learning algorithms operating in different domains have not been simultaneously explored. In this paper, we present a hybrid-domain convolutional neural network (hdNet) for the reduction of streak artifacts in limited-angle computed tomography. The network consists of three components: the first component is a convolutional neural network operating in the sinogram domain, the second is a domain transformation operation, and the last is a convolutional neural network operating in the CT image domain. After training the network, we can obtain artifact-suppressed CT images directly from the sinogram domain. Verification results based on numerical, experimental and clinical data confirm that the proposed method can significantly reduce serious artifacts.