A Scheme for Quantum Teleportation and Remote Quantum State Preparation of IoT Multiple Devices.

With the continuous development of the Internet of Things (IoT) technology, the industry's awareness of the security of the IoT is also increasing, and the adoption of quantum communication technology can significantly improve the communication security of various devices in the IoT. This paper proposes a scheme of controlled remote quantum state preparation and quantum teleportation based on multiple communication parties, and a nine-qubit entanglement channel is used to achieve secure communication of multiple devices in the IoT. The channel preparation, measurement operation, and unitary operation of the scheme were successfully simulated on the IBM Quantum platform, and the entanglement degree and reliability of the channel were verified through 8192 shots. The scheme's application in the IoT was analyzed, and the steps and examples of the scheme in the secure communication of multiple devices in the IoT are discussed. By simulating two different attack modes, the effect of the attack on the communication scheme in the IoT was deduced, and the scheme's high security and anti-interference ability was analyzed. Compared with other schemes from the two aspects of principle and transmission efficiency, it is highlighted that the advantages of the proposed scheme are that it overcomes the single fixed one-way or two-way transmission protocol form of quantum teleportation in the past and can realize quantum communication with multiple devices, ensuring both security and transmission efficiency.

Hierarchical Controlled Hybrid Quantum Communication Based on Six-Qubit Entangled States in IoT.

The rapid development and extensive application of the Internet of Things (IoT) have brought new challenges and opportunities to the field of communication. By integrating quantum secure communication with the IoT, we can provide a higher level of security and privacy protection to counteract security threats in the IoT. In this paper, a hybrid quantum communication scheme using six-qubit entangled states as a channel is proposed for specific IoT application scenarios. This scheme achieves hierarchical control of communication protocols on a single quantum channel. In the proposed scheme, device A transmits data to device B through quantum teleportation, while device B issues control commands to device A through remote quantum state preparation technology. These two tasks are controlled by control nodes C and D, respectively. The transmission of information from device A to device B is a relatively less important task, which can be solely controlled by control node C. On the other hand, issuing control commands from device B to device A is a more crucial task requiring joint control from control nodes C and D. This paper describes the proposed scheme and conducts simulation experiments using IBM's Qiskit Aer quantum computing simulator. The results demonstrate that the fidelity of the quantum teleportation protocol (QTP) and the remote state preparation protocol (RSP) reach an impressive value of 0.999, fully validating the scheme's feasibility. Furthermore, the factors affecting the fidelity of the hybrid communication protocol in an IoT environment with specific quantum noise are analyzed. By combining the security of quantum communication with the application scenarios of the IoT, this paper presents a new possibility for IoT communication.

Unsupervised and Self-supervised Learning in Low-Dose Computed Tomography Denoising: Insights from Training Strategies.

In recent years, X-ray low-dose computed tomography (LDCT) has garnered widespread attention due to its significant reduction in the risk of patient radiation exposure. However, LDCT images often contain a substantial amount of noises, adversely affecting diagnostic quality. To mitigate this, a plethora of LDCT denoising methods have been proposed. Among them, deep learning (DL) approaches have emerged as the most effective, due to their robust feature extraction capabilities. Yet, the prevalent use of supervised training paradigms is often impractical due to the challenges in acquiring low-dose and normal-dose CT pairs in clinical settings. Consequently, unsupervised and self-supervised deep learning methods have been introduced for LDCT denoising, showing considerable potential for clinical applications. These methods' efficacy hinges on training strategies. Notably, there appears to be no comprehensive reviews of these strategies. Our review aims to address this gap, offering insights and guidance for researchers and practitioners. Based on training strategies, we categorize the LDCT methods into six groups: (i) cycle consistency-based, (ii) score matching-based, (iii) statistical characteristics of noise-based, (iv) similarity-based, (v) LDCT synthesis model-based, and (vi) hybrid methods. For each category, we delve into the theoretical underpinnings, training strategies, strengths, and limitations. In addition, we also summarize the open source codes of the reviewed methods. Finally, the review concludes with a discussion on open issues and future research directions.

Ruthenium single-atom doping-driven modulation of Co3O4 spinel tetrahedral site 3d-orbital occupancy in lithium-oxygen batteries.

Metal single-atom catalysts have attracted widespread attention in the field of lithium-oxygen batteries due to their unique active sites, high catalytic selectivity, and near total atomic utilization efficiency. Isolated metal atoms not only serve as the active sites themselves, but also function as modulators, reversely regulating the surface electronic structure of the support to enhance its inherent electrocatalytic activities. Despite the potential of isolated metal atom-driven active sites, understanding the structure-activity relationship remains a challenge. In this study, we present a ruthenium single-atom doping-driven cost-effective and durable tricobalt tetroxide electrocatalyst with excellent oxygen electrode electrocatalytic activity. The lithium-oxygen battery with this catalyst as the oxygen electrode demonstrates high performance, achieving a capacity of up to 25 000 mA h g-1 and maintaining good stability over 400 cycles at a current density of 100 mA g-1. This improvement is attributed to the exquisite control of the morphology and structure of the discharge product, lithium peroxide. The aresults of physical characterization and theoretical calculations reveal that isolated ruthenium atoms bond with the tetrahedral cobalt site, resulting in spin polarization enhancement and rearrangement of d orbital energy levels in cobalt. This rearrangement reduces the dz2 orbital occupancy and promotes their transfer to the octahedral cobalt site, thereby enhancing its adsorption capacity for the oxygen-containing intermediates, and ultimately increasing the electrocatalytic activity of the oxygen evolution reaction. This work presents an innovative strategy to regulate the catalytic activity of metal oxides by introducing another metal single atom.

Self-supervised deep learning for joint 3D low-dose PET/CT image denoising.

Deep learning (DL)-based denoising of low-dose positron emission tomography (LDPET) and low-dose computed tomography (LDCT) has been widely explored. However, previous methods have focused only on single modality denoising, neglecting the possibility of simultaneously denoising LDPET and LDCT using only one neural network, i.e., joint LDPET/LDCT denoising. Moreover, DL-based denoising methods generally require plenty of well-aligned LD-normal-dose (LD-ND) sample pairs, which can be difficult to obtain. To this end, we propose a self-supervised two-stage training framework named MAsk-then-Cycle (MAC), to achieve self-supervised joint LDPET/LDCT denoising. The first stage of MAC is masked autoencoder (MAE)-based pre-training and the second stage is self-supervised denoising training. Specifically, we propose a self-supervised denoising strategy named cycle self-recombination (CSR), which enables denoising without well-aligned sample pairs. Unlike other methods that treat noise as a homogeneous whole, CSR disentangles noise into signal-dependent and independent noises. This is more in line with the actual imaging process and allows for flexible recombination of noises and signals to generate new samples. These new samples contain implicit constraints that can improve the network's denoising ability. Based on these constraints, we design multiple loss functions to enable self-supervised training. Then we design a CSR-based denoising network to achieve joint 3D LDPET/LDCT denoising. Existing self-supervised methods generally lack pixel-level constraints on networks, which can easily lead to additional artifacts. Before denoising training, we perform MAE-based pre-training to indirectly impose pixel-level constraints on networks. Experiments on an LDPET/LDCT dataset demonstrate its superiority over existing methods. Our method is the first self-supervised joint LDPET/LDCT denoising method. It does not require any prior assumptions and is therefore more robust.

Multimodal Brain Image Fusion Based on Improved Rolling Guidance Filter and Wiener Filter.

Medical image fusion technology can integrate complementary information from different modality medical images to provide a more complete and accurate description of the specific diagnosed object, which is very helpful for image-guided clinical diagnosis and treatment. This paper proposes an effective brain image fusion framework based on improved rolling guidance filter (IRGF). Firstly, input images are decomposed into base layers and detail layers using the IRGF and Wiener filter. Secondly, the visual saliency maps of the input image are computed by pixel-level saliency value, and the weight maps of detail layers are constructed by max-absolute strategy and are further smoothed with Gaussian filter, the purpose of which is to make the fused image appear more naturally and more suitable for human visual perception. Lastly, base layers are fused by visual saliency map based fusion rule and the corresponding weight maps from detail layers are fused by the weighted least squares optimization scheme. Experimental results testify that our method is superior to some state-of-the-art methods in both subjective and objective assessments.

The relationship between cortical thickness and language comprehension varies with sex in healthy young adults: a large sample analysis.

Functional neuroimaging studies evaluating sex differences in language processing have been previously reported. However, it remains largely unclear whether there are structural bases for language comprehension and whether these are partially affected by sexual dimorphism in cortical thickness. To this end, we performed correlation analysis between cortical thickness and language comprehension in a large (N = 1017, 549 females, 468 males) young and healthy subjects from Human Connectome Project, with a specific focus on the impact of sex. We identified significant relationship between cortical thickness of the posterior cingulate cortex (PCC) and vocabulary comprehension in females (r = 0.318, r = 10%), while the association was significantly reduced in males (P = 0.017, Cohen's q = 0.154). Furthermore, thickness difference in the PCC was observed to be smaller in females (P < 0.0001, t = -7.12, Cohen's d = 0.45); however, the difference disappeared when controlling for brain size (Cohen's d = 0.002). Our findings indicated that variability in cortical thickness may affect cognitive function much more in females than in males, and highlighted the importance of brain size in explaining sex-specific cortical thickness.