Computed Tomography (CT) is a widely utilized imaging modality in clinical settings. Using densely acquired rotational X-ray arrays, CT can capture 3D spatial features. However, it is confronted with challenged such as significant time consumption and high radiation exposure. CT reconstruction methods based on sparse-view X-ray images have garnered substantial attention from researchers as they present a means to mitigate costs and risks. In recent years, diffusion models, particularly the Latent Diffusion Model (LDM), have demonstrated promising potential in the domain of 3D CT reconstruction. Nonetheless, due to the substantial differences between the 2D latent representation of X-ray modalities and the 3D latent representation of CT modalities, the vanilla LDM is incapable of achieving effective alignment within the latent space. To address this issue, we propose the Consistent Latent Space Diffusion Model (CLS-DM), which incorporates cross-modal feature contrastive learning to efficiently extract latent 3D information from 2D X-ray images and achieve latent space alignment between modalities. Experimental results indicate that CLS-DM outperforms classical and state-of-the-art generative models in terms of standard voxel-level metrics (PSNR, SSIM) on the LIDC-IDRI and CTSpine1K datasets. This methodology not only aids in enhancing the effectiveness and economic viability of sparse X-ray reconstructed CT but can also be generalized to other cross-modal transformation tasks, such as text-to-image synthesis. We have made our code publicly available at https://anonymous.4open.science/r/CLS-DM-50D6/ to facilitate further research and applications in other domains.

The pelvis is a high-incidence region for trauma, and its precise segmentation is vital for clinical diagnosis and treatment. Although deep learning has achieved high precision for pelvis segmentation, it is still confronted with challenges such as the scarcity of medical data and the high cost of annotation. Recently, generative models have offered a viable solution through synthetic data augmentation. Thus, we propose a novel text-conditioned generative framework that simultaneously produces high-fidelity CT images and their corresponding segmentation masks, with a special focus on accurately generating symmetric pelvic structures, including the left hip bone, right hip bone and sacrum. Firstly, we fine-tuned a medical text encoder to transform detailed descriptions into precise generation conditions. Considering the easily overlooked symmetry attribute in pelvic bone images, we introduced a novel coordinate-aware enhancement module that incorporates bone-specific centroid coordinates for symmetrical awareness. Finally, we expanded it to multi-task learning for generation of paired pelvic images and segmentation labels. To test our method, we released a pelvic CT image dataset with textual description (CT-PelvisText) and then transferred it to downstream segmentation using the generative pelvic image. Our experiments demonstrate the reliability of our method, which contributes to accurate pelvic segmentation. This work can also be easily extended to other medical images with symmetry properties, which provides potential for efficient learning in small-sample datasets. Our code and dataset are available at: https://github.com/CurellaSong/TSA_LDM.

Time-of-flight magnetic resonance angiography (TOF-MRA) is a common cerebrovascular imaging. Accurate and automatic cerebrovascular segmentation in TOF-MRA images is an important auxiliary method in clinical practice. Due to the complex semantics and noise interference, the existing segmentation methods often fail to pay attention to topological correlation, resulting in the neglect of branch vessels and vascular topology destruction. In this paper, we proposed a topology regularization adversarial model for cerebrovascular segmentation in TOF-MRA images. Firstly, we trained a self-supervised model to learn spatial semantic layout in TOF-MRA images by image context restoration. Subsequently, we exploited initialization based on the self-supervised model and constructed an adversarial model to accomplish parameter optimization. Considering the limitations of uneven distribution of cerebrovascular classes, we introduced skeleton structures as discriminative features to enhance vessel topological strength. We constructed some latest models to test our method over two datasets. Results show that the proposed model attains the highest score. Therefore, our method can obtain accurate connectivity information and higher graph similarity, leading more meaningful clinical utility.

With the rapid growth of mobile internet, Internet of Things (IoT), and cloud computing, the demand for data services has arisen sharply. As the core data service infrastructure, the number of data centers (DCs) has surged and led to higher energy consumption, which is not conducive to energy conservation, emission reduction, and sustainable development. In this paper, we proposed an energy-saving solution based on Artificial Intelligence (AI) and digital twin in DC scenarios, called Smart DC. The proposed solution can reduce DCs' energy consumption by optimizing air distribution and reducing cooling redundancy. Specifically, the digital twin model in this solution is used to verify and optimize AI strategies, and solve the problem of insufficient data in physical data center. Data for AI training and information mining is limited because the environment in the DCs change little. Moreover, in order to ensure that the DCs operate at a safe temperature, the adjustment of parameters should be conservative, so there is still room for cooling redundancy. We combined digital twin and AI, exploring the temperature rise boundary in the digital DCs and mine more data pairs, which has proven to increase the robustness of the AI model and achieve better energy-saving effect. The simulation and experiment results show that the proposed solution can ensure safe and efficient operation and keep the energy-saving rate of the cooling system to reach 41.07%.

Convolutional neural networks (CNNs) have been applied in numerous Internet of Things (IoT) devices for multifarious downstream tasks. However, with the increasing amount of data on edge devices, CNNs can hardly complete some tasks in time with limited computing and storage resources. Recently, filter pruning has been regarded as an effective technique to compress and accelerate CNNs, but existing methods rarely prune CNNs from the perspective of compressing high-dimensional tensors. In this article, we propose a novel theory to find redundant information in 3-D tensors, namely, quantified similarity between feature maps (QSFM), and utilize this theory to guide the filter pruning procedure. We perform QSFM on data sets (CIFAR-10, CIFAR-100, and ILSVRC-12) and edge devices and demonstrate that the proposed method can find the redundant information in the neural networks effectively with comparable compression and tolerable drop of accuracy. Without any fine-tuning operation, QSFM can compress ResNet-56 on CIFAR-10 significantly (48.7% FLOPs and 57.9% parameters are reduced) with only a loss of 0.54% in the top-1 accuracy. For the practical application of edge devices, QSFM can accelerate MobileNet-V2 inference speed by 1.53 times with only a loss of 1.23% in the ILSVRC-12 top-1 accuracy.

Terahertz (THz) radiation, which occupies a relatively unexplored portion of the electromagnetic spectrum between the mid-infrared and microwave bands, offers innovative sensing and imaging technologies that can provide information unavailable through conventional methods such as microwave and X-ray techniques. With the advancement of THz technologies, THz sensing and imaging will impact a broad range of interdisciplinary fields, including chemical and biological detections and identifications. In particular, THz radiation offers the opportunity for transformational advances in defense and security. Recent work shows that THz technologies are promising for the standoff detection and identification of explosive targets.

This paper introduces kernel partial least squares (K-PLS) for the identification of mixture content from terahertz spectra. Kernel partial least squares is a nonlinear extension of the partial least squares (PLS) method, commonly used in chemometrics. K-PLS and PLS are considered superior to peak matching methods for mixture spectra of multiple compounds because it avoids having to address the problem of overlapping peaks explicitly. Terahertz (THz) radiation is capable of transmitting easily through most dielectric materials and is used as a new tool to collect the original spectral readings from transmission, diffusion and reflection. A multi-output kernel partial least squares method is presented to model mixture composition based on pure substance training patterns, under the assumption of linear spectral mixture behavior. Preprocessing consists of a wavelet transform of the THz spectra and an independent component analysis (ICA) transform. Preliminary results show that the ICA+K-PLS approach is able to classify pure spectra accurately and allows for an accurate estimate of the composition from THz mixed spectra even where there are severe overlapped peaks in these spectra.