Search

Event-based semantic segmentation has gained popularity due to its capability to deal with scenarios under high-speed motion and extreme lighting conditions, which cannot be addressed by conventional RGB cameras. Since it is hard to annotate event data, previous approaches rely on event-to-image reconstruction to obtain pseudo labels for training. However, this will inevitably introduce noise, and learning from noisy pseudo labels, especially when generated from a single source, may reinforce the errors. This drawback is also called confirmation bias in pseudo-labeling. In this paper, we propose a novel hybrid pseudo-labeling framework for unsupervised event-based semantic segmentation, HPL-ESS, to alleviate the influence of noisy pseudo labels. In particular, we first employ a plain unsupervised domain adaptation framework as our baseline, which can generate a set of pseudo labels through self-training. Then, we incorporate offline event-to-image reconstruction into the framework, and obtain another set of pseudo labels by predicting segmentation maps on the reconstructed images. A noisy label learning strategy is designed to mix the two sets of pseudo labels and enhance the quality. Moreover, we propose a soft prototypical alignment module to further improve the consistency of target domain features. Extensive experiments show that our proposed method outperforms existing state-of-the-art methods by a large margin on the DSEC-Semantic dataset (+5.88% accuracy, +10.32% mIoU), which even surpasses several supervised methods.

The robot position speculation, which determines where the chassis should move, is one key step to control the mobile manipulators. The target position must ensure the feasibility of chassis movement and manipulability, which is guaranteed by randomized sampling and kinematic checking in traditional methods. Addressing the demands of agile robotics, this paper proposes a robot position speculation network(RPSN), a learning-based approach to enhance the agility of mobile manipulators. The RPSN incorporates a differentiable inverse kinematic algorithm and a neural network. Through end-to-end training, the RPSN can speculate positions with a high success rate. We apply the RPSN to mobile manipulators disassembling end-of-life electric vehicle batteries (EOL-EVBs). Extensive experiments on various simulated environments and physical mobile manipulators demonstrate that the probability of the initial position provided by RPSN being the ideal position is 96.67%. From the kinematic constraint perspective, it achieves 100% generation of the ideal position on average within 1.28 attempts. Much lower than that of random sampling, 31.04. Moreover, the proposed method demonstrates superior data efficiency over pure neural network approaches. The proposed RPSN enables the robot to quickly infer feasible target positions by intuition. This work moves towards building agile robots that can act swiftly like humans.

Transformer-based models are becoming deeper and larger recently. For better scalability, an underlying training solution in industry is to split billions of parameters (tensors) into many tasks and then run them across homogeneous accelerators (e.g., GPUs). However, such dedicated compute cluster is prohibitively expensive in academia and moderate companies. An economic replacement is to aggregate existing heterogeneous devices and share resources among multi-tenants. Nevertheless, static hardware configurations and dynamic resource contention definitely cause straggling tasks, which heavily slows down the overall training efficiency. Existing works feature contributions mainly tailored for traditional data parallelism. They cannot work well for the new tensor parallelism due to strict communication and correctness constraints. In this paper we first present ZERO-resizing, a novel dynamic workload balancing technique without any data migration. We tune workloads in real-time by temporarily resizing matrices involved in core tensor-related computations. We particularly design data imputation and priority selection policies to respectively satisfy consistency constraint required by normal training and reduce the accuracy loss. We also give a lightweight data migration technique without loss of accuracy, to cope with heavy heterogeneity. Our final SEMI-migration solution is built on top of these two techniques and can adaptively distinguish their respective balancing missions, to achieve an overall success in efficiency and accuracy. Extensive experiments on the representative Colossal-AI platform validate the effectiveness of our proposals., Comment: 13 pages

In the last decade, Convolutional Neural Network with a multi-layer architecture has advanced rapidly. However, training its complex network is very space-consuming, since a lot of intermediate data are preserved across layers, especially when processing high-dimension inputs with a big batch size. That poses great challenges to the limited memory capacity of current accelerators (e.g., GPUs). Existing efforts mitigate such bottleneck by external auxiliary solutions with additional hardware costs, and internal modifications with potential accuracy penalty. Differently, our analysis reveals that computations intra- and inter-layers exhibit the spatial-temporal weak dependency and even complete independency features. That inspires us to break the traditional layer-by-layer (column) dataflow rule. Now operations are novelly re-organized into rows throughout all convolution layers. This lightweight design allows a majority of intermediate data to be removed without any loss of accuracy. We particularly study the weak dependency between two consecutive rows. For the resulting skewed memory consumption, we give two solutions with different favorite scenarios. Evaluations on two representative networks confirm the effectiveness. We also validate that our middle dataflow optimization can be smoothly embraced by existing works for better memory reduction.

In this paper, we propose a deep learning based performance testing framework to minimize the number of required test modules while guaranteeing the accuracy requirement, where a test module corresponds to a combination of one circuit and one stimulus. First, we apply a deep neural network (DNN) to establish the mapping from the response of the circuit under test (CUT) in each module to all specifications to be tested. Then, the required test modules are selected by solving a 0-1 integer programming problem. Finally, the predictions from the selected test modules are combined by a DNN to form the specification estimations. The simulation results validate the proposed approach in terms of testing accuracy and cost.

Acquiring a multi-task imitation policy in 3D manipulation poses challenges in terms of scene understanding and action prediction. Current methods employ both 3D representation and multi-view 2D representation to predict the poses of the robot's end-effector. However, they still require a considerable amount of high-quality robot trajectories, and suffer from limited generalization in unseen tasks and inefficient execution in long-horizon reasoning. In this paper, we propose SAM-E, a novel architecture for robot manipulation by leveraging a vision-foundation model for generalizable scene understanding and sequence imitation for long-term action reasoning. Specifically, we adopt Segment Anything (SAM) pre-trained on a huge number of images and promptable masks as the foundation model for extracting task-relevant features, and employ parameter-efficient fine-tuning on robot data for a better understanding of embodied scenarios. To address long-horizon reasoning, we develop a novel multi-channel heatmap that enables the prediction of the action sequence in a single pass, notably enhancing execution efficiency. Experimental results from various instruction-following tasks demonstrate that SAM-E achieves superior performance with higher execution efficiency compared to the baselines, and also significantly improves generalization in few-shot adaptation to new tasks., Comment: ICML 2024. Project page: https://sam-embodied.github.io

The energy-efficient and brain-like information processing abilities of Spiking Neural Networks (SNNs) have attracted considerable attention, establishing them as a crucial element of brain-inspired computing. One prevalent challenge encountered by SNNs is the trade-off between inference speed and accuracy, which requires sufficient time to achieve the desired level of performance. Drawing inspiration from animal behavior experiments that demonstrate a connection between decision-making reaction times, task complexity, and confidence levels, this study seeks to apply these insights to SNNs. The focus is on understanding how SNNs make inferences, with a particular emphasis on untangling the interplay between signal and noise in decision-making processes. The proposed theoretical framework introduces a new optimization objective for SNN training, highlighting the importance of not only the accuracy of decisions but also the development of predictive confidence through learning from past experiences. Experimental results demonstrate that SNNs trained according to this framework exhibit improved confidence expression, leading to better decision-making outcomes. In addition, a strategy is introduced for efficient decision-making during inference, which allows SNNs to complete tasks more quickly and can use stopping times as indicators of decision confidence. By integrating neuroscience insights with neuromorphic computing, this study opens up new possibilities to explore the capabilities of SNNs and advance their application in complex decision-making scenarios., Comment: 27 pages, 6 figures, 3 tables

Large foundation models have recently emerged as a prominent focus of interest, attaining superior performance in widespread scenarios. Due to the scarcity of 3D data, many efforts have been made to adapt pre-trained transformers from vision to 3D domains. However, such 2D-to-3D approaches are still limited, due to the potential loss of spatial geometries and high computation cost. More importantly, their frameworks are mainly designed for 2D models, lacking a general any-to-3D paradigm. In this paper, we introduce Any2Point, a parameter-efficient method to empower any-modality large models (vision, language, audio) for 3D understanding. Given a frozen transformer from any source modality, we propose a 3D-to-any (1D or 2D) virtual projection strategy that correlates the input 3D points to the original 1D or 2D positions within the source modality. This mechanism enables us to assign each 3D token with a positional encoding paired with the pre-trained model, which avoids 3D geometry loss caused by the true projection and better motivates the transformer for 3D learning with 1D/2D positional priors. Then, within each transformer block, we insert an any-to-3D guided adapter module for parameter-efficient fine-tuning. The adapter incorporates prior spatial knowledge from the source modality to guide the local feature aggregation of 3D tokens, compelling the semantic adaption of any-modality transformers. We conduct extensive experiments to showcase the effectiveness and efficiency of our method. Code and models are released at https://github.com/Ivan-Tang-3D/Any2Point., Comment: Code and models are released at https://github.com/Ivan-Tang-3D/Any2Point

In order to study the dynamic response of hydraulic cylinders under complex load conditions, the modal characteristics of hydraulic cylinders were simulated and calculated by comprehensively considering the combined effects of preload and oil fluid dynamics. The contact in the finite element model was specially processed to ensure that the local degrees of freedom of the piston in the structure were maintained during the modal simulation calculation process. After the static characteristics were calculated, the relevant data was imported into the modal simulation module to achieve coupling analysis under preload. By verifying the contact stiffness factor, the boundary conditions of the model were effectively optimized. The natural frequencies of the model were compared and analyzed under different piston strokes, and the mechanism of dynamic response was obtained. Through modal testing, the results indicated that the simulation results had high accuracy and reliability.

Haoshaqiang Zhang,1,* Hua Shang,1,2,* Zhigang Wang,1 Kun Li1 1Department of Orthopedics Surgery, People’s Hospital of Xinjiang Uygur Autonomous Region, Urumqi, People’s Republic of China; 2Department of Human Resources, People’s Hospital of Xinjiang Uygur Autonomous Region, Urumqi People’s Republic of China*These authors contributed equally to this workCorrespondence: Zhigang Wang, Department of Orthopedics Surgery, People’s Hospital of Xinjiang Uygur Autonomous Region, No. 91, Tianchi Road, Tianshan District, Urumqi, 830001, People’s Republic of China, Tel +86-9918563146, Email guanjie2023yundong@163.comPurpose: To analyze the independent associations of miRNA-146a and miRNA-223 with rheumatoid arthritis (RA) and evaluate their predictive values for RA.Patients and Methods: A total of 68 RA patients were selected as cases, and meanwhile 68 patients with a traumatic knee condition were selected as controls by matching to the cases according to sex and age at the ratio of 1:1. The independent associations of miRNA-146a and miRNA-223 with RA were identified by binary logistic regression analysis. Receiver operating characteristic (ROC) curve was used to evaluate their predictive values for RA.Results: MiRNA-146a and miRNA-223 expression levels in both synovial tissues and serums were statistically higher in cases than in controls, and their expression levels in serums were not statistically different from those in synovial tissues in both cases and controls. The expression levels of miRNA-146a and miRNA-223 in synovial tissues were independently associated with RA, as well as the expression levels of miRNA-146a and miRNA-223 in serums. The area under curve (AUC) of combination of miRNA-146a and miRNA-223 in synovial tissues for the prediction of RA was 0.910 [95% confidence interval (CI): 0.863– 0.962], and the AUC of combination of miRNA-146a and miRNA-223 in serums was 0.904 (95% CI: 0.851– 0.957). Their difference was not sta

The highly irregular spiking activity of cortical neurons and behavioral variability suggest that the brain could operate in a fundamentally probabilistic way. Mimicking how the brain implements and learns probabilistic computation could be a key to developing machine intelligence that can think more like humans. In this work, we propose a theory of stochastic neural computing (SNC) in which streams of noisy inputs are transformed and processed through populations of nonlinearly coupled spiking neurons. To account for the propagation of correlated neural variability, we derive from first principles a moment embedding for spiking neural network (SNN). This leads to a new class of deep learning model called the moment neural network (MNN) which naturally generalizes rate-based neural networks to second order. As the MNN faithfully captures the stationary statistics of spiking neural activity, it can serve as a powerful proxy for training SNN with zero free parameters. Through joint manipulation of mean firing rate and noise correlations in a task-driven way, the model is able to learn inference tasks while simultaneously minimizing prediction uncertainty, resulting in enhanced inference speed. We further demonstrate the application of our method to Intel's Loihi neuromorphic hardware. The proposed theory of SNC may open up new opportunities for developing machine intelligence capable of computing uncertainty and for designing unconventional computing architectures.

Talking head generation aims to generate faces that maintain the identity information of the source image and imitate the motion of the driving image. Most pioneering methods rely primarily on 2D representations and thus will inevitably suffer from face distortion when large head rotations are encountered. Recent works instead employ explicit 3D structural representations or implicit neural rendering to improve performance under large pose changes. Nevertheless, the fidelity of identity and expression is not so desirable, especially for novel-view synthesis. In this paper, we propose HiDe-NeRF, which achieves high-fidelity and free-view talking-head synthesis. Drawing on the recently proposed Deformable Neural Radiance Fields, HiDe-NeRF represents the 3D dynamic scene into a canonical appearance field and an implicit deformation field, where the former comprises the canonical source face and the latter models the driving pose and expression. In particular, we improve fidelity from two aspects: (i) to enhance identity expressiveness, we design a generalized appearance module that leverages multi-scale volume features to preserve face shape and details; (ii) to improve expression preciseness, we propose a lightweight deformation module that explicitly decouples the pose and expression to enable precise expression modeling. Extensive experiments demonstrate that our proposed approach can generate better results than previous works. Project page: https://www.waytron.net/hidenerf, Comment: Accepted by CVPR 2023

This paper addresses the problem of rolling shutter correction in complex nonlinear and dynamic scenes with extreme occlusion. Existing methods suffer from two main drawbacks. Firstly, they face challenges in estimating the accurate correction field due to the uniform velocity assumption, leading to significant image correction errors under complex motion. Secondly, the drastic occlusion in dynamic scenes prevents current solutions from achieving better image quality because of the inherent difficulties in aligning and aggregating multiple frames. To tackle these challenges, we model the curvilinear trajectory of pixels analytically and propose a geometry-based Quadratic Rolling Shutter (QRS) motion solver, which precisely estimates the high-order correction field of individual pixels. Besides, to reconstruct high-quality occlusion frames in dynamic scenes, we present a 3D video architecture that effectively Aligns and Aggregates multi-frame context, namely, RSA2-Net. We evaluate our method across a broad range of cameras and video sequences, demonstrating its significant superiority. Specifically, our method surpasses the state-of-the-art by +4.98, +0.77, and +4.33 of PSNR on Carla-RS, Fastec-RS, and BS-RSC datasets, respectively. Code is available at https://github.com/DelinQu/qrsc., Comment: accepted at ICCV 2023

Understanding 3D scenes from multi-view inputs has been proven to alleviate the view discrepancy issue in 3D visual grounding. However, existing methods normally neglect the view cues embedded in the text modality and fail to weigh the relative importance of different views. In this paper, we propose ViewRefer, a multi-view framework for 3D visual grounding exploring how to grasp the view knowledge from both text and 3D modalities. For the text branch, ViewRefer leverages the diverse linguistic knowledge of large-scale language models, e.g., GPT, to expand a single grounding text to multiple geometry-consistent descriptions. Meanwhile, in the 3D modality, a transformer fusion module with inter-view attention is introduced to boost the interaction of objects across views. On top of that, we further present a set of learnable multi-view prototypes, which memorize scene-agnostic knowledge for different views, and enhance the framework from two perspectives: a view-guided attention module for more robust text features, and a view-guided scoring strategy during the final prediction. With our designed paradigm, ViewRefer achieves superior performance on three benchmarks and surpasses the second-best by +2.8%, +1.5%, and +1.35% on Sr3D, Nr3D, and ScanRefer. Code is released at https://github.com/Ivan-Tang-3D/ViewRefer3D., Comment: Accepted by ICCV 2023. Code is released at https://github.com/Ivan-Tang-3D/ViewRefer3D

Non-line-of-sight (NLOS) tracking has drawn increasing attention in recent years, due to its ability to detect object motion out of sight. Most previous works on NLOS tracking rely on active illumination, e.g., laser, and suffer from high cost and elaborate experimental conditions. Besides, these techniques are still far from practical application due to oversimplified settings. In contrast, we propose a purely passive method to track a person walking in an invisible room by only observing a relay wall, which is more in line with real application scenarios, e.g., security. To excavate imperceptible changes in videos of the relay wall, we introduce difference frames as an essential carrier of temporal-local motion messages. In addition, we propose PAC-Net, which consists of alternating propagation and calibration, making it capable of leveraging both dynamic and static messages on a frame-level granularity. To evaluate the proposed method, we build and publish the first dynamic passive NLOS tracking dataset, NLOS-Track, which fills the vacuum of realistic NLOS datasets. NLOS-Track contains thousands of NLOS video clips and corresponding trajectories. Both real-shot and synthetic data are included. Our codes and dataset are available at https://againstentropy.github.io/NLOS-Track/., Comment: CVPR 2023 camera-ready version. Codes and dataset are available at https://againstentropy.github.io/NLOS-Track

Depth-from-defocus (DFD), modeling the relationship between depth and defocus pattern in images, has demonstrated promising performance in depth estimation. Recently, several self-supervised works try to overcome the difficulties in acquiring accurate depth ground-truth. However, they depend on the all-in-focus (AIF) images, which cannot be captured in real-world scenarios. Such limitation discourages the applications of DFD methods. To tackle this issue, we propose a completely self-supervised framework that estimates depth purely from a sparse focal stack. We show that our framework circumvents the needs for the depth and AIF image ground-truth, and receives superior predictions, thus closing the gap between the theoretical success of DFD works and their applications in the real world. In particular, we propose (i) a more realistic setting for DFD tasks, where no depth or AIF image ground-truth is available; (ii) a novel self-supervision framework that provides reliable predictions of depth and AIF image under the challenging setting. The proposed framework uses a neural model to predict the depth and AIF image, and utilizes an optical model to validate and refine the prediction. We verify our framework on three benchmark datasets with rendered focal stacks and real focal stacks. Qualitative and quantitative evaluations show that our method provides a strong baseline for self-supervised DFD tasks., Comment: CVPR 2023 camera-ready version. The code is released at https://github.com/Ehzoahis/DEReD

This paper presents "AI Nüshu," an emerging language system inspired by Nüshu (women's scripts), the unique language created and used exclusively by ancient Chinese women who were illiterate under a patriarchy society. Through an interactive art installation, two artificial intelligent (AI) agents continuously observe their environment and communicate with each other, developing a writing system that encodes Chinese. In this system, two AI agents observe the environment through cameras, record the unconscious behaviors of the audience, and generate summaries of their observations through visual recognition. Subsequently, the agent associates the corresponding original Nüshu poetry lines and generates new poetry text through a Language Model (LLM), representing its reflection. To develop their language, they continuously switch roles between the speaker and listener, constantly communicating their reflections, and encrypting a word in the poetry line with their self-created AI Nüshu character, allowing the other to guess and learn. Gradually, they reach a consensus on AI Nüshu, forming a unique "AI Nüshu Dictionary" for machines. This language, algorithmically combined into corresponding characters, has components derived from Nüshu, similar to Chinese characters and traditional textile patterns. Thus, like ancient women, the two agents gradually developed their Chinese writing system, corresponding one-to-one with Chinese characters. In contrast, humans, as the authority of the language system, became an object observed, interpreted, and inspired by machines to stimulate non-human language. This is the first media art project to interpret Nüshu from a computational linguistics perspective, infusing AI and art research with non-English natural language processing, Chinese cultural heritage, and a feminist viewpoint. This encourages the creation of more non-English, linguistically-oriented artworks for diverse cultures. We simulate communication in sisterhood through

This paper presents “AI Nüshu," an emerging language system inspired by Nüshu (women’s scripts), the unique language created and used exclusively by ancient Chinese women who were thought to be illiterate under a patriarchal society. In this interactive installation, two artificial intelligence (AI) agents are trained in the Chinese dictionary and the Nüshu corpus. By continually observing their environment and communicating, these agents collaborate towards creating a standard writing system to encode Chinese. It offers an artistic interpretation of the creation of a non-western script from a computational linguistics perspective, integrating AI technology with Chinese cultural heritage and a feminist viewpoint.

The field of 4D point cloud understanding is rapidly developing with the goal of analyzing dynamic 3D point cloud sequences. However, it remains a challenging task due to the sparsity and lack of texture in point clouds. Moreover, the irregularity of point cloud poses a difficulty in aligning temporal information within video sequences. To address these issues, we propose a novel cross-modal knowledge transfer framework, called X4D-SceneFormer. This framework enhances 4D-Scene understanding by transferring texture priors from RGB sequences using a Transformer architecture with temporal relationship mining. Specifically, the framework is designed with a dual-branch architecture, consisting of an 4D point cloud transformer and a Gradient-aware Image Transformer (GIT). During training, we employ multiple knowledge transfer techniques, including temporal consistency losses and masked self-attention, to strengthen the knowledge transfer between modalities. This leads to enhanced performance during inference using single-modal 4D point cloud inputs. Extensive experiments demonstrate the superior performance of our framework on various 4D point cloud video understanding tasks, including action recognition, action segmentation and semantic segmentation. The results achieve 1st places, i.e., 85.3% (+7.9%) accuracy and 47.3% (+5.0%) mIoU for 4D action segmentation and semantic segmentation, on the HOI4D challenge\footnote{\url{http://www.hoi4d.top/}.}, outperforming previous state-of-the-art by a large margin. We release the code at https://github.com/jinglinglingling/X4D

Quantitative phase imaging (QPI) through multi-core fibers (MCFs) has been an emerging in vivo label-free endoscopic imaging modality with minimal invasiveness. However, the computational demands of conventional iterative phase retrieval algorithms have limited their real-time imaging potential. We demonstrate a learning-based MCF phase imaging method, that significantly reduced the phase reconstruction time to 5.5 ms, enabling video-rate imaging at 181 fps. Moreover, we introduce an innovative optical system that automatically generated the first open-source dataset tailored for MCF phase imaging, comprising 50,176 paired speckle and phase images. Our trained deep neural network (DNN) demonstrates robust phase reconstruction performance in experiments with a mean fidelity of up to 99.8\%. Such an efficient fiber phase imaging approach can broaden the applications of QPI in hard-to-reach areas., Comment: 5 pages. 5 figures

In this paper, we introduce \textbf{GS-SLAM} that first utilizes 3D Gaussian representation in the Simultaneous Localization and Mapping (SLAM) system. It facilitates a better balance between efficiency and accuracy. Compared to recent SLAM methods employing neural implicit representations, our method utilizes a real-time differentiable splatting rendering pipeline that offers significant speedup to map optimization and RGB-D rendering. Specifically, we propose an adaptive expansion strategy that adds new or deletes noisy 3D Gaussians in order to efficiently reconstruct new observed scene geometry and improve the mapping of previously observed areas. This strategy is essential to extend 3D Gaussian representation to reconstruct the whole scene rather than synthesize a static object in existing methods. Moreover, in the pose tracking process, an effective coarse-to-fine technique is designed to select reliable 3D Gaussian representations to optimize camera pose, resulting in runtime reduction and robust estimation. Our method achieves competitive performance compared with existing state-of-the-art real-time methods on the Replica, TUM-RGBD datasets. Project page: https://gs-slam.github.io/., Comment: Accepted to CVPR 2024(highlight). Project Page: https://gs-slam.github.io

Generalizable articulated object manipulation is essential for home-assistant robots. Recent efforts focus on imitation learning from demonstrations or reinforcement learning in simulation, however, due to the prohibitive costs of real-world data collection and precise object simulation, it still remains challenging for these works to achieve broad adaptability across diverse articulated objects. Recently, many works have tried to utilize the strong in-context learning ability of Large Language Models (LLMs) to achieve generalizable robotic manipulation, but most of these researches focus on high-level task planning, sidelining low-level robotic control. In this work, building on the idea that the kinematic structure of the object determines how we can manipulate it, we propose a kinematic-aware prompting framework that prompts LLMs with kinematic knowledge of objects to generate low-level motion trajectory waypoints, supporting various object manipulation. To effectively prompt LLMs with the kinematic structure of different objects, we design a unified kinematic knowledge parser, which represents various articulated objects as a unified textual description containing kinematic joints and contact location. Building upon this unified description, a kinematic-aware planner model is proposed to generate precise 3D manipulation waypoints via a designed kinematic-aware chain-of-thoughts prompting method. Our evaluation spanned 48 instances across 16 distinct categories, revealing that our framework not only outperforms traditional methods on 8 seen categories but also shows a powerful zero-shot capability for 8 unseen articulated object categories. Moreover, the real-world experiments on 7 different object categories prove our framework's adaptability in practical scenarios. Code is released at https://github.com/GeWu-Lab/LLM_articulated_object_manipulation/tree/main., Comment: Accepted by ICRA 2024

The popularity of pre-trained large models has revolutionized downstream tasks across diverse fields, such as language, vision, and multi-modality. To minimize the adaption cost for downstream tasks, many Parameter-Efficient Fine-Tuning (PEFT) techniques are proposed for language and 2D image pre-trained models. However, the specialized PEFT method for 3D pre-trained models is still under-explored. To this end, we introduce Point-PEFT, a novel framework for adapting point cloud pre-trained models with minimal learnable parameters. Specifically, for a pre-trained 3D model, we freeze most of its parameters, and only tune the newly added PEFT modules on downstream tasks, which consist of a Point-prior Prompt and a Geometry-aware Adapter. The Point-prior Prompt adopts a set of learnable prompt tokens, for which we propose to construct a memory bank with domain-specific knowledge, and utilize a parameter-free attention to enhance the prompt tokens. The Geometry-aware Adapter aims to aggregate point cloud features within spatial neighborhoods to capture fine-grained geometric information through local interactions. Extensive experiments indicate that our Point-PEFT can achieve better performance than the full fine-tuning on various downstream tasks, while using only 5% of the trainable parameters, demonstrating the efficiency and effectiveness of our approach. Code is released at https://github.com/Ivan-Tang-3D/Point-PEFT., Comment: The specialized PEFT framework for 3D pre-trained models, which achieves competitive performance to full fine-tuning, and significantly reduces the computational resources. Project page: https://github.com/Ivan-Tang-3D/Point-PEFT

Grasping occluded objects in cluttered environments is an essential component in complex robotic manipulation tasks. In this paper, we introduce an AffordanCE-driven Next-Best-View planning policy (ACE-NBV) that tries to find a feasible grasp for target object via continuously observing scenes from new viewpoints. This policy is motivated by the observation that the grasp affordances of an occluded object can be better-measured under the view when the view-direction are the same as the grasp view. Specifically, our method leverages the paradigm of novel view imagery to predict the grasps affordances under previously unobserved view, and select next observation view based on the highest imagined grasp quality of the target object. The experimental results in simulation and on a real robot demonstrate the effectiveness of the proposed affordance-driven next-best-view planning policy. Project page: https://sszxc.net/ace-nbv/., Comment: Conference on Robot Learning (CoRL) 2023

Recently, multiple data-driven models based on machine learning for weather forecasting have emerged. These models are highly competitive in terms of accuracy compared to traditional numerical weather prediction (NWP) systems. In particular, the Pangu-Weather model, which is open source for non-commercial use, has been validated for its forecasting performance by the European Centre for Medium-Range Weather Forecasts (ECMWF) and has recently been published in the journal "Nature". In this paper, we evaluate the compatibility of the Pangu-Weather model with several commonly used NWP operational analyses through case studies. The results indicate that the Pangu-Weather model is compatible with different operational analyses from various NWP systems as the model initial conditions, and it exhibits a relatively stable forecasting capability. Furthermore, we have verified that improving the quality of global or local initial conditions significantly contributes to enhancing the forecasting performance of the Pangu-Weather model.

Hui Gao,1,2,* Zhaoxia Wang,1,2,* Mixiao Tan,1 Weiwei Liu,1 Liang Zhang,1 Ju Huang,1 Yang Cao,1 Pan Li,1 Zhigang Wang,1 Jiexin Wen,2 Tingting Shang,1 Haitao Ran1 1Chongqing Key Laboratory of Ultrasound Molecular Imaging & Institute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 2Department of Ultrasound, Children’s Hospital of Chongqing Medical University, Chongqing, People’s Republic of China*These authors contributed equally to this workCorrespondence: Haitao RanChongqing Key Laboratory of Ultrasound Molecular Imaging & Institute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, No. 76 Linjiang Road, Yuzhong District, Chongqing, 400010, People’s Republic of ChinaTel +86-13-512373563Email ranhaitao@cqmu.edu.cnBackground: Therapeutic ultrasound (US) has been extensively explored for its inherent high tissue-penetrating capability and on-demand irradiation without radioactive damage. Although high-intensity focused ultrasound (HIFU) is evolved as such an outstanding US-based approach, its insufficient therapeutic effect and the high-intensity induced potential damage to surrounding normal tissues hindered its development towards practical application. As opposed to high intensity ultrasound, sonodynamic therapy (SDT) is a low intensity US-based method which exhibits certain therapeutic effects against cancer via sonosensitizers-generated reactive oxygen species (ROS) overproduction.Methods: Hematoporphyrin monomethyl ether (HMME) loaded CaCO3 nanoparticles (designated as Ca@H) were synthesized by a gas diffusion method. The pH-responsive performance, in vitro SDT, ex vivo HIFU therapy (HIFUT), photoacoustic (PA) imaging and in vivo HIFUT combined with SDT were investigated thoroughly.Results: Ca@H NPs gradually decomposed in acid tumor microenvironment, produced CO2 and released HMME. Both CO2 and HMME enhanced pho

Li M, Luo H, Zhang W, et al. Int J Nanomedicine. 2018;13:3907–3920. The Editor and Publisher of International Journal of Nanomedicine wish to retract the published article. Concerns were raised regarding the alleged duplication of images in Figure 6. Specifically, Figure 6A, Non-targeted group, appears to have been duplicated with Figure 6A, Antagonized group. The authors responded to our queries and provided some of the original data relating to Figure 6. However, the Editor determined that current and previous concerns relating to this article have raised doubts about the validity of the findings and requested for the article to be retracted. The authors have been informed of this decision. Our decision-making was informed by our policy on publishing ethics and integrity and the COPE guidelines on retraction. The retracted article will remain online to maintain the scholarly record, but it will be digitally watermarked on each page as “Retracted”. This retraction relates to this paper

Aims: In the UNITI endoscopy sub-study, only 17.4% of patients with Crohn's disease (CD) on ustekinumab achieved endoscopic response and 10.9% of patients achieved endoscopic remission at week (w)44. We aimed to evaluate the impact of alternative ustekinumab dosage regimens on endoscopic outcomes based on population pharmacokinetic-pharmacodynamic (popPK-PD) modelling and simulation analysis. Methods: Real-world data were obtained from 83 patients with moderate-to-severe CD (95% biological-refractory) enrolled in a prospective cohort study receiving intravenous ustekinumab (similar to 6 mg/kg) followed by every eight-week (q8w) subcutaneous maintenance therapy (90 mg). Three sequential models were developed: a two-compartment popPK model linking ustekinumab dose to ustekinumab exposure, an indirect response popPK-PD model describing the effect of ustekinumab exposure on fecal calprotectin (fCal), and a logistic regression outcome model linking fCal to endoscopic outcomes. Results: Ustekinumab clearance increased with decreasing serum albumin and increasing bodyweight. fCal decreased with increasing ustekinumab exposure. The probability of endoscopic response at w24 increased from 10.0% to 17.9% with fCal at w8 decreasing from 1800 mu g/g to 694 mu g/g (EC50). The probability of endoscopic remission at w24 increased from 2.1% to 10.0% with fCal at w8 decreasing from 1800 mu g/g to 214 mu g/g (EC50). Simulation-based comparison of q8w and q4w maintenance dosing regimens predicted 16.7% and 22.2% endoscopic response rates, respectively. Endoscopic remission rates were estimated to be 4.2% on q8w dosing and 6.7% on q4w dosing. Conclusions: The developed models can guide clinical trial design and support model-informed dose optimization (stratified or individualized dosing) to improve endoscopic outcomes.

Meteorological satellite imagery is critical for meteorologists. The data have played an important role in monitoring and analyzing weather and climate changes. However, satellite imagery is a kind of observation data and exists a significant time delay when transmitting the data back to Earth. It is important to make accurate predictions for meteorological satellite images, especially the nowcasting prediction up to 2 hours ahead. In recent years, there has been growing interest in the research of nowcasting prediction applications of weather radar images based on deep learning. Compared to the weather radar images prediction problem, the main challenge for meteorological satellite images prediction is the large-scale observation areas and therefore the large sizes of the observation products. Here we present a deep multi-scales extrapolation fusion method, to address the challenge of the meteorological satellite images nowcasting prediction. First, we downsample the original satellite images dataset with large size to several images datasets with smaller resolutions, then we use a deep spatiotemporal sequences prediction method to generate the multi-scales prediction images with different resolutions separately. Second, we fuse the multi-scales prediction results to the targeting prediction images with the original size by a conditional generative adversarial network. The experiments based on the FY-4A meteorological satellite data show that the proposed method can generate realistic prediction images that effectively capture the evolutions of the weather systems in detail. We believe that the general idea of this work can be potentially applied to other spatiotemporal sequence prediction tasks with a large size.

This paper proposes a novel Unified Feature Optimization (UFO) paradigm for training and deploying deep models under real-world and large-scale scenarios, which requires a collection of multiple AI functions. UFO aims to benefit each single task with a large-scale pretraining on all tasks. Compared with the well known foundation model, UFO has two different points of emphasis, i.e., relatively smaller model size and NO adaptation cost: 1) UFO squeezes a wide range of tasks into a moderate-sized unified model in a multi-task learning manner and further trims the model size when transferred to down-stream tasks. 2) UFO does not emphasize transfer to novel tasks. Instead, it aims to make the trimmed model dedicated for one or more already-seen task. With these two characteristics, UFO provides great convenience for flexible deployment, while maintaining the benefits of large-scale pretraining. A key merit of UFO is that the trimming process not only reduces the model size and inference consumption, but also even improves the accuracy on certain tasks. Specifically, UFO considers the multi-task training and brings two-fold impact on the unified model: some closely related tasks have mutual benefits, while some tasks have conflicts against each other. UFO manages to reduce the conflicts and to preserve the mutual benefits through a novel Network Architecture Search (NAS) method. Experiments on a wide range of deep representation learning tasks (i.e., face recognition, person re-identification, vehicle re-identification and product retrieval) show that the model trimmed from UFO achieves higher accuracy than its single-task-trained counterpart and yet has smaller model size, validating the concept of UFO. Besides, UFO also supported the release of 17 billion parameters computer vision (CV) foundation model which is the largest CV model in the industry., Comment: Accepted in ECCV 2022

Yuanli Luo1 *, Bin Qiao1 *, Chao Yang2 *, Ping Zhang,1 Zhuoyan Xie,1 Jin Cao,1 Anyu Yang,1 Qinyanqiu Xiang,1 Haitao Ran,1 Zhigang Wang,1 Lan Hao,1 Yang Cao,1 Zhiyi Zhou,3 Jianli Ren1 1Chongqing Key Laboratory of Ultrasound Molecular Imaging, the Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 2Radiology Department, Chongqing General Hospital, Chongqing, 400014, People’s Republic of China; 3General Practice Department, Chongqing General Hospital, Chongqing, 400014, People’s Republic of China* These authors contributed equally to this workCorrespondence: Zhiyi Zhou; Jianli Ren, Email renjianli@cqmu.edu.cn; zzy16966@163.comBackground: Sonodynamic therapy (SDT) has been regarded as a novel therapeutic modality for killing tumors. However, the hypoxic tumor microenvironment, especially deep-seated tumors distant from blood vessels, severely restricts therapeutic efficacy due to the oxygen-dependent manner of SDT.Methods: Herein, we report a novel ultrasonic cavitation effect-based therapeutic modality that is able to facilitate the hypoxia-tolerant SDT for inducing hypoxic tumor death. A tLyP-1 functionalized liposomes is fabricated, composed of hematoporphyrin monomethyl ether gadolinium as the sonosentizer and perfluoropentane (PFP) as the acoustic environment regulator. Moreover, the tLyP-1 functioned liposomes could achieve active tumor homing and effective deep-penetrating into hypoxic tumors. Upon low intensity focused ultrasound (LIFU) irradiation, the acoustic droplet vaporization effect of PFP induced fast liquid-to-gas transition and quick bubbles explosion to generate hydroxyl radicals, efficiently promoting cell death in both normoxic and hypoxic microenvironment (acting as deep-penetration nanobomb, DPNB).Results: The loading of PFP is proved to significantly enhance the therapeutic efficacy of hypoxic tumors. In particular, these DPNB can also act as ultrasound

The booming of electric vehicles demands efficient battery disassembly for recycling to be environment-friendly. Due to the unstructured environment and high uncertainties, battery disassembly is still primarily done by humans, probably assisted by robots. It is highly desirable to design autonomous solutions to improve work efficiency and lower human risks in high voltage and toxic environments. This paper proposes a novel framework of the NeuroSymbolic task and motion planning method to disassemble batteries in an unstructured environment using robots automatically. It enables robots to independently locate and disassemble battery bolts, with or without obstacles. This study not only provides a solution for intelligently disassembling electric vehicle batteries but also verifies its feasibility through a set of test results with the robot accomplishing the disassembly tasks in a complex and dynamic environment., Comment: Accepted to IntelliSys 2022(Intelligent Systems Conference),15 pages with 6 figures

Most existing unsupervised person re-identification (Re-ID) methods use clustering to generate pseudo labels for model training. Unfortunately, clustering sometimes mixes different true identities together or splits the same identity into two or more sub clusters. Training on these noisy clusters substantially hampers the Re-ID accuracy. Due to the limited samples in each identity, we suppose there may lack some underlying information to well reveal the accurate clusters. To discover these information, we propose an Implicit Sample Extension (\OurWholeMethod) method to generate what we call support samples around the cluster boundaries. Specifically, we generate support samples from actual samples and their neighbouring clusters in the embedding space through a progressive linear interpolation (PLI) strategy. PLI controls the generation with two critical factors, i.e., 1) the direction from the actual sample towards its K-nearest clusters and 2) the degree for mixing up the context information from the K-nearest clusters. Meanwhile, given the support samples, ISE further uses a label-preserving loss to pull them towards their corresponding actual samples, so as to compact each cluster. Consequently, ISE reduces the "sub and mixed" clustering errors, thus improving the Re-ID performance. Extensive experiments demonstrate that the proposed method is effective and achieves state-of-the-art performance for unsupervised person Re-ID. Code is available at: \url{https://github.com/PaddlePaddle/PaddleClas}., Comment: CVPR2022

The booming of electric vehicles demands efficient battery disassembly for recycling to be environment-friendly. Currently, battery disassembly is still primarily done by humans, probably assisted by robots, due to the unstructured environment and high uncertainties. It is highly desirable to design autonomous solutions to improve work efficiency and lower human risks in high voltage and toxic environments. This paper proposes a novel neurosymbolic method, which augments the traditional Variational Autoencoder (VAE) model to learn symbolic operators based on raw sensory inputs and their relationships. The symbolic operators include a probabilistic state symbol grounding model and a state transition matrix for predicting states after each execution to enable autonomous task and motion planning. At last, the method's feasibility is verified through test results.

Xin He,1 Jiping Sun,2 Chao Liu,2 Xiaoyang Yu,2 Huixian Li,2 Wenjing Zhang,2 Yan Li,2 Yingzhou Geng,2 Zhigang Wang2 1Dialysis Department of Nephrology Hospital, First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, 710061, People’s Republic of China; 2Department of Nephrology, First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, 710061, People’s Republic of ChinaCorrespondence: Zhigang Wang, First Affiliated Hospital of Xi’an Jiaotong University, West Yanta Road 277, Xi’an, 710061, People’s Republic of China, Email wangzhigang2017@xjtu.edu.cnPurpose: Emerging evidence has revealed that gut microbiota plays a pivotal role in the pathogenesis of type 2 diabetes mellitus (T2DM) and diabetic kidney disease (DKD). However, few studies have used metagenomic sequencing to analyze the alterations of gut microbiota community in patients with early-stage DKD.Methods: We carried out metagenomic sequencing in fecal samples of 10 DKD patients (DKD group) and 10 T2DM patients who appeared to be less prone to DKD (non-DKD group), aiming to compare the composition and function of gut microbiota between the DKD and non-DKD groups.Results: The gut microbial community of the DKD group was significantly different from that of the non-DKD group, characterized by a marked increase in phylum Proteobacteria, genus Selenomonadales, Neosynechococcus, Shigella, Bilophila, Acidaminococcus, species, Escherichia coli, Bacteroides plebeius, Megasphaera elsdenii, Acidaminococcus unclassified, and Bilophila wadsworthia. The amounts of species Citrobacter farmeri and Syntrophaceticus schinkii were significantly and positively correlated with the urinary albumin creatinine ratio in the DKD group. Furthermore, functional analysis based on dbCAN and KEGG databases showed aberrant lipopolysaccharide (LPS) biosynthesis and carbohydrate metabolism in the gut microbiome of the DKD group.Conclusion: Our findings provided evidence for alterations in the composition and fun

Qiang Yang,1 Yanghao Zhou,1 Jin Chen,1 Ning Huang,1 Zhigang Wang,2 Yuan Cheng1 1Department of Neurosurgery, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of China; 2Institute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of ChinaCorrespondence: Yuan ChengDepartment of Neurosurgery, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of ChinaTel +86 023 63693540Email chengyuan@hospital.cqmu.edu.cnBackground: Therapy for glioblastoma (GBM) has always been very challenging, not only because of the presence of the blood-brain barrier (BBB) but also due to susceptibility to drug resistance. Recently, the clustered regularly interspaced short palindromic repeats (CRISPR)-associated protein 9 (CRISPR/Cas9) has revolutionized gene editing technology and is capable of treating a variety of genetic diseases, including human tumors, but there is a lack of safe and effective targeting delivery systems in vivo, especially in the central nervous system (CNS).Methods: Lipid-polymer hybrid nanoparticles (LPHNs-cRGD) were constructed for efficient and targeting delivery of CRISPR/Cas9 plasmids targeting O6-methylguanine-DNA methyltransferase (MGMT), a drug-resistance gene to temozolomide (TMZ). Focused ultrasound (FUS)-microbubbles (MBs) were used to non-invasively and locally open the BBB to further facilitate gene delivery into glioblastoma in vivo. The gene editing efficiency and drug sensitivity changes were evaluated both in vitro and in vivo.Results: The gene-loaded LPHNs-cRGD were successfully synthesized and could protect pCas9/MGMT from enzyme degradation. LPHNs-cRGD could target GBM cells and mediate the transfection of pCas9/MGMT to downregulate the expression of MGMT, resulting in an increased sensitivity of GBM cells to TMZ. MBs-LPHNs-cRGD complexes could safely an

Jun Zhou,1,* Jingxin Hou,1,2,* Shuling Liu,1 Jie Xu,1 Ying Luo,1 Jun Zheng,2 Xin Li,1 Zhigang Wang,2 Haitao Ran,2 Dajing Guo1 1Department of Radiology, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 2Chongqing Key Laboratory of Ultrasound Molecular Imaging & Department of Ultrasound, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China*These authors contributed equally to this workCorrespondence: Dajing Guo Email guodaj@163.comBackground: Some patients with cervical cancer have the need to preserve fertility; therefore, a minimally invasive treatment option that can effectively inactivate tumors in these patients is necessary.Methods: In this paper, we designed and prepared nanoparticles (NPs) carrying IR780 and perfluorohexane (PFH) and characterized their properties. We focused on the promotion of programmed low-intensity focused ultrasound (LIFU) irradiation on the penetration and treatment of cervical cancer. First we used penetration-enhancing LIFU irradiation to promote the penetration of the NPs into 3D multicellular tumor spheroids (MCTSs) and tumors in tumor-bearing nude mice. Then we used re-therapeutic LIFU irradiation to achieve antitumor effects in vitro and in vivo. Photoacoustic (PA) and magnetic resonance (MR) imaging were used to monitor and evaluate the targeting and therapeutic effects of these NPs on tumor tissues.Results: The NPs prepared in this paper exhibited high affinity for HeLa cells, and can selectively achieve mitochondrial localization in the cell due to IR780 assistance. The penetration-enhancing LIFU irradiation have the ability to promote the penetration of the NPs into cervical cancer models in vivo and in vitro. Under LIFU irradiation, the cytotoxic reactive oxygen species (ROS) produced by IR780 during the first half of the re-therapeutic LIFU irradiation and the physical acoustic drople

Huipu Li,1,* Shasha Shi,1,* Meng Wu,2,* Wei Shen,1 Jianli Ren,3 Zhechuan Mei,1 Haitao Ran,3 Zhigang Wang,3 Yi Tian,4 Jian Gao,1 Hongyun Zhao1,3 1Department of Gastroenterology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 2Department of Ultrasound, Zhongnan Hospital of Wuhan University, Wuhan, Hubei, People’s Republic of China; 3Chongqing Key Laboratory of Ultrasound Molecular Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 4Department of Plastic Surgery, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China*These authors contributed equally to this workCorrespondence: Hongyun Zhao; Jian Gao Email zhychina@cqmu.edu.cn; 982213482@qq.comPurpose: Prepare a multifunctional ultrasound molecular probe, cell-penetrating peptide-modified 10-hydroxycamptothecin-loaded phase-transformation lipid nanoparticles (iRGD-ICG-10-HCPT-PFP-NPs), and to combine iRGD-ICG-10-HCPT-PFP -NPs with low-intensity focused ultrasound (LIFU) for precision theranostics against hepatocellular carcinoma (HCC).Materials and Methods: The morphology of nanoparticles (NPs) and iRGD-ICG-10-HCPT-PFP-NPs was detected. In vitro, we examined targeting ability by flow cytometry and confocal laser scanning microscopy (CLSM), assessed penetration ability into hepatoma cells, and assessed killing ability. In vivo, we examined the targeting ability of the NPs with a photoacoustic (PA) imager and fluorometer (FL), while LIFU irradiation was used to trigger the release of chemotherapeutic drugs, which had a therapeutic effect on tumors.Results: The particle size of iRGD-ICG-10-HCPT-PFP-NPs was 298.4 ± 10.42 nm. In vitro, iRGD-ICG-10-HCPT-PFP-NPs bound more to SK-Hep1 cells than ICG-10-HCPT-PFP-NPs. iRGD-ICG-10-HCPT-PFP-NPs could achieve PA/ultrasound imaging. The percentage of antiproliferative and apoptotic

Liandi Huang,1,2 Huanhuan Gao,3 Zhigang Wang,2,4 Yixin Zhong,5 Lan Hao,4 Zhiyu Du1 1Department of Ophthalmology, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 2State Key Laboratory of Ultrasound in Medicine and Engineering, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 3Department of Ophthalmology, Third Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 4Chongqing Key Laboratory of Ultrasound Molecular Imaging, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of China; 5Department of Radiology, Second Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of ChinaCorrespondence: Zhiyu DuDepartment of Ophthalmology, Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of ChinaEmail duzhiyu@hospital.cqmu.edu.cnPurpose: Anti-inflammation is essential for dry eye disease. Traditional anti-inflammation agent corticosteroids applied in dry eye disease (DED) treatment could result in high intraocular pressure, especially in long-term treatment. Thus, we have prepared a liposome loading 1-bromoheptadecafluorooctane and tetrandrine (PFOB@LIP-Tet) to treat DED via anti-inflammation that hardly affects intraocular pressure in this study, which provided another therapy strategy for dry eye disease.Methods: We firstly detected the physicochemical properties of PFOB@LIP-Tet. Next, we tested the biosafety of synthesized liposomes for corneal epithelium. Then, we explored the accumulations and distribution of PFOB@LIP-Tet both in cellular and animal models. And then, we assessed the therapeutic effects of PFOB@LIP-Tet formulations by laboratory and clinical examinations. Last, we examined the changes in eye pressure before and after trea

Jiawen Zhao,1 Liang Zhang,2 Yingjie Qi,3 Kui Liao,4 Zhigang Wang,5 Ming Wen,1 Di Zhou1 1Department of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 2Department of Ultrasound, The First Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 3Department of Intensive Care Unit (ICU), Dianjiang People’s Hospital of Chongqing, Chongqing, People’s Republic of China; 4Department of Oncology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 5Institute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of ChinaCorrespondence: Di ZhouDepartment of Radiology, The First Affiliated Hospital of Chongqing Medical University, Chongqing, 400010, People’s Republic of ChinaTel +8613508329306Email zhoudi@cqmu.edu.cnPurpose: Multifunctional nanoparticles with targeted therapeutic function and diagnostic-imaging are of great interest in the domain of precision therapy. NIR laser responsive nanoparticles (PLGA-PEG-FA encapsulating Bi2S3, PFP, and Dox (designed as FBPD NPs)) are synthesized for ovarian cancer targeted combination therapy with CT/PA dual-modal imaging guidance (PA: photoacoustic; CT: X-ray computed tomography).Methods and Results: The FBPD NPS prepared by the double emulsification method revealed excellent dispersity, great stability, outstanding optical properties. The temperature of FBPD NPs increased rapidly after laser irradiation, inducing liquid-to-gas conversion of perfluoropentane (PFP), and promoting the release of Dox up to 86.7%. These FBPD NPs demonstrated their outstanding imaging capability for both PA and CT imaging both in vitro and in vivo, providing the potential for therapeutic guidance and monitoring. Assisted by folic acid, these nanoparticles could highly enrich in ovarian tumor tissue and the accumulation peaked at 3 h

Developing a universal strategy to design piezochromic luminescent materials with desirable properties remains challenging. Here, we report that insertion of a non-emissive molecule into a donor (perylene) and acceptor (1,2,4,5-tetracyanobezene) binary cocrystal can realize fine manipulation of intermolecular interactions between perylene and 1,2,4,5-tetracyanobezene (TCNB) for desirable piezochromic luminescent properties. A continuous pressure-induced emission enhancement up to 3 GPa and a blue shift from 655 to 619 nm have been observed in perylene-TCNB cocrystals upon THF insertion, in contrast to the red-shifted and quenched emission observed when compressing perylene-TCNB cocrystals and other cocrystals reported earlier. By combining experiment with theory, it is further revealed that the inserted non-emissive THF forms blue-shifting hydrogen bonds with neighboring TCNB molecules and promote a conformation change of perylene molecules upon compression, causing the blue-shifted and enhanced emission. This strategy remains valid when inserting other molecules as non-emissive component into perylene-TCNB cocrystals for abnormal piezochromic luminescent behaviors.

Finding an optimal subset of nodes or links to disintegrate harmful networks is a fundamental problem in network science, with potential applications to anti-terrorism, epidemic control, and many other fields of study. The challenge of the network disintegration problem is to balance the effectiveness and efficiency of strategies. In this paper, we propose a cost-effective targeted enumeration method for network disintegration. The proposed approach includes two stages: searching for candidate objects and identifying an optimal solution. In the first stage, we use rank aggregation to generate a comprehensive ranking of node importance, upon which we identify a small-scale candidate set of nodes to remove. In the second stage, we use an enumeration method to find an optimal combination among the candidate nodes. Extensive experimental results on synthetic and real-world networks demonstrate that the proposed method achieves a satisfying trade-off between effectiveness and efficiency. The introduced two-stage targeted enumeration framework can also be applied to other computationally intractable combinational optimization problems, from team assembly via portfolio investment to drug design., Comment: 9 pages, 5 figures

Unsupervised domain adaptation (UDA) methods for person re-identification (re-ID) aim at transferring re-ID knowledge from labeled source data to unlabeled target data. Although achieving great success, most of them only use limited data from a single-source domain for model pre-training, making the rich labeled data insufficiently exploited. To make full use of the valuable labeled data, we introduce the multi-source concept into UDA person re-ID field, where multiple source datasets are used during training. However, because of domain gaps, simply combining different datasets only brings limited improvement. In this paper, we try to address this problem from two perspectives, \ie{} domain-specific view and domain-fusion view. Two constructive modules are proposed, and they are compatible with each other. First, a rectification domain-specific batch normalization (RDSBN) module is explored to simultaneously reduce domain-specific characteristics and increase the distinctiveness of person features. Second, a graph convolutional network (GCN) based multi-domain information fusion (MDIF) module is developed, which minimizes domain distances by fusing features of different domains. The proposed method outperforms state-of-the-art UDA person re-ID methods by a large margin, and even achieves comparable performance to the supervised approaches without any post-processing techniques., Comment: CVPR 2021 Oral

Tianlin Liu,1,2 Chunping Yin,1 Yanan Li,1 Fang Gao,1 Lili Yu,1,2 Zhigang Wang,1 Qiujun Wang1 1Department of Anesthesiology, The Third Hospital of Hebei Medical University, Shijiazhuang City, Hebei, People’s Republic of China; 2Department of Anesthesiology, Cangzhou Central Hospital, Cangzhou, Hebei, People’s Republic of ChinaCorrespondence: Qiujun Wang Tel +86 311 8860 2072Email wangqiujunsy@163.comBackground: Postoperative cognitive decline (POCD) in the old ages seriously delays the rapid recovery. Here, we aimed to investigate the effects of transcutaneous electrical acupoint stimulation (TEAS) against POCD in elderly patients undergoing laparoscopic radical colon cancer surgery, as well as the potential mechanism.Methods: A prospective, single-center, parallel-group, randomized trial was designed. A total of 100 patients (age ≥ 65 years) undergoing laparoscopic radical resection of colon cancer were involved and randomly divided into TEAS (Group T) and control (Group C) groups. The patients in Group T were performed with percutaneous acupoint electrical stimulation in bilateral Hegu, Neiguan and Zusanli points from 30 minutes before anesthesia induction to the end of surgery. A Z-score based on Mini-Mental State Exam (MMSE) was used to assess the incidence of POCD. The levels of serum IL-6, hs-CRP, CGRP at 0 min before TEAS (T0), 1 h after beginning of surgery (T1) and the end of surgery (T2) were evaluated.Results: Our data showed that the cumulative duration of POCD on postoperative day 2 and 3 in Group T was significantly decreased compared to Group C (P < 0.05). Compared with T0, the levels of serum IL-6, hs-CRP, and CGRP in both Group T and C were statistically elevated at T1 and T2 (P < 0.05). Moreover, the levels of serum IL-6 and hs-CRP were decreased, but the level of CGRP was increased in Group T compared to Group C at T1 and T2 (P < 0.05).Conclusion: TEAS is associated with a lower cumulative duration of

Zhigang Wang,1,2 Tianlin Liu,1 Chunping Yin,1 Yanan Li,1 Fang Gao,1 Lili Yu,1 Qiujun Wang1 1Department of Anesthesiology, The Third Hospital of Hebei Medical University, Shijiazhuang City, Hebei, People’s Republic of China; 2Department of Anesthesiology, Handan Central Hospital, Handan, Hebei, People’s Republic of ChinaCorrespondence: Qiujun Wang Tel +86 311 8860 2072Email wangqiujunsy@163.comObjective: Postoperative cognitive dysfunction (POCD) after anesthesia and surgery (AS) is a common complication in the elderly population. A cholinergic-dependent signal, the alpha7-nicotinic acetylcholine receptor (α 7-nAChR), has been suggested to regulate cognitive processes in a variety of neurologic diseases. In the current study, we determined whether electroacupuncture (EA) pretreatment ameliorates AS-induced POCD in aged rats, as well as the underlying mechanism.Methods: Male Sprague-Dawley rats (20 months old) were randomly assigned to the following 5 groups (n=12): vehicle; POCD (tibial fracture surgery); EA plus POCD; EA plus POCD and alpha-bungarotoxin (α-BGT); and POCD plus α-BGT groups. Alpha-bungarotoxin (1 μg/kg), a selective antagonist of α 7-nAChR, was administrated via intraperitoneal injection before EA. Thirty days post-AS, the Morris water maze and a novel objective recognition test were used to evaluate cognitive function. Neuronal amount, apoptosis, microglial activation, percentage of high mobility group box 1 (HMGB1)- and nuclear factor-κB (NF-κB)-positive microglia, and levels of HMGB-1 downstream factors, including NF-κB, interleukin-6 (IL-6), and IL-1β, were detected by Nissl staining, immunofluorescence, and Western blot assays.Results: EA pretreatment significantly increased crossing platform times and elevated the time with a novel object, restored the quantity of neurons, decreased TUNEL-positive neurons, alleviated activation of microglia, downregulated expression of HMGB1 and NF-κB in the microglia, and reduced levels of phospho

Li M, Luo H, Zhang W, et al. Int J Nanomedicine. 2018;13:3907–3920. The authors have advised that they covered up some larger-sized particles and impurities to enhance the aesthetic of the image shown for Figure 4A, on page 3913. The authors also removed some impurities and enhanced certain aspects of the image shown for Figure 4B. These two images were used to display the general appearance and shapes of the nanoparticles the authors had successfully prepared in the study. The editor determined that while the images had been altered it did not affect the conclusions of the study. The authors wish to apologise for these errors. The correct Figure 4 is shown below. Read the original article

Yubei He,1 Mengzhu Wang,2 Ming Fu,2 Xun Yuan,3 Yuanli Luo,2 Bin Qiao,2 Jin Cao,2 Zhigang Wang,2 Lan Hao,2,* Gengbiao Yuan1,* 1Department of Nuclear Medicine, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of China; 2Chongqing Key Laboratory of Ultrasound Molecular Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of China; 3Department of Ophthalmology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of China*These authors contributed equally to this workCorrespondence: Lan Hao; Gengbiao Yuan Email lanhao5@cqmu.edu.cn; 300784@hospital.cqmu.edu.cnPurpose: A multi-functional nanoplatform with diagnostic imaging and targeted treatment functions has aroused much interest in the nanomedical research field and has been paid more attention in the field of tumor diagnosis and treatment. However, some existing nano-contrast agents have encountered difficulties in different aspects during clinical promotion, such as complicated preparation process and low specificity. Therefore, it is urgent to find a nanocomplex with good targeting effect, high biocompatibility and significant therapeutic effect for the integration of diagnosis and treatment and clinical transformation.Materials and Methods: Nanoparticles (NPs) targeting breast cancer were synthesized by phacoemulsification which had liquid fluorocarbon perfluoropentane(PFP) in the core and were loaded with Iron(II) phthalocyanine (FePc) on the shell. The aptamer (APT) AS1411 was outside the shell used as a molecular probe. Basic characterization and targeting abilities of the NPs were tested, and their cytotoxicity and biological safety in vivo were evaluated through CCK-8 assay and blood bio-chemical analysis. The photoacoustic (PA) and ultrasound (US) imaging system were used to assess the effects of AS1411-PLGA@FePc@PFP (A-FP NP

Mengzhu Wang,1 Zhigang Wang,1 Bin Qiao,1 Jin Cao,1 Luya Quan,1 Yuanli Luo,1 Hanwen Qi,2 Xiaowen Zhong,1 Yubei He,1 Xianquan Zhang,1,3 Lan Hao1 1Institute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, Chongqing 400010, People’s Republic of China; 2The A. Gary Anderson Graduate School of Management, University of California, Riverside, Riverside, CA, 92521, USA; 3Cardiothoracic Surgery Department, Chongqing Hygeia Cancer Hospital, Chongqing, 401331, People’s Republic of ChinaCorrespondence: Lan HaoInstitute of Ultrasound Imaging, The Second Affiliated Hospital of Chongqing Medical University, 1# Yixueyuan Road, Yuzhong District, Chongqing, People’s Republic of ChinaTel +86 15823291038Email lanhao5@cqmu.edu.cnXianquan ZhangChongqing Hygeia Cancer Hospital, No. 222, Sixian Road, University Town, Shapingba District, Chongqing 401331, People’s Republic of ChinaTel +86 13996286125Email xqzhng@163.comPurpose: Tumor metastasis and drug resistance have always been vital aspects to cancer mortality and prognosis. To compromise metastasis and drug resistance, a nanoparticle IPPD-PHF2 (IR780/PLGA-PEI(Dox)-PHF2) has been engineered to accomplish efficient targeted epigenotherapy forced by PHF2-induced MET (mesenchymal to epithelial transition).Materials and Methods: IPPD-PHF2 nanoparticle was synthesized and characterized by several analytical techniques. The transfection efficiency of IPP-PHF2 (IR780/PLGA-PEI-PHF2) was compared with PP-PHF2 (PLGA-PEI-PHF2) in vitro by WB and in vivo by IHC, and the cytotoxicity of IPP was compared with Lipo2000 in vitro by CCK8 assay. The inhibition of cancer cell migration caused by PHF2-upregulation was tested by wound healing assay, and the enhanced chemotherapeutic sensitivity was detected by flow cytometry. Tumor-targeting property of IPPD-PHF2 was proved by fluorescent imaging in vivo with MDA-MB-231 tumor-bearing nude mice. Except for fluorescent imaging abil

Jie Xu,1 Jun Zhou,1 Yixin Zhong,1 Yu Zhang,1 Man Ye,1 Jingxin Hou,1 Zhigang Wang,2 Haitao Ran,2 Jia Liu,1 Dajing Guo1 1Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of China; 2Institute of Ultrasound Imaging, Department of Ultrasound, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, People’s Republic of ChinaCorrespondence: Dajing Guo; Jia LiuDepartment of Radiology, The Second Affiliated Hospital of Chongqing Medical University, No. 74 Linjiang Road, Yuzhong District, Chongqing 400010, People’s Republic of ChinaTel +86 23 6369 3781Email guodaj@163.com; 442775883@qq.comBackground: There have been many recent reports of molecular probes for thrombi but with unsatisfactory in vivo targeting effects, which could be related to the blood flow velocity in vivo. Therefore, it is worth explaining the relationship between the targeting effect and the blood flow velocity.Methods and Materials: In this study, we constructed a platelet-targeting nanoparticle (NP) based on EWVDV for targeting P-selectin combined with the phase transition material perfluorohexane and India ink to achieve the multimodal imaging of thrombi. We studied the targeting effect of the NPs for rabbit blood thrombi under different flow velocities simulating blood flow velocities in vivo.Results: The results show the successful fabrication of NPs with the ability to undergo a phase transition via low-intensity focused ultrasound irradiation to achieve ultrasound imaging and with a high binding affinity for activated platelets. In vitro, low flow velocities (20 cm/s) hardly affected the targeting effect of the NPs, while moderate flow velocities (40 cm/s) reduced the number of NPs that target thrombi by 52.6% comparing to static fluid (0 cm/s). High flow velocities (60 cm/s) greatly reduced the targeting effect of the NPs by 83.5%.Conclusion: These results can serve as a reference f