Search

Background: By combining theory-driven and data-driven methods, this study aimed to develop dementia predictive algorithms among Chinese older adults guided by the cognitive footprint theory. Methods: Electronic medical records from the Clinical Data Analysis and Reporting System in Hong Kong were employed. We included patients with dementia diagnosed at 65+ between 2010 and 2018, and 1:1 matched dementia-free controls. We identified 51 features, comprising exposures to established modifiable factors and other factors before and after 65 years old. The performances of four machine learning models, including LASSO, Multilayer perceptron (MLP), XGBoost, and LightGBM, were compared with logistic regression models, for all patients and subgroups by age. Findings: A total of 159,920 individuals (40.5% male; mean age [SD]: 83.97 [7.38]) were included. Compared with the model included established modifiable factors only (area under the curve [AUC] 0.689, 95% CI [0.684, 0.694]), the predictive accuracy substantially improved for models with all factors (0.774, [0.770, 0.778]). Machine learning and logistic regression models performed similarly, with AUC ranged between 0.773 (0.768, 0.777) for LASSO and 0.780 (0.776, 0.784) for MLP. Antipsychotics, education, antidepressants, head injury, and stroke were identified as the most important predictors in the total sample. Age-specific models identified different important features, with cardiovascular and infectious diseases becoming prominent in older ages. Interpretation: The models showed satisfactory performances in identifying dementia. These algorithms can be used in clinical practice to assist decision making and allow timely interventions cost-effectively.

BACKGROUND: Most studies on the impact of the COVID-19 pandemic on depression burden focused on the earlier pandemic phase specific to lockdowns, but the longer-term impact of the pandemic is less well-studied. In this population-based cohort study, we examined the short-term and long-term impacts of COVID-19 on depression incidence and healthcare service use among patients with depression. METHODS: Using the territory-wide electronic medical records in Hong Kong, we identified all patients aged ≥ 10 years with new diagnoses of depression from 2014 to 2022. We performed an interrupted time-series (ITS) analysis to examine changes in incidence of medically attended depression before and during the pandemic. We then divided all patients into nine cohorts based on year of depression incidence and studied their initial and ongoing service use patterns until the end of 2022. We applied generalized linear modeling to compare the rates of healthcare service use in the year of diagnosis between patients newly diagnosed before and during the pandemic. A separate ITS analysis explored the pandemic impact on the ongoing service use among prevalent patients with depression. RESULTS: We found an immediate increase in depression incidence (RR = 1.21, 95% CI: 1.10-1.33, p < 0.001) in the population after the pandemic began with non-significant slope change, suggesting a sustained effect until the end of 2022. Subgroup analysis showed that the increases in incidence were significant among adults and the older population, but not adolescents. Depression patients newly diagnosed during the pandemic used 11% fewer resources than the pre-pandemic patients in the first diagnosis year. Pre-existing depression patients also had an immediate decrease of 16% in overall all-cause service use since the pandemic, with a positive slope change indicating a gradual rebound over a 3-year period. CONCLUSIONS: During the pandemic, service provision for depression was suboptimal in the face of increa

This paper investigates the impact of children's recess activity patterns on particulate matter (PM) resuspension in indoor environments, highlighting the complex, multi-dimensional nature of these activities and their interaction with environmental parameters. Despite the recognized role of indoor human activity in PM resuspension, research specifically addressing the effects of children's movements has been sparse. Through experimental scenarios that account for the characteristics of student activities, such as movement speed, trajectory, the number of participants, aisle widths, and varying humidity levels, this study uncovers significant differences in PM resuspension rates. It reveals that not only do movement speed and trajectory have a profound impact, but also the interaction between humidity and these factors plays a critical role, especially under lower humidity conditions. Additionally, the study demonstrates how the combination of people density and spatial configurations can significantly influence resuspension rates. The findings offer valuable insights for designing strategies to mitigate particle pollution in classrooms and similar indoor environments., QC 20240708

Supervised image captioning approaches have made great progress, but it is challenging to collect high-quality human-annotated image-text data. Recently, large-scale vision and language models (e.g., CLIP) and large-scale generative language models (e.g., GPT-2) have shown strong performances in various tasks, which also provide some new solutions for image captioning with web paired data, unpaired data or even text-only data. Among them, the mainstream solution is to project image embeddings into the text embedding space with the assistance of consistent representations between image-text pairs from the CLIP model. However, the current methods still face several challenges in adapting to the diversity of data configurations in a unified solution, accurately estimating image-text embedding bias, and correcting unsatisfactory prediction results in the inference stage. This paper proposes a new Text data-centric approach with Interactive Prompts for image Captioning, named TIPCap. 1) We consider four different settings which gradually reduce the dependence on paired data. 2) We construct a mapping module driven by multivariate Gaussian distribution to mitigate the modality gap, which is applicable to the above four different settings. 3) We propose a prompt interaction module that can incorporate optional prompt information before generating captions. Extensive experiments show that our TIPCap outperforms other weakly or unsupervised image captioning methods and achieves a new state-of-the-art performance on two widely used datasets, i.e., MS-COCO and Flickr30K.

Existing hyperspectral image (HSI) super-resolution (SR) methods struggle to effectively capture the complex spectral-spatial relationships and low-level details, while diffusion models represent a promising generative model known for their exceptional performance in modeling complex relations and learning high and low-level visual features. The direct application of diffusion models to HSI SR is hampered by challenges such as difficulties in model convergence and protracted inference time. In this work, we introduce a novel Group-Autoencoder (GAE) framework that synergistically combines with the diffusion model to construct a highly effective HSI SR model (DMGASR). Our proposed GAE framework encodes high-dimensional HSI data into low-dimensional latent space where the diffusion model works, thereby alleviating the difficulty of training the diffusion model while maintaining band correlation and considerably reducing inference time. Experimental results on both natural and remote sensing hyperspectral datasets demonstrate that the proposed method is superior to other state-of-the-art methods both visually and metrically., Comment: Accepted by AAAI2024

Diffusion models have emerged as state-of-the-art generative models for image generation. However, sampling from diffusion models is usually time-consuming due to the inherent autoregressive nature of their sampling process. In this work, we propose a novel approach that accelerates the sampling of diffusion models by parallelizing the autoregressive process. Specifically, we reformulate the sampling process as solving a system of triangular nonlinear equations through fixed-point iteration. With this innovative formulation, we explore several systematic techniques to further reduce the iteration steps required by the solving process. Applying these techniques, we introduce ParaTAA, a universal and training-free parallel sampling algorithm that can leverage extra computational and memory resources to increase the sampling speed. Our experiments demonstrate that ParaTAA can decrease the inference steps required by common sequential sampling algorithms such as DDIM and DDPM by a factor of 4$\sim$14 times. Notably, when applying ParaTAA with 100 steps DDIM for Stable Diffusion, a widely-used text-to-image diffusion model, it can produce the same images as the sequential sampling in only 7 inference steps. The code is available at https://github.com/TZW1998/ParaTAA-Diffusion., Comment: ICML 2024

Reconfigurable intelligent surfaces, with their large number of antennas, offer an interesting opportunity for high spatial-resolution imaging. In this paper, we propose a novel RIS-aided integrated imaging and communication system that can reduce the RIS beam training overhead for communication by leveraging the imaging of the surrounding environment. In particular, using the RIS as a wireless imaging device, our system constructs the scene depth map of the environment, including the mobile user. Then, we develop a user detection algorithm that subtracts the background and extracts the mobile user attributes from the depth map. These attributes are then utilized to design the RIS interaction vector and the beam selection strategy with low overhead. Simulation results show that the proposed approach can achieve comparable beamforming gain to the optimal/exhaustive beam selection solution while requiring 1000 times less beam training overhead., Comment: 6 pages, 4 figures. To appear in Asilomar 2023

In this paper, we explore an integrated sensing and communication (ISAC) system with backscattering RFID tags. In this setup, an access point employs a communication beam to serve a user while leveraging a sensing beam to detect an RFID tag. Under the total transmit power constraint of the system, our objective is to design sensing and communication beams by considering the tag detection and communication requirements. First, we adopt zero-forcing to design the beamforming vectors, followed by solving a convex optimization problem to determine the power allocation between sensing and communication. Then, we study a joint beamforming design problem with the goal of minimizing the total transmit power while satisfying the tag detection and communication requirements. To resolve this, we re-formulate the non-convex constraints into convex second-order cone constraints. The simulation results demonstrate that, under different communication SINR requirements, joint beamforming optimization outperforms the zero-forcing-based method in terms of achievable detection distance, offering a promising approach for the ISAC-backscattering systems., Comment: 5 pages, 5 figures. To appear in IEEE ICC 2024

Compared with the remarkable progress made in parallel numerical solvers of partial differential equations,the development of algorithms for generating unstructured triangular/tetrahedral meshes has been relatively sluggish. In this paper, we propose a novel, consistent parallel advancing front technique (CPAFT) by combining the advancing front technique, the domain decomposition method based on space-filling curves, the distributed forest-of-overlapping-trees approach, and the consistent parallel maximal independent set algorithm. The newly proposed CPAFT algorithm can mathematically ensure that the generated unstructured triangular/tetrahedral meshes are independent of the number of processors and the implementation of domain decomposition. Several numerical tests are conducted to validate the parallel consistency and outstanding parallel efficiency of the proposed algorithm, which scales effectively up to two thousand processors. This is, as far as we know, the first parallel unstructured triangular/tetrahedral mesh generator with scalability to O(1,000) CPU processors.

In recent years, various intelligent autonomous robots have begun to appear in daily life and production. Desktop-level robots are characterized by their flexible deployment, rapid response, and suitability for light workload environments. In order to meet the current societal demand for service robot technology, this study proposes using a miniaturized desktop-level robot (by ROS) as a carrier, locally deploying a natural language model (NLP-BERT), and integrating visual recognition (CV-YOLO) and speech recognition technology (ASR-Whisper) as inputs to achieve autonomous decision-making and rational action by the desktop robot. Three comprehensive experiments were designed to validate the robotic arm, and the results demonstrate excellent performance using this approach across all three experiments. In Task 1, the execution rates for speech recognition and action performance were 92.6% and 84.3%, respectively. In Task 2, the highest execution rates under the given conditions reached 92.1% and 84.6%, while in Task 3, the highest execution rates were 95.2% and 80.8%, respectively. Therefore, it can be concluded that the proposed solution integrating ASR, NLP, and other technologies on edge devices is feasible and provides a technical and engineering foundation for realizing multimodal desktop-level robots.

We give a continuous perspective on the Inertial Corrected Primal-Dual Proximal Splitting (IC-PDPS) proposed by Valkonen ({\it SIAM J. Optim.}, 30(2): 1391--1420, 2020) for solving saddle-point problems. The algorithm possesses nonergodic convergence rate and admits a tight preconditioned proximal point formulation which involves both inertia and additional correction. Based on new understandings on the relation between the discrete step size and rescaling effect, we rebuild IC-PDPS as a semi-implicit Euler scheme with respect to its iterative sequences and integrated parameters. This leads to two novel second-order ordinary differential equation (ODE) models that are equivalent under proper time transformation, and also provides an alternative interpretation from the continuous point of view. Besides, we present the convergence analysis of the Lagrangian gap along the continuous trajectory by using proper Lyapunov functions.

The correspondence between input text and the generated image exhibits opacity, wherein minor textual modifications can induce substantial deviations in the generated image. While, text embedding, as the pivotal intermediary between text and images, remains relatively underexplored. In this paper, we address this research gap by delving into the text embedding space, unleashing its capacity for controllable image editing and explicable semantic direction attributes within a learning-free framework. Specifically, we identify two critical insights regarding the importance of per-word embedding and their contextual correlations within text embedding, providing instructive principles for learning-free image editing. Additionally, we find that text embedding inherently possesses diverse semantic potentials, and further reveal this property through the lens of singular value decomposition (SVD). These uncovered properties offer practical utility for image editing and semantic discovery. More importantly, we expect the in-depth analyses and findings of the text embedding can enhance the understanding of text-to-image diffusion models.

Compressive sensing is a promising solution for the channel estimation in multiple-input multiple-output (MIMO) systems with large antenna arrays and constrained hardware. Utilizing site-specific channel data from real-world systems, deep learning can be employed to learn the compressive sensing measurement vectors with minimum redundancy, thereby focusing sensing power on promising spatial directions of the channel. Collecting real-world channel data, however, is challenging due to the high overhead resulting from the large number of antennas and hardware constraints. In this paper, we propose leveraging a site-specific digital twin to generate synthetic channel data, which shares a similar distribution with real-world data. The synthetic data is then used to train the deep learning models for learning measurement vectors and hybrid precoder/combiner design in an end-to-end manner. We further propose a model refinement approach to fine-tune the model pre-trained on the digital twin data with a small amount of real-world data. The evaluation results show that, by training the model on the digital twin data, the learned measurement vectors can be efficiently adapted to the environment geometry, leading to high performance of hybrid precoding for real-world deployments. Moreover, the model refinement approach can enable the digital twin aided model to achieve comparable performance to the model trained on the real-world dataset with a significantly reduced amount of real-world data., Comment: 7 pages, 5 figures

Daniel Hsiang-Te Tsai,1,2 J Simon Bell,3 Shahab Abtahi,4 Brenda N Baak,5 Marloes T Bazelier,4 Ruth Brauer,6 Adrienne YL Chan,6– 9 Esther W Chan,9– 12 Haoqian Chen,13 Celine SL Chui,9,14,15 Sharon Cook,16 Stephen Crystal,16 Poonam Gandhi,13 Sirpa Hartikainen,17 Frederick K Ho,18 Shao-Ti Hsu,1 Jenni Ilomäki,3 Ju Hwan Kim,19 Olaf H Klungel,4 Marjaana Koponen,17 Wallis CY Lau,6,8,9 Kui Kai Lau,20,21 Terry YS Lum,22 Hao Luo,22 Kenneth KC Man,6,8,9 Jill P Pell,18 Soko Setoguchi,13,23 Shih-Chieh Shao,1,24 Chin-Yao Shen,1 Ju-Young Shin,19,25,26 Patrick C Souverein,4 Anna-Maija Tolppanen,17 Li Wei,6,9 Ian CK Wong,6,8,9,27 Edward Chia-Cheng Lai1 1School of Pharmacy, Institute of Clinical Pharmacy and Pharmaceutical Sciences, College of Medicine, National Cheng Kung University, Tainan, Taiwan; 2Centre for Neonatal and Paediatric Infection, St George’s University of London, London, UK; 3Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, Melbourne, Australia; 4Division of Pharmacoepidemiology and Clinical Pharmacology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, Utrecht, the Netherlands; 5PHARMO Institute for Drug Outcomes Research, Utrecht, the Netherlands; 6Research Department of Practice and Policy, UCL School of Pharmacy, London, UK; 7Groningen Research Institute of Pharmacy, Unit of Pharmacotherapy, ‐Epidemiology and ‐Economics, University of Groningen, Groningen, the Netherlands; 8Department of Pharmacology and Pharmacy, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong, Special Administrative Region, People’s Republic of China; 9Laboratory of Data Discovery for Health (D24H), Hong Kong Science Park, Hong Kong, Special Administrative Region, People’s Republic of China; 10Centre for Safe Medication Practice and Research, Department of Pharmacology and Pharmacy, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong, Special Administrative Region, Peopleâ

Rapidly shrinking Arctic sea ice has had substantial impacts on the Earth system. Therefore, reliably estimating the Arctic sea-ice thickness (SIT) using a combination of available observations and numerical modeling is urgently needed. Here, for the first time, we assimilate the latest CryoSat-2 summer SIT data into a coupled ice-ocean model. In particular, an incremental analysis update scheme is implemented to overcome the discontinuity resulting from the combined assimilation of biweekly SIT and daily sea-ice concentration (SIC) data. Along with improved estimates of sea-ice volume, our SIT estimates corrected the overestimation of SIT produced by the reanalysis that assimilates only SIC in summer in areas where the sea ice is roughest and experiences strong deformation, e.g., around the Fram Strait and Greenland. This study suggests that the newly developed CryoSat-2 SIT product, when assimilated properly using our approach, has great potential for Arctic sea-ice simulation and prediction.

Given the role played by the historical and extensive coverage of sea ice concentration (SIC) observations in reconstructing the long‐term variability of Antarctic sea ice, and the limited attention given to model‐dependent parameters in current sea ice data assimilation studies, this study focuses on enhancing the performance of the Data Assimilation System for the Southern Ocean in assimilating SIC through optimizing the localization and observation error estimate, and two assimilation experiments were conducted from 1979 to 2018. By comparing the results with the sea ice extent of the Southern Ocean and the sea ice thickness in the Weddell Sea, it becomes evident that the experiment with optimizations outperforms that without optimizations due to achieving more reasonable error estimates. Investigating uncertainties of the sea ice volume anomaly modeling reveals the importance of the sea ice‐ocean interaction in the SIC assimilation, implying the necessity of assimilating more oceanic and sea‐ice observations.

PURPOSE: To describe and categorize detailed components of databases in the Neurological and Mental Health Global Epidemiology Network (NeuroGEN).METHODS: An online 132-item questionnaire was sent to key researchers and data custodians of NeuroGEN in North America, Europe, Asia and Oceania. From the responses, we assessed data characteristics including population coverage, data follow-up, clinical information, validity of diagnoses, medication use and data latency. We also evaluated the possibility of conversion into a common data model (CDM) to implement a federated network approach. Moreover, we used radar charts to visualize the data capacity assessments, based on different perspectives.RESULTS: The results indicated that the 15 databases covered approximately 320 million individuals, included in 7 nationwide claims databases from Australia, Finland, South Korea, Taiwan and the US, 6 population-based electronic health record databases from Hong Kong, Scotland, Taiwan, the Netherlands and the UK, and 2 biomedical databases from Taiwan and the UK.CONCLUSION: The 15 databases showed good potential for a federated network approach using a common data model. Our study provided publicly accessible information on these databases for those seeking to employ real-world data to facilitate current assessment and future development of treatments for neurological and mental disorders.

As Massive Open Online Courses (MOOCs) expand and diversify, more and more researchers study recommender systems that take advantage of interaction data to keep students interested and boost their performance. In a typical roadmap, courses and videos are recommended using a graph model, but this does not take into account the user’s learning needs with some particular subjects. However, all existing graph models degrade performances either by ignoring the data sparsity issue caused by a large number of concepts, which may lead to biased recommendations, or by constructing improper contrasting pairs, which may result in graph noise. To overcome both challenges, we propose a gRaph cOntrastive Multi-view framEwork (ROME) from hyperbolic angular space to learn user and concept representations based on user-user and concept-concept relationships. The first step is to use hyperbolic and Euclidean space representations as different views of graph and maximize the mutual information between them. Furthermore, we maximize the angular decision margin in graph contrastive training objects to enhance pairwise discriminative power. Our experiments on a large-scale real-world MOOC dataset show that the proposed approach outperforms several baselines and state-of-the-art methods for predicting and recommending concepts of interest to users.

The changing landscape of transportation technology and traveler behavior, accelerated by recent events like COVID-19, has led to significant shifts in travel demand and vehicle miles traveled in Indiana. This study seeks to understand the long-term implications of these changes and their potential impact on passenger, freight, and micro-mobility movements across the state. To achieve this objective, this project focused on forecasting future transportation demand conditions and carrying out long-range scenario planning by accomplishing four tasks: forecasting travel demand shifts based on location-based data, evaluating medium-term inter- and intra-urban transportation demand shifts, forecasting county-level industry shifts using scenario-based growth models, and providing recommendations and guidance to the Indiana Department of Transportation (INDOT) based on the study results. Results offer improved planning for infrastructure investments and operations, the incorporation of emerging technologies into transportation planning processes, and an enhanced understanding of passenger and freight movements at the statewide and regional levels. Deliverables from this study include valuable tools and models that can help INDOT navigate potential transportation system changes and accommodate the evolving needs of the future.

Evidence on the healthcare utilization associated with comorbidity in people with dementia is lacking in Chinese societies. This study aimed to quantify healthcare utilization associated with comorbidity that is common in people living with dementia. We conducted a cohort study using population-based data from Hong Kong public hospitals. Individuals aged 35+ with a dementia diagnosis between 2010 and 2019 were included. Among 88,151 participants, people with at least two comorbidities accounted for 81.2%. Estimates from negative binomial regressions showed that compared to those with one or no comorbid condition other than dementia, adjusted rate ratios of hospitalizations among individuals with six or seven and eight or more conditions were 1.97 [98.75% CI, 1.89–2.05] and 2.74 [2.63–2.86], respectively; adjusted rate ratios of Accident and Emergency department visits among individuals with six or seven and eight or more conditions were 1.53 [1.44–1.63] and 1.92 [1.80–2.05], respectively. Comorbid chronic kidney diseases were associated with the highest adjusted rate ratios of hospitalizations (1.81 [1.74–1.89]), whereas comorbid chronic ulcer of the skin was associated with the highest adjusted rate ratios of Accident and Emergency department visits (1.73 [1.61–1.85]). Healthcare utilization for individuals with dementia differed substantially by both the number of comorbid chronic conditions and the presence of some specific comorbid conditions. These findings further highlight the importance of taking account of multiple long-term conditions in tailoring the care approach and developing healthcare plans for people with dementia.

Ran Cheng,1 Ning Tan,2 Hao Luo,2 Qian Kang,2 Xiaoyuan Xu3 1Department of Infectious Diseases, Peking University Third Hospital, Beijing, People’s Republic of China; 2Department of Infectious Diseases, Peking University First Hospital, Beijing, People’s Republic of China; 3Department of Gastroenterology, Peking University First Hospital, Beijing, People’s Republic of ChinaCorrespondence: Xiaoyuan Xu, Department of Gastroenterology, Peking University First Hospital, 8 Xishiku Street, Beijing, 100034, People’s Republic of China, Tel +86-010-83575787, Fax +86-010-83575787, Email xiaoyuanxu6@163.comBackground and Aims: The performance of noninvasive assessments to rule-in or rule-out fibrosis may improve when combined. We aimed to evaluate the efficiencies of sequential algorithms based on the aspartate aminotransferase-to-platelet ratio index (APRI), the fibrosis index based on four factors (FIB-4), and transient elastography (TE) for the assessment of advanced fibrosis (AF) and cirrhosis.Methods: This study enrolled 179 CHB subjects who underwent liver biopsy (LB) before antiviral treatment.Results: AF and cirrhosis were identified in 71 (39.7%) and 28 (15.7%) patients, respectively. Compared with TE alone, sequential FIB-4-TE and APRI-TE algorithms saved a slightly higher number of liver biopsies for the identification of advanced fibrosis (69.3% or 68.2% vs 63.7%, P=0.263 or P=0.372, respectively). For the identification of cirrhosis, sequential FIB-4-TE and APRI-TE algorithms saved a significantly higher number of liver biopsies than TE alone (83.2% or 88.3% vs 69.8%, P=0.003 or P=0.000, respectively). No significant difference was found between the sequential algorithms and TE alone in the diagnostic accuracy for the detection of AF and cirrhosis.Conclusion: The sequential algorithms could significantly reduce the need for liver biopsy with high accuracy for diagnosis of AF and cirrhosis in CHB patients, which would be optimal especially in resource-limited

Purpose. Exercise is emerging as an adjunct therapy to cancer treatment; however, its role in older patients with advanced pancreatic cancer undergoing first-line chemotherapy is unclear. The aim of this study was to primarily provide evidence on feasibility with an exploratory examination of the initial efficacy of exercise in this clinical setting. Materials and Methods. Six patients aged 60–75 years with de novo or recurrent advanced pancreatic cancer undergoing first-line chemotherapy consented to participate in twice-weekly exercise that included resistance and aerobic training and boxing-related activities for up to 12 weeks. Patients were monitored for attendance, adherence, and adverse events. Body composition, muscle strength, functional ability, patient-reported outcome measures, and patient-reported experience measures were assessed at baseline and/or postintervention. Results. Of the 6 patients, 1 withdrew after baseline testing and 5 attended 42%–95% of planned sessions and adhered to 28%–83% of the prescribed exercise. There were no serious exercise-emergent adverse events. All 5 patients increased or maintained lean mass (0.1%–4.4%) and 4 reduced fat mass (−0.4%–−8.6%). Improvements were observed in 4 or all 5 patients for muscle strength (7.1%–75%), 5 times sit-to-stand (1.3%–21.4%), 6-m backward walk (16.5%–35.8%), and patient-reported outcomes. Furthermore, all patients perceived exercise as very helpful in managing their cancer and expressed a strong willingness to continue exercise in the future. Conclusion. A multimodal exercise program appears feasible with potential physical and psychological benefits for older patients with advanced pancreatic cancer undergoing first-line chemotherapy. Further research including a larger sample size is warranted.

In reactor-scale CFD modeling of biomass pyrolysis with thermally-thick particles, zero-dimensional (0D) models coupled with lumped kinetics are commonly used, as they are simple and computationally efficient. However, intra-particle heat transfer, which cannot be directly implemented in 0D models, has significant effects on pyrolysis behaviors of thermally-thick biomass particles. Additionality, lumped kinetics usually fails to predict detailed composition of pyrolysis products. To overcome these issues, a widely-used one-dimensional (1D) model that can directly incorporate intra-particle heat transfer was employed with a detailed pyrolysis kinetics in this work to develop a corrected 0D (Cor-0D) model for accurate CFD modeling of biomass pyrolysis inside thermally-thick particles. Correction coefficients of external heat transfer, particle diameter, and pyrolysis reactions were introduced by comparing predictions of the 1D model with those of the 0D model quantitatively to reflect the effects of respective factors. The comparison demonstrates that if correction coefficients are properly determined, predictions of the developed Cor-0D model are in good agreement with experimental data as well as those of the employed 1D model under various conditions, while the 0D model overestimates mass loss rate and particle heating rate for thermally-thick biomass particles. Considering that correction coefficients are case dependent and determination of their values are tedious, artificial neural network (ANN) was used to correlate correction coefficients as functions of convective heat transfer coefficient, particle size, gas temperature, moisture content, and particle’s dimensionless temperature to derive an ANN-Cor-0D model. Results show that the ANN-Cor-0D model has the same performance as the Cor-0D model.

Rechargeable zinc-air batteries (ZABs) hold great promise for energy storage and conversion due to their high theoretical energy density, cost-effectiveness, and inherent safety. However, progress is constrained by sluggish oxygen electrocatalysis and instability at the air cathode. To address such issues, we resort to a directly integrated pseudo-3D composite electrocatalyst based on carbon cloth, on which Fe/Fe3C- and N- co-doped carbon nanotubes are directly induced and then used to further intercalate NiFe hydroxide clusters, FexNC@NiFe(OH)(x). This hierarchical electrocatalyst shows enhanced oxygen electrocatalysis (Delta E is 636 mV), rendering high efficiency and durability of ZABs. Such improvement can be attributed to the rationally integrated pseudo-3D structure with high conductivity, high density of active sites, interconnected porosity, and well-bonded components for accelerating electron transfer and ion diffusion while ensuring structural integrity. Moreover, the hierarchical structure increases the electrochemical surface area with superior surface hydrophilicity. As a result, the composite electrocatalyst shows great potential as a binder-free air electrode, as demonstrated in a rechargeable ZAB of a high power density of 85.1 mW cm(-2) and a long period of operation beyond 2000 cycles (350 h) without notable degradation, outperforming noble metal electrodes.

Background: Chemotherapy alone or in conjunction with radiotherapy plays an integral role in all clinical stages of pancreatic cancer (PanCa). Problematically, available chemotherapy drugs, in particular, the oft-used intensive combination regimens, are associated with extensive toxicities and patients often present with debilitating syndromes/symptoms, collectively reducing treatment tolerability, quality of life (QoL), and in turn, survival. Exercise and sportbased interventions are emerging as potential therapies to prevent or alleviate cancer-related and treatment-related sequalae. However, current evidence on physical activity interventions in oncology care is primarily derived from patients with prevalent cancers and better prognoses (e.g., breast cancer). Clinical research in PanCa remains scant and has focused largely on conventional resistance and/or aerobic exercise in patients with potentially curable disease, a relatively younger age and/or higher functioning level. Purpose: This doctoral research aimed to investigate the feasibility, tolerability, and safety of exercise and sport medicine programs in patients with different stages of PanCa, in particular, older individuals with advanced disease during chemo(radio)therapy as well as the efficacy for health-related parameters and QoL. Methods: A series of 4 interrelated studies were undertaken: (1) a systematic review of clinical trials examining the health-related effects of physical activity interventions in PanCa to identify research gaps; (2) an integrative review summarising current evidence of sport-based programs in cancer to inform intervention development; (3) an original analysis to examine the feasibility, tolerability, and safety as well as the health-related effects of exercise at a moderate-intensity to vigorous-intensity in patients with PanCa during chemo(radio)therapy; and (4) a case series to assess the feasibility and initial efficacy of a combined exercise and sport-based intervention

Neural Architecture Search (NAS) has been increasingly appealing to the society of object Re-Identification (ReID), for that task-specific architectures significantly improve the retrieval performance. Previous works explore new optimizing targets and search spaces for NAS ReID, yet they neglect the difference of training schemes between image classification and ReID. In this work, we propose a novel Twins Contrastive Mechanism (TCM) to provide more appropriate supervision for ReID architecture search. TCM reduces the category overlaps between the training and validation data, and assists NAS in simulating real-world ReID training schemes. We then design a Multi-Scale Interaction (MSI) search space to search for rational interaction operations between multi-scale features. In addition, we introduce a Spatial Alignment Module (SAM) to further enhance the attention consistency confronted with images from different sources. Under the proposed NAS scheme, a specific architecture is automatically searched, named as MSINet. Extensive experiments demonstrate that our method surpasses state-of-the-art ReID methods on both in-domain and cross-domain scenarios. Source code available in https://github.com/vimar-gu/MSINet., Comment: Accepted by CVPR2023

Parameter-Efficient Transfer Learning (PETL) aims at efficiently adapting large models pre-trained on massive data to downstream tasks with limited task-specific data. In view of the practicality of PETL, previous works focus on tuning a small set of parameters for each downstream task in an end-to-end manner while rarely considering the task distribution shift issue between the pre-training task and the downstream task. This paper proposes a novel two-stage paradigm, where the pre-trained model is first aligned to the target distribution. Then the task-relevant information is leveraged for effective adaptation. Specifically, the first stage narrows the task distribution shift by tuning the scale and shift in the LayerNorm layers. In the second stage, to efficiently learn the task-relevant information, we propose a Taylor expansion-based importance score to identify task-relevant channels for the downstream task and then only tune such a small portion of channels, making the adaptation to be parameter-efficient. Overall, we present a promising new direction for PETL, and the proposed paradigm achieves state-of-the-art performance on the average accuracy of 19 downstream tasks., Comment: 11 pages

Decentralized policy optimization has been commonly used in cooperative multi-agent tasks. However, since all agents are updating their policies simultaneously, from the perspective of individual agents, the environment is non-stationary, resulting in it being hard to guarantee monotonic policy improvement. To help the policy improvement be stable and monotonic, we propose model-based decentralized policy optimization (MDPO), which incorporates a latent variable function to help construct the transition and reward function from an individual perspective. We theoretically analyze that the policy optimization of MDPO is more stable than model-free decentralized policy optimization. Moreover, due to non-stationarity, the latent variable function is varying and hard to be modeled. We further propose a latent variable prediction method to reduce the error of the latent variable function, which theoretically contributes to the monotonic policy improvement. Empirically, MDPO can indeed obtain superior performance than model-free decentralized policy optimization in a variety of cooperative multi-agent tasks., Comment: 24 pages

Deep supervised learning algorithms typically require a large volume of labeled data to achieve satisfactory performance. However, the process of collecting and labeling such data can be expensive and time-consuming. Self-supervised learning (SSL), a subset of unsupervised learning, aims to learn discriminative features from unlabeled data without relying on human-annotated labels. SSL has garnered significant attention recently, leading to the development of numerous related algorithms. However, there is a dearth of comprehensive studies that elucidate the connections and evolution of different SSL variants. This paper presents a review of diverse SSL methods, encompassing algorithmic aspects, application domains, three key trends, and open research questions. Firstly, we provide a detailed introduction to the motivations behind most SSL algorithms and compare their commonalities and differences. Secondly, we explore representative applications of SSL in domains such as image processing, computer vision, and natural language processing. Lastly, we discuss the three primary trends observed in SSL research and highlight the open questions that remain. A curated collection of valuable resources can be accessed at https://github.com/guijiejie/SSL.

Transformer has been considered the dominating neural architecture in NLP and CV, mostly under supervised settings. Recently, a similar surge of using Transformers has appeared in the domain of reinforcement learning (RL), but it is faced with unique design choices and challenges brought by the nature of RL. However, the evolution of Transformers in RL has not yet been well unraveled. In this paper, we seek to systematically review motivations and progress on using Transformers in RL, provide a taxonomy on existing works, discuss each sub-field, and summarize future prospects., Comment: Accepted by TMLR

Human-centric visual tasks have attracted increasing research attention due to their widespread applications. In this paper, we aim to learn a general human representation from massive unlabeled human images which can benefit downstream human-centric tasks to the maximum extent. We call this method SOLIDER, a Semantic cOntrollable seLf-supervIseD lEaRning framework. Unlike the existing self-supervised learning methods, prior knowledge from human images is utilized in SOLIDER to build pseudo semantic labels and import more semantic information into the learned representation. Meanwhile, we note that different downstream tasks always require different ratios of semantic information and appearance information. For example, human parsing requires more semantic information, while person re-identification needs more appearance information for identification purpose. So a single learned representation cannot fit for all requirements. To solve this problem, SOLIDER introduces a conditional network with a semantic controller. After the model is trained, users can send values to the controller to produce representations with different ratios of semantic information, which can fit different needs of downstream tasks. Finally, SOLIDER is verified on six downstream human-centric visual tasks. It outperforms state of the arts and builds new baselines for these tasks. The code is released in https://github.com/tinyvision/SOLIDER., Comment: accepted by CVPR2023

One of the essential missions in the AI research community is to build an autonomous embodied agent that can attain high-level performance across a wide spectrum of tasks. However, acquiring reward/penalty in all open-ended tasks is unrealistic, making the Reinforcement Learning (RL) training procedure impossible. In this paper, we propose a novel cross-modal contrastive learning framework architecture, CLIP4MC, aiming to learn an RL-friendly vision-language model that serves as a reward function for open-ended tasks. Therefore, no further task-specific reward design is needed. Intuitively, it is more reasonable for the model to address the similarity between the video snippet and the language prompt at both the action and entity levels. To this end, a motion encoder is proposed to capture the motion embeddings across different intervals. The correlation scores are then used to construct the auxiliary reward signal for RL agents. Moreover, we construct a neat YouTube dataset based on the large-scale YouTube database provided by MineDojo. Specifically, two rounds of filtering operations guarantee that the dataset covers enough essential information and that the video-text pair is highly correlated. Empirically, we show that the proposed method achieves better performance on RL tasks compared with baselines.

Cross-lingual cross-modal retrieval has garnered increasing attention recently, which aims to achieve the alignment between vision and target language (V-T) without using any annotated V-T data pairs. Current methods employ machine translation (MT) to construct pseudo-parallel data pairs, which are then used to learn a multi-lingual and multi-modal embedding space that aligns visual and target-language representations. However, the large heterogeneous gap between vision and text, along with the noise present in target language translations, poses significant challenges in effectively aligning their representations. To address these challenges, we propose a general framework, Cross-Lingual to Cross-Modal (CL2CM), which improves the alignment between vision and target language using cross-lingual transfer. This approach allows us to fully leverage the merits of multi-lingual pre-trained models (e.g., mBERT) and the benefits of the same modality structure, i.e., smaller gap, to provide reliable and comprehensive semantic correspondence (knowledge) for the cross-modal network. We evaluate our proposed approach on two multilingual image-text datasets, Multi30K and MSCOCO, and one video-text dataset, VATEX. The results clearly demonstrate the effectiveness of our proposed method and its high potential for large-scale retrieval., Comment: Accepted by AAAI2024

Text-based Person Search (TBPS) aims to retrieve images of target pedestrian indicated by textual descriptions. It is essential for TBPS to extract fine-grained local features and align them crossing modality. Existing methods utilize external tools or heavy cross-modal interaction to achieve explicit alignment of cross-modal fine-grained features, which is inefficient and time-consuming. In this work, we propose a Vision-Guided Semantic-Group Network (VGSG) for text-based person search to extract well-aligned fine-grained visual and textual features. In the proposed VGSG, we develop a Semantic-Group Textual Learning (SGTL) module and a Vision-guided Knowledge Transfer (VGKT) module to extract textual local features under the guidance of visual local clues. In SGTL, in order to obtain the local textual representation, we group textual features from the channel dimension based on the semantic cues of language expression, which encourages similar semantic patterns to be grouped implicitly without external tools. In VGKT, a vision-guided attention is employed to extract visual-related textual features, which are inherently aligned with visual cues and termed vision-guided textual features. Furthermore, we design a relational knowledge transfer, including a vision-language similarity transfer and a class probability transfer, to adaptively propagate information of the vision-guided textual features to semantic-group textual features. With the help of relational knowledge transfer, VGKT is capable of aligning semantic-group textual features with corresponding visual features without external tools and complex pairwise interaction. Experimental results on two challenging benchmarks demonstrate its superiority over state-of-the-art methods., Comment: Accepted to IEEE TIP

Scanning electron microscopy (SEM) is indispensable in diverse applications ranging from microelectronics to food processing because it provides large depth-of-field images with a resolution beyond the optical diffraction limit. However, the technology requires coating conductive films on insulator samples and a vacuum environment. We use deep learning to obtain the mapping relationship between optical super-resolution (OSR) images and SEM domain images, which enables the transformation of OSR images into SEM-like large depth-of-field images. Our custom-built scanning superlens microscopy (SSUM) system, which requires neither coating samples by conductive films nor a vacuum environment, is used to acquire the OSR images with features down to ~80 nm. The peak signal-to-noise ratio (PSNR) and structural similarity index measure values indicate that the deep learning method performs excellently in image-to-image translation, with a PSNR improvement of about 0.74 dB over the optical super-resolution images. The proposed method provides a high level of detail in the reconstructed results, indicating that it has broad applicability to chip-level defect detection, biological sample analysis, forensics, and various other fields., Comment: 13 pages,7 figures

Unsupervised Contrastive learning has gained prominence in fields such as vision, and biology, leveraging predefined positive/negative samples for representation learning. Data augmentation, categorized into hand-designed and model-based methods, has been identified as a crucial component for enhancing contrastive learning. However, hand-designed methods require human expertise in domain-specific data while sometimes distorting the meaning of the data. In contrast, generative model-based approaches usually require supervised or large-scale external data, which has become a bottleneck constraining model training in many domains. To address the problems presented above, this paper proposes DiffAug, a novel unsupervised contrastive learning technique with diffusion mode-based positive data generation. DiffAug consists of a semantic encoder and a conditional diffusion model; the conditional diffusion model generates new positive samples conditioned on the semantic encoding to serve the training of unsupervised contrast learning. With the help of iterative training of the semantic encoder and diffusion model, DiffAug improves the representation ability in an uninterrupted and unsupervised manner. Experimental evaluations show that DiffAug outperforms hand-designed and SOTA model-based augmentation methods on DNA sequence, visual, and bio-feature datasets. The code for review is released at \url{https://github.com/zangzelin/code_diffaug}., Comment: accepted by ICML24

Pre-trained vision transformers have strong representation benefits to various downstream tasks. Recently, many parameter-efficient fine-tuning (PEFT) methods have been proposed, and their experiments demonstrate that tuning only 1\% extra parameters could surpass full fine-tuning in low-data resource scenarios. However, these methods overlook the task-specific information when fine-tuning diverse downstream tasks. In this paper, we propose a simple yet effective method called "Salient Channel Tuning" (SCT) to leverage the task-specific information by forwarding the model with the task images to select partial channels in a feature map that enables us to tune only 1/8 channels leading to significantly lower parameter costs. Experiments on 19 visual transfer learning downstream tasks demonstrate that our SCT outperforms full fine-tuning on 18 out of 19 tasks by adding only 0.11M parameters of the ViT-B, which is 780$\times$ fewer than its full fine-tuning counterpart. Furthermore, experiments on domain generalization and few-shot classification further demonstrate the effectiveness and generic of our approach. The code is available at https://github.com/showlab/SCT., Comment: This work has been accepted by IJCV

Current research on cross-modal retrieval is mostly English-oriented, as the availability of a large number of English-oriented human-labeled vision-language corpora. In order to break the limit of non-English labeled data, cross-lingual cross-modal retrieval (CCR) has attracted increasing attention. Most CCR methods construct pseudo-parallel vision-language corpora via Machine Translation (MT) to achieve cross-lingual transfer. However, the translated sentences from MT are generally imperfect in describing the corresponding visual contents. Improperly assuming the pseudo-parallel data are correctly correlated will make the networks overfit to the noisy correspondence. Therefore, we propose Dual-view Curricular Optimal Transport (DCOT) to learn with noisy correspondence in CCR. In particular, we quantify the confidence of the sample pair correlation with optimal transport theory from both the cross-lingual and cross-modal views, and design dual-view curriculum learning to dynamically model the transportation costs according to the learning stage of the two views. Extensive experiments are conducted on two multilingual image-text datasets and one video-text dataset, and the results demonstrate the effectiveness and robustness of the proposed method. Besides, our proposed method also shows a good expansibility to cross-lingual image-text baselines and a decent generalization on out-of-domain data.

Person Re-identification (ReID) plays a more and more crucial role in recent years with a wide range of applications. Existing ReID methods are suffering from the challenges of misalignment and occlusions, which degrade the performance dramatically. Most methods tackle such challenges by utilizing external tools to locate body parts or exploiting matching strategies. Nevertheless, the inevitable domain gap between the datasets utilized for external tools and the ReID datasets and the complicated matching process make these methods unreliable and sensitive to noises. In this paper, we propose a Region Generation and Assessment Network (RGANet) to effectively and efficiently detect the human body regions and highlight the important regions. In the proposed RGANet, we first devise a Region Generation Module (RGM) which utilizes the pre-trained CLIP to locate the human body regions using semantic prototypes extracted from text descriptions. Learnable prompt is designed to eliminate domain gap between CLIP datasets and ReID datasets. Then, to measure the importance of each generated region, we introduce a Region Assessment Module (RAM) that assigns confidence scores to different regions and reduces the negative impact of the occlusion regions by lower scores. The RAM consists of a discrimination-aware indicator and an invariance-aware indicator, where the former indicates the capability to distinguish from different identities and the latter represents consistency among the images of the same class of human body regions. Extensive experimental results for six widely-used benchmarks including three tasks (occluded, partial, and holistic) demonstrate the superiority of RGANet against state-of-the-art methods.

Unsupervised domain adaptive person re-identification (Re-ID) methods alleviate the burden of data annotation through generating pseudo supervision messages. However, real-world Re-ID systems, with continuously accumulating data streams, simultaneously demand more robust adaptation and anti-forgetting capabilities. Methods based on image rehearsal addresses the forgetting issue with limited extra storage but carry the risk of privacy leakage. In this work, we propose a Color Prompting (CoP) method for data-free continual unsupervised domain adaptive person Re-ID. Specifically, we employ a light-weighted prompter network to fit the color distribution of the current task together with Re-ID training. Then for the incoming new tasks, the learned color distribution serves as color style transfer guidance to transfer the images into past styles. CoP achieves accurate color style recovery for past tasks with adequate data diversity, leading to superior anti-forgetting effects compared with image rehearsal methods. Moreover, CoP demonstrates strong generalization performance for fast adaptation into new domains, given only a small amount of unlabeled images. Extensive experiments demonstrate that after the continual training pipeline the proposed CoP achieves 6.7% and 8.1% average rank-1 improvements over the replay method on seen and unseen domains, respectively. The source code for this work is publicly available in https://github.com/vimar-gu/ColorPromptReID.

Vision Transformers (ViTs) are normally regarded as a stack of transformer layers. In this work, we propose a novel view of ViTs showing that they can be seen as ensemble networks containing multiple parallel paths with different lengths. Specifically, we equivalently transform the traditional cascade of multi-head self-attention (MSA) and feed-forward network (FFN) into three parallel paths in each transformer layer. Then, we utilize the identity connection in our new transformer form and further transform the ViT into an explicit multi-path ensemble network. From the new perspective, these paths perform two functions: the first is to provide the feature for the classifier directly, and the second is to provide the lower-level feature representation for subsequent longer paths. We investigate the influence of each path for the final prediction and discover that some paths even pull down the performance. Therefore, we propose the path pruning and EnsembleScale skills for improvement, which cut out the underperforming paths and re-weight the ensemble components, respectively, to optimize the path combination and make the short paths focus on providing high-quality representation for subsequent paths. We also demonstrate that our path combination strategies can help ViTs go deeper and act as high-pass filters to filter out partial low-frequency signals. To further enhance the representation of paths served for subsequent paths, self-distillation is applied to transfer knowledge from the long paths to the short paths. This work calls for more future research to explain and design ViTs from new perspectives., Comment: Accepted by ICCV 2023, oral presentation

The rapid development of transformative transportation technologies, such as bike-sharing, shared e-scooters, and ride-hailing systems, is reshaping the transportation landscape. These transformative transportation technologies have the potential to significantly change travel behavior and travel demand and affect transportation agencies’ planning, operations, and decision-making. The objective of this project is to develop a framework and models to quantify the potential travel demand and mobility impacts of transformative transportation technologies in Indiana cities. This project analyzed historical system usage data and conducted survey studies to evaluate the availability and use of transformative transportation technologies in select Indiana cities. The project also proposed a data-driven model to study the relationship between shared micro-mobility and the existing transit system and developed a simulation model to analyze the potential mode choice change under different future development scenarios. Additionally, based on a comprehensive literature review, a list of operations; environmental, health and safety; and accessibility and equity metrics were identified as the Key Performance Indicators to evaluate transformative transportation technologies. Furthermore, as this study was conducted in the midst of the COVID-19 pandemic, the impacts of the pandemic on both traditional and transformative transportation systems were also examined as documented in the literature and stated in our survey.

Australia has a reputation for producing a reliable supply of high-quality barley in a contaminant-free climate. As a result, Australian barley is highly sought after by malting, brewing, distilling, and feed industries worldwide. Barley is traded as a variety-specific commodity on the international market for food, brewing and distilling end-use, as the intrinsic quality of the variety determines its market value. Manual identification of barley varieties by the naked eye is challenging and time-consuming for all stakeholders, including growers, grain handlers and traders. Current industrial methods for identifying barley varieties include molecular protein weights or DNA based technology, which are not only time-consuming and costly but need specific laboratory equipment. On grain receival, there is a need for efficient and low-cost solutions for barley classification to ensure accurate and effective variety segregation. This paper proposes an efficient deep learning-based technique that can classify barley varieties from RGB images. Our proposed technique takes only four milliseconds to classify an RGB image. The proposed technique outperforms the baseline method and achieves a barley classification accuracy of 94% across 14 commercial barley varieties (some highly genetically related).

Introduction: Adequate, transparent, and consistent reporting of adverse events (AEs) in exercise oncology trials is critical to assess the safety of exercise interventions for people following a cancer diagnosis. However, there is little understanding of how AEs are reported in exercise oncology trials. Thus, we propose to conduct a scoping review to summarise and evaluate current practice of reporting of AEs in published exercise oncology trials with further exploration of factors associated with inadequate reporting of AEs. The study findings will serve to inform the need for future research on standardisation of the definition, collection, and reporting of AEs for exercise oncology research. Materials and Methods: The ADVANCE (ADverse eVents reporting of clinicAl trials iN exerCise oncology rEsearch) study will be conducted and reported following the PRISMA extension for scoping reviews guideline. Any type of clinical trial involving an exercise intervention in people living with and beyond cancer with a full-text report in English will be included. Six electronic databases (Embase, PubMed, Google Scholar, Web of Science Core Collection, SPORTDiscus, and CINAHL Plus) will be searched for studies. Two independent review authors will assess eligibility of identified studies, chart data using pre-established extraction forms, and evaluate adequacy of reporting of AEs-related data against a 20-item scoring checklist derived from the CONSORT (Consolidated Standards of Reporting Trials) harms extension. We will summarise results using descriptive and inferential analysis methods. Ethics and Dissemination: No ethics approval will be required to conduct the ADVANCE study owing to inclusion of only published data. The study results will be disseminated via publications in peer-reviewed journals and presentations at national and internationa conferences. Systematic Review Registration: Open Science Framework: https://osf.io/NXEJD/ (doi:10.17605/OSF.IO/NXEJD).

Objectives: The aim was to characterize multimorbidity patterns in a large sample of older individuals living in nursing homes (NHs) and to investigate their association with mortality, also considering the effect of functional status. Design: Observational and retrospective study. Setting and Participants: We analyzed data on 4131 NH residents in Italy, aged 60 years and older, assessed through the interRAI long-term care facility instrument. Entry date was between 2014 and 2018, and participants were followed until 2019. Methods: Multimorbidity patterns were identified through principal component analysis; for the identified components, subjects were stratified in quintiles (Q) with respect to their loading values, with the higher quantiles indicating greater expression of the component's pattern. Their association [hazard ratio (HR) and 95% CI] with mortality was tested in Cox regression models. Analyses were stratified by disability status. Results: Four patterns of multimorbidity were identified: (1) heart diseases; (2) dementia and sensory impairments; (3) heart, respiratory, and psychiatric diseases; and (4) diabetes, musculoskeletal, and vascular diseases. For the heart diseases pattern [HR Q5 vs Q1 = 1.83 (1.53–2.20)] and the dementia and sensory impairments pattern [HR Q5 vs Q1 = 1.23 (1.06–1.42)], as the specific multimorbidity expression increases, the risk of mortality increases. On stratifying by disability status, the association between the multimorbidity patterns and mortality was not always present. Conclusions and Implications: Different multimorbidity patterns are differentially associated with mortality in older residents of NHs, confirming that multimorbidity's prognosis is strictly dependent on the underlying disease combinations. This knowledge may be useful to implement personalized preventive and therapeutic care pathways for institutionalized older adults, which respond to individuals’ health needs.

Objectives: Evidence about prescribing patterns of dementia medication in China is lacking. This study aimed to examine prescribing rates of antidementia and psychotropic drugs and factors associated with drug prescription for dementia in China. Design: A multicenter observational study. Setting and Participants: This study employed cross-sectional data from the Clinical Pathway for Alzheimer's Disease in China study that was conducted in 28 memory clinics at tertiary hospitals across 14 provinces between 2012 and 2013. Patients aged ≥45 years with a diagnosis of dementia were included. Methods: Antidementia and psychotropic drugs were classified according to the Anatomical Therapeutic Chemical codes. Odds ratios (ORs) of putative factors associated with prescription patterns were estimated using logistic regressions. Results: A total of 751 respondents were included in this study, 77.8% of whom were prescribed antidementia drugs, and 33.0% were prescribed at least 1 psychotropic drug. The concomitant prescription rate of antidementia and psychotropic drugs was 24.1%. Frontotemporal dementia [OR 9.92 (99.17% CI 3.08-42.70)], severe dementia [4.25 (1.88-9.79)], and apathy [1.94 (1.18-3.20)] were significantly associated with an elevated likelihood of memantine prescription. Psychotic symptoms [1.84 (1.02-3.35)], agitation [1.91 (1.08-3.40)], and depressive symptoms [2.10 (1.12-3.94)] were significantly associated with the coprescription of antidementia and psychotropic agents. Conclusions and Implications: The prescribing rate of antidementia drugs in the study sample was higher, whereas the rate of coprescription of psychotropic and antidementia drugs was lower than reported in Western studies. Dementia prescription practice was generally consistent with clinical guidelines in memory clinics in China, whereas the prescription of antidementia and psychotropic medication mainly depended on patients’ clinical symptoms.

Hongyu Chen,1 Jianxiang Liu,2 Qian Kang,1 Hao Luo,1 Ning Tan,1 Jiali Pan,1 Yuqing Yang,1 Min Yu,1 Dan Liu,1 Hongli Xi,1 Yifan Han,1 Ran Cheng,1 Yanyan Yu,1 Xiaoyuan Xu1 1Departments of Infectious Diseases, Peking University First Hospital, Beijing, People’s Republic of China; 2Department of Gastroenterology, Peking University First Hospital, Beijing, People’s Republic of ChinaCorrespondence: Xiaoyuan Xu; Yanyan Yu, Email xiaoyuanxu6@163.com; yyy@bjmu.edu.cnBackground: Since oral direct-acting antiviral agents (DAAs) became available, the global hepatitis C treatment situation has undergone tremendous changes. However there are still many issues worthy of attention in treatment.Methods: We selected 53 HCV-infected patients who were treated and followed up in the Peking University First Hospital from December 2017 to January 2021 to detect the RASs in HCV. Pearson correlation analysis was used to analyze HCV RNA and HCV cAg, the Fisher exact test and chi-square test was used to compare the effects of RASs on the rate of decline of HCV RNA and HCV core antigen (cAg) during DAA treatment.Results: The RASs and its prevalence on the NS3 are mainly Y56F 2.56% (1/39), Q80K 23.08% (9/39), S122G 71.79% (28/39), and V170I 38.46% (15/39). On the NS5A were R30Q 10.53% (4/38), P32A 5.26% (2/38), P58S 2.63% (1/39), and Y93H 21.05% (8/38). On NS5B were C316N 71.05% (27/38), C451H 2.63% (1/38), and I585C 2.63% (1/38). There was no significant correlation between the RASs (Y93H, V179I, Q80K, S122G, C316N) and HCV genotype (p > 0.05). The baseline serum HCV RNA and HCV cAg had a significant medium-degree correlation (r = 0.601, p = 0.002). After 1 week of DAA treatment was weak correlation (r = 0.413, p = 0.032). Q80K, S122G, V170I, Y93H, and C316N had no effect on the clearance of HCV RNA and HCV cAg within the first week of DAA treatment (p> 0.05).Conclusion: The HCV genotype may have a limited impact on the presence of the five RASs (Y93H, V179I, Q80K, S122G, and C316N) a

RATIONALE: Vascular smooth muscle cells (VSMCs) phenotype switch from contractile to proliferative phenotype is a pathological hallmark in various cardiovascular diseases. Recently, a subset of long noncoding RNAs was identified to produce functional polypeptides. However, the functional impact and regulatory mechanisms of long noncoding RNAs in VSMCs phenotype switching remain to be fully elucidated. OBJECTIVES: To illustrate the biological function and mechanism of a VSMC-enriched long noncoding RNA and its encoded peptide in VSMC phenotype switching and vascular remodeling. RESULTS: We identified a VSMC-enriched transcript encoded by a previously uncharacterized gene, which we called phenotype switching regulator (PSR), which was markedly upregulated during vascular remodeling. Although PSR was annotated as a long noncoding RNA, we demonstrated that the lncPSR (PSR transcript) also encoded a protein, which we named arteridin. In VSMCs, both arteridin and lncPSR were necessary and sufficient to induce phenotype switching. Mechanistically, arteridin and lncPSR regulate downstream genes by directly interacting with a transcription factor YBX1 (Y-box binding protein 1) and modulating its nuclear translocation and chromatin targeting. Intriguingly, the PSR transcription was also robustly induced by arteridin. More importantly, the loss of PSR gene or arteridin protein significantly attenuated the vascular remodeling induced by carotid arterial injury. In addition, VSMC-specific inhibition of lncPSR using adeno-associated virus attenuated Ang II (angiotensin II)-induced hypertensive vascular remodeling. CONCLUSIONS: PSR is a VSMC-enriched gene, and its transcript IncPSR and encoded protein (arteridin) coordinately regulate transcriptional reprogramming through a shared interacting partner, YBX1. This is a previously uncharacterized regulatory circuit in VSMC phenotype switching during vascular remodeling, with lncPSR/arteridin as potential therapeutic targets for the t