Prompt-and-Transfer: Dynamic Class-aware Enhancement for Few-shot Segmentation.

For more efficient generalization to unseen domains (classes), most Few-shot Segmentation (FSS) would directly exploit pre-trained encoders and only fine-tune the decoder, especially in the current era of large models. However, such fixed feature encoders tend to be class-agnostic, inevitably activating objects that are irrelevant to the target class. In contrast, humans can effortlessly focus on specific objects in the line of sight. This paper mimics the visual perception pattern of human beings and proposes a novel and powerful prompt-driven scheme, called "Prompt and Transfer" (PAT), which constructs a dynamic class-aware prompting paradigm to tune the encoder for focusing on the interested object (target class) in the current task. Three key points are elaborated to enhance the prompting: 1) Cross-modal linguistic information is introduced to initialize prompts for each task. 2) Semantic Prompt Transfer (SPT) that precisely transfers the class-specific semantics within the images to prompts. 3) Part Mask Generator (PMG) that works in conjunction with SPT to adaptively generate different but complementary part prompts for different individuals. Surprisingly, PAT achieves competitive performance on 4 different tasks including standard FSS, Cross-domain FSS (e.g., CV, medical, and remote sensing domains), Weak-label FSS, and Zero-shot Segmentation, setting new state-of-the-arts on 11 benchmarks.

An App Platform-Facilitated Collaborative Palliative Care Intervention for Outpatients With Interstitial Lung Disease: A Pilot Randomized Trial.

Rationale: Outpatients with interstitial lung disease often experience serious symptoms, yet infrequently receive palliative care. Objective: To determine the feasibility and clinical impact of a mobile application (PCplanner) in an outpatient setting. Methods: We conducted a pilot randomized controlled trial among adults with interstitial lung disease in a single-center academic clinic. Clinical outcomes included change in Needs at the End-of-Life Screening Tool (NEST) scale between baseline and 3 months as well as frequency of advance care planning discussions and referrals to palliative care services. Results: Observed feasibility outcomes were similar to targeted benchmarks including randomization rates (82.1% vs 80%) and retention (84.8% vs 80%). Mean NEST scores between the intervention and control group were 38.9 (SD, 18.9) vs 41.5 (SD, 20.5) at baseline, 34.6 (SD, 18.9) vs 33.6 (SD, 19.4) at 1 month after clinic visit, 40.5 (SD, 21.6) vs 35.3 (SD, 25.0) at 3 months after clinic visit. Changes in NEST scores between baseline and 3 months showed no difference in the primary outcome (P = 0.481, 95% CI [-8.45, 17.62]). Conclusion: Among patients with interstitial lung disease, a mobile app designed to focus patients and clinicians on palliative care principles demonstrated evidence of feasibility. Although changes in self-reported needs were similar between intervention and control groups, more patients in the intervention group updated their advance directives and code status compared to the control group. Clinical Trial Registration: Palliative Care Planner (PCplanner) NCT05095363. https://www.clinicaltrials.gov/study/NCT05095363.

Remote Sensing Change Detection With Bitemporal and Differential Feature Interactive Perception.

Recently, the transformer has achieved notable success in remote sensing (RS) change detection (CD). Its outstanding long-distance modeling ability can effectively recognize the change of interest (CoI). However, in order to obtain the precise pixel-level change regions, many methods directly integrate the stacked transformer blocks into the UNet-style structure, which causes the high computation costs. Besides, the existing methods generally consider bitemporal or differential features separately, which makes the utilization of ground semantic information still insufficient. In this paper, we propose the multiscale dual-space interactive perception network (MDIPNet) to fill these two gaps. On the one hand, we simplify the stacked multi-head transformer blocks into the single-layer single-head attention module and further introduce the lightweight parallel fusion module (LPFM) to perform the efficient information integration. On the other hand, based on the simplified attention mechanism, we propose the cross-space perception module (CSPM) to connect the bitemporal and differential feature spaces, which can help our model suppress the pseudo changes and mine the more abundant semantic consistency of CoI. Extensive experiment results on three challenging datasets and one urban expansion scene indicate that compared with the mainstream CD methods, our MDIPNet obtains the state-of-the-art (SOTA) performance while further controlling the computation costs.

Osteoking prevents bone loss and enhances osteoblastic bone formation by modulating the AGEs/IGF-1/ß-catenin/OPG pathway in type 2 diabetic db/db mice.

Type 2 diabetic osteoporosis (T2DOP) is a skeletal metabolic syndrome characterized by impaired bone remodeling due to type 2 diabetes mellitus, and there are drawbacks in the present treatment. Osteoking (OK) is widely used for treating fractures and femoral head necrosis. However, OK is seldom reported in the field of T2DOP, and its role and mechanism of action need to be elucidated. Consequently, this study investigated whether OK improves bone remodeling and the mechanisms of diabetes-induced injury. We used db/db mice as a T2DOP model and stimulated MC3T3-E1 cells (osteoblast cell line) with high glucose (HG, 50 mM) and advanced glycation end products (AGEs, 100 µg/mL), respectively. The effect of OK on T2DOP was assessed using a combined 3-point mechanical bending test, hematoxylin and eosin staining, and enzyme-linked immunosorbent assay. The effect of OK on enhancing MC3T3-E1 cell differentiation and mineralization under HG and AGEs conditions was assessed by an alkaline phosphatase activity assay and alizarin red S staining. The AGEs/insulin-like growth factor-1(IGF-1)/ß-catenin/osteoprotegerin (OPG) pathway-associated protein levels were assayed by western blot analysis and immunohistochemical staining. We found that OK reduced hyperglycemia, attenuated bone damage, repaired bone remodeling, increased tibial and femoral IGF-1, ß-catenin, and OPG expression, and decreased receptor activator of nuclear kappa B ligand and receptor activator of nuclear kappa B expression in db/db mice. Moreover, OK promoted the differentiation and mineralization of MC3T3-E1 cells under HG and AGEs conditions, respectively, and regulated the levels of AGEs/IGF-1/ß-catenin/OPG pathway-associated proteins. In conclusion, our results suggest that OK may lower blood glucose, alleviate bone damage, and attenuate T2DOP, in part through activation of the AGEs/IGF-1/ß-catenin/OPG pathway.

The use of real-world data for clinical investigation of effectiveness in drug development.

With the growing interest in leveraging real-world data (RWD) to support effectiveness evaluations for new indications, new target populations, and post-market performance, the United States Food and Drug Administration has published several guidance documents on RWD sources and real-world studies (RWS) to assist sponsors in generating credible real-world evidence (RWE). Meanwhile, the randomized controlled trial (RCT) remains the gold standard in drug evaluation. Along this line, we propose a hybrid two-stage adaptive design to evaluate effectiveness based on evidence from both RCT and RWS. At the first stage, a typical non-inferiority test is conducted using RCT data to test for not-ineffectiveness. Once not-ineffectiveness is established, the study proceeds to the second stage to conduct an RWS and test for effectiveness using integrated information from RCT and RWD. The composite likelihood approach is implemented as a down-weighing strategy to account for the impact of high variability in RWS population. An optimal sample size determination procedure for RCT and RWS is introduced, aiming to achieve the minimal expected sample size. Through extensive numerical study, the proposed design demonstrates the ability to control type I error inflation in most cases and consistently maintain statistical power above the desired level. In general, this RCT/RWS hybrid two-stage adaptive design is beneficial for effectiveness evaluations in drug development, especially for oncology and rare diseases.

On the use of RWD in support of regulatory submission in drug development.

For the approval of a drug product, the United States Food and Drug Administration requires substantial evidence (SE) regarding effectiveness and safety of the test drug to be provided. In recent years, the use of real-world data in support of regulatory submission of pharmaceutical development has received much attention, and real-world evidence (RWE) is treated as complementary to SE by evaluating the real-world performance of the test treatment. In this article, we start by summarizing current regulatory perspectives on drug evaluation and some potential challenges in using RWE. To test for superiority in co-primary endpoints, a two-stage hybrid RCT/RWS adaptive design that combines randomized control trial for providing SE and real-world study for generating RWE is proposed. We use superiority in effectiveness and non-inferiority in safety as an example to illustrate how to implement this design. Numerical studies have shown that the proposed design has merits in reducing the required sample size compared with traditional co-primary endpoint tests while maintaining statistical power and controlling type I error inflation. The proposed design can be implemented in drug development considering co-primary endpoints, especially for oncology and rare disease drug development.

Chinese Title Generation for Short Videos: Dataset, Metric and Algorithm.

Previous work for video captioning aims to objectively describe the video content but the captions lack human interest and attractiveness, limiting its practical application scenarios. The intention of video title generation (video titling) is to produce attractive titles, but there is a lack of benchmarks. This work offers CREATE, the first large-scale Chinese shoRt vidEo retrievAl and Title gEneration dataset, to assist research and applications in video titling, video captioning, and video retrieval in Chinese. CREATE comprises a high-quality labeled 210 K dataset and two web-scale 3 M and 10 M pre-training datasets, covering 51 categories, 50K+ tags, 537K+ manually annotated titles and captions, and 10M+ short videos with original video information. This work presents ACTEr, a unique Attractiveness-Consensus-based Title Evaluation, to objectively evaluate the quality of video title generation. This metric measures the semantic correlation between the candidate (model-generated title) and references (manual-labeled titles) and introduces attractive consensus weights to assess the attractiveness and relevance of the video title. Accordingly, this work proposes a novel multi-modal ALignment WIth Generation model, ALWIG, as one strong baseline to aid future model development. With the help of a tag-driven video-text alignment module and a GPT-based generation module, this model achieves video titling, captioning, and retrieval simultaneously. We believe that the release of the CREATE dataset, ACTEr metric, and ALWIG model will encourage in-depth research on the analysis and creation of Chinese short videos.

Unmanned Aerial Vehicle-Based Compressed Data Acquisition for Environmental Monitoring in WSNs.

With the increasing concerns for the environment, the amount of the data monitored by wireless sensor networks (WSNs) is becoming larger and the energy required for data transmission is greater. However, sensor nodes have limited storage capacity and battery power. The WSNs are faced with the challenge of handling larger data volumes while minimizing energy consumption for transmission. To address this issue, this paper employs data compression technology to eliminate redundant information in the environmental data, thereby reducing energy consumption of sensor nodes. Additionally, an unmanned aerial vehicle (UAV)-assisted compressed data acquisition algorithm is put forward. In this algorithm, compressive sensing (CS) is introduced to decrease the amount of data in the network and the UAV serves as a mobile aerial base station for efficient data gathering. Based on CS theory, the UAV selectively collects measurements from a subset of sensor nodes along a route planned using the optimized greedy algorithm with variation and insertion strategies. Once the UAV returns, the sink node reconstructs sensory data from these measurements using the reconstruction algorithms. Extensive experiments are conducted to verify the performance of this algorithm. Experimental results show that the proposed algorithm has lower energy consumption compared to other approaches. Furthermore, we employ different data reconstruction algorithms to recover data and discover that the data can be better reconstructed in a shorter time.

Innovative thinking of clinical investigation for rare disease drug development.

For the development of a test treatment or drug product, it is necessary to conduct composite hypothesis testing to test for effectiveness and safety simultaneously, since some approved drug products have been recalled due to safety concerns. One of the major issues in conducting a composite hypothesis testing for effectiveness and safety is the requirement of a huge sample size to achieve the desired power for detecting clinically meaningful differences in both safety and effectiveness. Situation can be much difficult in orphan drug development. In this article, a generalized two-stage innovative approach to test for effectiveness and safety simultaneously is proposed. Additionally, to alleviate the requirement of a large randomized clinical trial (RCT) and revealing effectiveness, real-world data is suggested to use in conjunction with RCT data for orphan drug development. The proposed approach can help investigators test for effectiveness and safety at the same time without worrying about the sample size. It also helps reduce the probability of approving a drug product with safety concerns.

Mechanism of cognitive impairment induced by d-galactose and l-glutamate through gut-brain interaction in tree shrews.

d-Galactose (d-gal) and l-glutamate (l-glu) impair learning and memory. The mechanism of interaction between the gut microbiome and brain remains unclear. In this study, a model of cognitive impairment was induced in tree shrews by intraperitoneal (ip) injection of d-gal (600 mg/kg/day), intragastric (ig) administration with l-glu (2000 mg/kg/day), and the combination of d-gal (ip, 600 mg/kg/day) and l-glu (ig, 2000 mg/kg/day). The cognitive function of tree shrews was tested by the Morris water maze method. The expression of Aß1-42 proteins, the intestinal barrier function proteins occludin and P-glycoprotein (P-gp), and the inflammatory factors NF-κB, TLR2, and IL-18 was determined by immunohistochemistry. The gut microbiome was analyzed by 16SrRNA high-throughput sequencing. After administering d-gal and l-glu, the escape latency increased (p < .01), and the times of crossing the platform decreased (p < .01). These changes were greater in the combined administration of d-gal and l-glu (p < .01). The expression of Aß1-42 was higher in the perinuclear region of the cerebral cortex (p < .01) and intestinal cell (p < .05). There was a positive correlation between the cerebral cortex and intestinal tissue. Moreover, the expression of NF-κB, TLR2, IL-18, and P-gp was higher in the intestine (p < .05), while the expression of occludin and the diversity of gut microbes were lower, which altered the biological barrier of intestinal mucosal cells. This study indicated that d-gal and l-glu could induce cognitive impairment, increase the expression of Aß1-42 in the cerebral cortex and intestinal tissue, decrease the gut microbial diversity, and alter the expression of inflammatory factors in the mucosal intestines. The dysbacteriosis may produce inflammatory cytokines to modulate neurotransmission, causing the pathogenesis of cognitive impairment. This study provides a theoretical basis to explore the mechanism of learning and memory impairment through the interaction of microbes in the gut and the brain.