Scientific Reports最新文献

To explore whether ultra-sensitive circulating tumor DNA (ctDNA) profiling enables earlier prediction of treatment response and detection of disease progression, we applied NeXT Personal, an ultra-sensitive bespoke tumor-informed liquid biopsy platform, to profile tumor samples from the KeyLargo study, a phase II trial in which metastatic esophagogastric cancer (mEGC) patients received capecitabine, oxaliplatin, and pembrolizumab. A total of 24 patients were evaluated, and all were ctDNA-positive at baseline. ctDNA levels varied from 406,067 down to 1.5 parts per million (PPM) with a median limit of detection of 2.03 PPM. ctDNA dynamics were highly correlated with changes in tumor size (ρ = 0.59, p = 7.3 × 10^- 9). Lack of early molecular response (50% or greater decrease in ctDNA levels at first available time point after 30 days, C2D1 or C4D1) was associated with worse overall survival (OS) (HR 4.5, 95% CI 1.2-16.7, p = 0.02) and progression-free survival (PFS) (HR 10.4, 95% CI 2.2-49.8, p = 0.003). Lack of molecular clearance of ctDNA was associated with worse OS (HR 7.1, 95% CI 1.6-31.7, p = 0.01) and PFS (HR 19.9, 95% CI 2.5-158.2, p = 0.005). Molecular progression (ctDNA increase) preceded imaging-derived progression by a median lead time of 65 days. These results suggest that ultra-sensitive liquid biopsy approaches could improve treatment decision-making for mEGC patients receiving chemotherapy and immunotherapy.

Post-grouting is an effective engineering technique for enhancing the bearing performance of pile foundations, and the grouting material is a key factor determining the performance of post-grouted piles. In this study, different grouting materials (fly ash-cement, geopolymer, cement, and cement-sodium silicate) were adopted in laboratory tests. Model tests of post-grouted piles were conducted under simulated groundwater flow conditions and long-term loading scenarios to investigate the influence of various grouting materials on pile bearing performance under vertical loading. The research focuses on the characteristics of load- settlement curves, the axial force transfer along the pile, and the mobilization behavior of side resistance and tip resistance, and employs scanning electron microscopy (SEM) tests to reveal the action mechanisms of different grouting materials. The test results indicate that the grouting process significantly enhances the pile-soil interaction. Compared with the ungrouted piles, the bearing capacity of the grouted piles increases markedly. Under flowing-water conditions, the cement-sodium silicate double-liquid grout provides the best reinforcement effect around the pile due to its short gel time, rapid solidification, and dense hydration products, whereas the fly ash-cement and geopolymer grouts mainly improve the soil via permeation grouting, with the hydration reaction of fly ash proceeding relatively slowly.

Barrier-involved crashes are shaped by evolving behavioral patterns, roadway conditions, and systemic changes in transportation environments. This study investigates temporal shifts in the factors influencing injury severity in barrier-related traffic crashes over three multi-year periods: 2017-2018, 2019-2020, and 2021-2022. A Multinomial Logit (MNL) model and its extensions, including the Random Parameter Logit (RPL), Random Parameter Logit with Heterogeneity in Means (RPLHM), Correlated Random Parameter Logit with Heterogeneity in Means (CRPLHM), and Random Parameter Logit with Heterogeneity in Means and Variance (RPLHMV), are employed to capture unobserved heterogeneity and temporal changes in predictor effects. The analysis reveals that contextual factors such as roadway alignment, posted speed limits, weather, lighting, and vehicle type exhibit time-varying associations with crash outcomes. Moderate speed zones (35-55 mph), previously not significant, became increasingly associated with lower severity in later periods. Meanwhile, the influence of lane departure, marked lanes, positive median barriers, and collisions with fixed objects also varied across periods. Notably, crashes outside intersections that initially corresponded with lower injury risk began to show elevated severity in the later years. Driver failure to maintain a single lane consistently emerged as a strong predictor of severity, with its effect intensifying over time. These findings highlight the importance of capturing temporal instability in crash determinants and support the need for adaptive, evidence-driven safety policies that reflect shifting roadway and behavioral dynamics.

Traceability is a key requirement for compliance in food supply chains, ensuring transparency, safety, and product authenticity across all stages of production and distribution. Although advanced technologies have been introduced to support compliance efforts, blockchain stands out for its potential to provide secure, tamper-proof, and verifiable traceability. However, adoption of blockchain remains limited due to technical complexity, high implementation costs, and fragmented integration between supply chains. To address these challenges, this study proposes a unified blockchain-enabled services framework for the implementation of streamlined, standardized, and accessible traceability. The framework includes a network architecture, traceability-driven operational rules, a UML sequence diagram defining system interactions, and algorithms to record and verify data. Demonstrated through an Australian seafood traceability scenario, the framework was validated by deploying a prototype digital platform and smart contracts on the blockchain. The results show that the framework enables reliable traceability with minimal integration effort while supporting interoperability and data integrity. Although not required, the study highlights the potential role of regulatory bodies as service providers to strengthen compliance alignment. In general, this approach presents a scalable, cost-effective, and practical solution to improve traceability in supply chains through blockchain technology.

Multimodal large language models (MLLMs) have achieved impressive performance in understanding and describing visual content, setting new state-of-the-art results on a variety of visual question answering (VQA) benchmarks. However, during decoding, these models often fail to attend to fine-grained visual details in the input image. Our analysis of intermediate attention layers reveals that MLLMs are not inherently incapable of perceiving target objects; rather, attention to visual details becomes diluted in deeper layers due to the dominance of language priors. To address this limitation, we propose a plug-and-play attention re-alignment module (ARA) that enhances suppressed visual grounding. ARA conducts a layer-wise analysis of the relative attention distribution of image-centric attention heads. It incorporates a confidence-aware layer selection mechanism based on attention peak and entropy, enabling the dynamic aggregation of attention maps from the most informative layers. These aggregated maps are subsequently leveraged to guide the generation of semantic masks, enabling the model to emphasize salient visual regions while suppressing irrelevant or noisy content. ARA can be seamlessly integrated into existing MLLMs and demonstrates consistent improvements across multiple VQA benchmarks, validating its effectiveness in enhancing visual detail sensitivity.