Search Results (1,755)

Soybean–corn intercropping in the hilly–mountainous regions of Southwest China poses unique challenges to mechanized harvesting because of complex topography and agronomic constraints. Addressing the soybean-harvesting bottleneck in these fields requires advanced sensing and perception rather than purely mechanical redesigns. Prior reviews emphasized flat-terrain machinery or single-crop systems, leaving a gap in sensor-centric solutions for intercropping on steep, irregular plots. This review analyzes how sensors enable the next generation of intelligent harvesters by linking field constraints to perception and control. We frame the core failures of conventional machines—instability, inconsistent cutting, and low efficiency—as perception problems driven by low pod height, severe slope effects, and header–row mismatches. From this perspective, we highlight five fronts: (1) terrain-profiling sensors integrated with adaptive headers; (2) IMUs and inclination sensors for chassis stability and traction on slopes; (3) multi-sensor fusion of LiDAR and machine vision with AI for crop identification, navigation, and obstacle avoidance; (4) vision and spectral sensing for selective harvesting and impurity pre-sorting; and (5) acoustic/vibration sensing for low-damage, high-efficiency threshing and cleaning. We conclude that compact, intelligent machinery powered by sensing, data fusion, and real-time control is essential, while acknowledging technological and socio-economic barriers to deployment. This review outlines a sensor-driven roadmap for sustainable, efficient soybean harvesting in challenging terrains. Full article

Generating realistic and diverse driving scenarios is essential for effective scenario-based testing and validation in autonomous driving and the development of driver assistance systems. Traditionally, parametric models are used as standard approaches for scenario generation, but they require detailed domain expertise, suffer from scalability issues, and often introduce biases due to idealizations. Recent research has demonstrated that AI models can generate more realistic driving scenarios with reduced manual effort. However, these models typically focused on single scenario types, such as cut-in maneuvers, which limits their applicability to diverse real-world driving situations. This paper, therefore, proposes a unified generative framework that can simultaneously generate multiple types of driving scenarios, including cut-in, cut-out, and cut-through maneuvers from both directions, thus covering six distinct driving behaviors. The model not only learns to generate realistic trajectories but also reflects the same statistical properties as observed in real-world data, which is essential for risk assessment. Comprehensive evaluations, including quantitative metrics and visualizations from detailed latent and physical space analyses, demonstrate that the unified model achieves comparable performance to individually trained models. The shown approach reduces modeling complexity and offers a scalable solution for generating diverse, safety-relevant driving scenarios, supporting robust testing and validation. Full article

Colorectal cancer (CRC) remains a leading cause of cancer-related morbidity and mortality worldwide, with patient outcomes highly dependent on early and accurate diagnosis. However, existing diagnostic methods, such as colonoscopy, fecal occult blood testing, and imaging, are often invasive, costly, or lack sufficient sensitivity and specificity, particularly in early-stage disease. In this context, aptamers, which are synthetic single-stranded oligonucleotides capable of binding to specific targets with high affinity, have emerged as a powerful alternative to antibodies for biosensing applications. This review provides a comprehensive overview of aptamer-based strategies for CRC detection, spanning from biomarker discovery to clinical translation. We first examine established and emerging CRC biomarkers, including those approved by regulatory agencies, described in patents, and shared across multiple cancer types. We then discuss recent advances in aptamer selection and design, with a focus on SELEX variants and in silico optimization approaches tailored to CRC-relevant targets. The integration of aptamers into cutting-edge sensing platforms, such as electrochemical, optical, and nanomaterial-enhanced aptasensors, is highlighted, with emphasis on recent innovations that enhance sensitivity, portability, and multiplexing capabilities. Furthermore, we explore the convergence of aptasensing with microfluidics, and wearable technologies to enable intelligent, miniaturized diagnostic systems. Finally, we consider the clinical and regulatory pathways for point-of-care implementation, as well as current challenges and opportunities for advancing the field. By outlining the technological and translational trajectory of aptamer-based CRC diagnostics, this review aims to provide a roadmap for future research and interdisciplinary collaboration in precision oncology. Full article

With the acceleration of urbanization, environmental degradation is increasingly restricting the improvement of residents’ quality of life, and promoting the transformation of old communities has become a key path for sustainable urban development. However, existing buildings generally face challenges, such as the deterioration of the performance of the envelope structure and the rising energy consumption of the air conditioning system, which pose a serious test for the realization of green renovation. Inspired by the application of bionics in the field of architecture, this study innovatively designed five types of bionic envelope structures for outdoor air conditioning units, namely scales, honeycombs, spider webs, leaves, and bird nests, based on the aerodynamic characteristics of biological prototypes. The ventilation performance of these structures was evaluated at three scales—namely, single building, townhouse, and community—under natural ventilation conditions, using a CFD simulation system. The study shows the following: (1) the spider web structure has the best comprehensive performance among all types of enclosures, which can significantly improve the uniformity of the flow field and effectively eliminate the low-speed stagnation area on the windward side; (2) the structure reorganizes the flow structure of the near-wall area through the cutting and diversion of the porous grid, reduces the wake range, and weakens the negative pressure intensity, making the pressure distribution around the building more balanced; (3) in the height range of 1.5–27 m, the spider web structure performs particularly well at the townhouse and community scales, with an average wind speed increase of 1.1–1.4%; and (4) the design takes into account both the safety of the enclosure and the comfort of the pedestrian area, achieving a synergistic optimization of function and performance. This study provides new ideas for the micro-renewal of buildings, based on bionic principles, and has theoretical and practical value for improving the wind environment quality of old communities and promoting low-carbon urban development. Full article

Background/Objectives: Artificial Intelligence (AI) is rapidly advancing in medicine, facilitating personalized care by leveraging complex clinical data, imaging, and patient monitoring. This study characterizes current practices in AI use within oncology clinical trials by analyzing completed U.S. trials within the Cancer Control Continuum (CCC), a framework that spans the stages of cancer etiology, prevention, detection, diagnosis, treatment, and survivorship. Methods: This cross-sectional study analyzed U.S.-based oncology trials registered on ClinicalTrials.gov between January 2015 and April 2025. Using AI-related MeSH terms, we identified trials addressing stages of the CCC. Results: Fifty completed oncology trials involving AI were identified; 66% were interventional and 34% observational. Machine Learning was the most common AI application, though specific algorithm details were often lacking. Other AI domains included Natural Language Processing, Computer Vision, and Integrated Systems. Most trials were single-center with limited participant enrollment. Few published results or reported outcomes, indicating notable reporting gaps. Conclusions: This analysis of ClinicalTrials.gov reveals a dynamic and innovative landscape of AI applications transforming oncology care, from cutting-edge Machine Learning models enhancing early cancer detection to intelligent chatbots supporting treatment adherence and personalized survivorship interventions. These trials highlight AI’s growing role in improving outcomes across the CCC in advancing personalized cancer care. Standardized reporting and enhanced data sharing will be important for facilitating the broader application of trial findings, accelerating the development and clinical integration of reliable AI tools to advance cancer care. Full article

Hybrid natural/synthetic fiber laminates were examined as a practical process to cut mass, reduce material footprint, and meet structural demands while addressing sustainability targets. Yet direct, like-for-like comparisons generated under a single process and accompanied by durability measurements were limited, leaving design choices uncertain. This study aimed to fabricate and benchmark five representative laminates—C1: flax/epoxy, C2: jute/glass/epoxy, C3: hemp/carbon/epoxy, C4: flax/glass/bio-epoxy, and C5: kenaf/basalt/polyester—under a controlled hot-press schedule with a fixed cavity and verified fiber volume fraction. Panels were characterized using ASTM D3039 tension, ASTM D790 flexure, instrumented impact, 168 h water immersion, and thermogravimetric mass retention. The results were normalized to enable direct multi-criteria comparison, and a model was calibrated to predict tensile strength. C3 delivered the highest strengths (tension ≈ 120 MPa; flexure ≈ 126 MPa), while C5 showed the greatest impact capacity (≈60 kJ/m²). End-of-test water uptake at 168 h was C1 ≈ 3.4%, C2 ≈ 2.6%, C3 ≈ 1.4%, C4 ≈ 2.1%, and C5 ≈ 2.3%. The tensile predictor was fitted to panel means, with an R² of 0.988, and maintained an R² of 0.96 under leave-one-configuration-out testing. These results indicated that carbon-containing hybrids played the most critical roles in terms of stiffness, with kenaf/basalt being most suitable for stiffness-critical components at a similar density, and flax/glass with a bio-resin maximized the sustainability score while maintaining adequate strength. Future research should focus on enhancing specific strength at high renewable content through interface treatments, and extended modeling to improve flexure and impact responses. Full article

The growing demand for air connectivity, coupled with the forecasted increase in passengers by 2040, implies an exigency in the aviation sector to adopt sustainable approaches for net zero emission by 2050. Sustainable Aviation Fuel (SAF) is currently the most promising short-term solution; however, ensuring its overall sustainability depends on reducing the life cycle carbon footprints. A key challenge prevails in hydrogen usage as a reactant for the approved ASTM routes of SAF. The processing, conversion and refinement of feed entailing hydrodeoxygenation (HDO), decarboxylation, hydrogenation, isomerisation and hydrocracking requires substantial hydrogen input. This hydrogen is sourced either in situ or ex situ, with the supply chain encompassing renewables or non-renewables origins. Addressing this hydrogen usage and recognising the emission implications thereof has therefore become a novel research priority. Aside from the preferred adoption of renewable water electrolysis to generate hydrogen, other promising pathways encompass hydrothermal gasification, biomass gasification (with or without carbon capture) and biomethane with steam methane reforming (with or without carbon capture) owing to the lower greenhouse emissions, the convincing status of the technology readiness level and the lower acidification potential. Equally imperative are measures for reducing hydrogen demand in SAF pathways. Strategies involve identifying the appropriate catalyst (monometallic and bimetallic sulphide catalyst), increasing the catalyst life in the deoxygenation process, deploying low-cost iso-propanol (hydrogen donor), developing the aerobic fermentation of sugar to 1,4 dimethyl cyclooctane with the intermediate formation of isoprene and advancing aqueous phase reforming or single-stage hydro processing. Other supportive alternatives include implementing the catalytic and co-pyrolysis of waste oil with solid feedstocks and selecting highly saturated feedstock. Thus, future progress demands coordinated innovation and research endeavours to bolster the seamless integration of the cutting-edge hydrogen production processes with the SAF infrastructure. Rigorous techno-economic and life cycle assessments, alongside technological breakthroughs and biomass characterisation, are indispensable for ensuring scalability and sustainability. Full article

Background: The hemoglobin–albumin–lymphocyte–platelet (HALP) score integrates the immunonutritional and inflammatory status. We evaluated whether baseline HALP predicts the 3-month response after lumbar medial branch radiofrequency ablation (RFA), defined as a Visual Analogue Scale (VAS) reduction of ≥50% and an Oswestry Disability Index (ODI) reduction of ≥40%, and identified a Youden-optimal cut-off. The discrimination and calibration of multivariable models were also assessed. Methods: This single-center retrospective cohort (N = 120) included rigorously selected patients (≥50% pain relief after two comparative medial branch blocks) undergoing standardized RFA. Multivariable logistic regression was adjusted for age, sex, Body Mass Index (BMI), smoking status, paraspinal tenderness, and baseline scores. We quantified the Area Under the Receiver Operating Characteristic Curve (AUC), Hosmer–Lemeshow (HL) goodness-of-fit, Brier score, and calibration slope; optimism was corrected using a 500-bootstrap method. Results: Responses occurred in 64.2% (VAS) and 65.8% (ODI) of participants. HALP independently predicted ODI (OR = 1.06, 95% CI 1.02–1.09; p < 0.001) and VAS (OR = 1.05, 95% CI 1.02–1.08; p = 0.001). As a single predictor, HALP showed fair discrimination (AUC 0.717 [VAS], 0.731 [ODI]). The Youden cut-off of 39.8 yielded high sensitivity (~0.87) with modest specificity (~0.58–0.61). Multivariable AUCs were 0.744 (VAS) and 0.774 (ODI), optimism-corrected to 0.680 and 0.720; calibration was acceptable (HL p > 0.05; slopes ≈ 0.74–0.78; Brier 0.188/0.179). Conclusions: HALP is a simple, low-cost adjunct that independently predicts short-term pain and functional outcomes after lumbar medial branch RFA. Incorporation into post-block triage may refine selection, especially for functional improvement, pending prospective external validation and recalibration of the cut-off. Full article

The header platform of a combine harvester is subjected to severe abrasive and corrosive wear from rice stalks and environmental factors, which significantly limits its service life and operational efficiency. Accurately predicting the complex distribution of this wear over time and across the platform’s surface, however, remains a significant challenge. This paper, for the first time, systematically establishes a quantitative mapping relationship from “material motion trajectory” to “component wear profile” and introduces a novel method for time-sequence wear validation based on corrosion color gradients, providing a complete research paradigm to address this challenge. To this end, an analytical model based on rigid-body dynamics was first developed to predict the motion trajectory of a single rice stalk. Subsequently, a full-scale Discrete Element (DEM) model of the header platform–flexible rice stalk system was constructed. This model simulated the complex flow process of the rice population with high fidelity and was used to analyze the influence of key operating parameters (spiral auger rotational speed, cutting width) on wear distribution. Finally, real-world wear data were obtained through in situ mapping of a header platform after long-term service (1300 h) and multi-period (0–1600 h) image analysis. Through a three-way quantitative comparison among the theoretical trajectory, simulated trajectory, and the actual wear profile, the results indicate that the simulated and theoretical trajectories are in good agreement in terms of their macroscopic trends (Mean Squared Error, MSE, ranging from 0.4 to 6.2); the simulated and actual wear profiles exhibit an extremely high degree of geometric similarity, with the simulated wear area showing a 95.1% match to the actual measured area (Edit Distance: 0.14; Hamming Distance: 1). This research not only confirms that the flow trajectory of rice is the determining factor for the wear distribution on the header platform but, more importantly, the developed analytical and numerical methods offer a robust theoretical basis and effective predictive tools for optimizing the wear resistance and predicting the service life of the header platform, thereby demonstrating significant engineering value. Full article

Urban recreational spaces (URSs) are pivotal for enhancing resident well-being, making the accurate assessment of public perceptions crucial for quality optimization. Compared to traditional surveys, social media data provide a scalable means for multi-dimensional perception assessment. However, existing studies predominantly rely on single-modal data, which limits the comprehensive capturing of complex perceptions and lacks interpretability. To address these gaps, this study employs cutting-edge large vision–language models (LVLMs) and develops an interpretable model, Qwen2.5-VL-7B-SFT, through supervised fine-tuning on a manually annotated dataset. The model integrates visual-linguistic features to assess four perceptual dimensions of URSs: esthetics, attractiveness, cultural significance, and restorativeness. Crucially, we generate textual evidence for our judgments by identifying the key spatial elements and emotional characteristics associated with specific perceptions. By integrating multi-source built environment data with Optuna-optimized machine learning and SHAP analysis, we further decipher the nonlinear relationships between built environment variables and perceptual outcomes. The results are as follows: (1) Interpretable LVLMs are highly effective for urban spatial perception research. (2) URSs within Beijing’s Third Ring Road fall into four typologies, historical heritage, commercial entertainment, ecological-natural, and cultural spaces, with significant correlations observed between physical elements and emotional responses. (3) Historical heritage accessibility and POI density are identified as key predictors of public perception. Positive perception significantly improves when a block’s POI functional density exceeds 4000 units/km² or when its 500 m radius encompasses more than four historical heritage sites. Our methodology enables precise quantification of multidimensional URS perceptions, links built environment elements to perceptual mechanisms, and provides actionable insights for urban planning. Full article

Background: Hepatocellular carcinoma (HCC) frequently develops in patients with chronic hepatitis B and C. Early detection is critical, but current methods, including ultrasound and AFP, have suboptimal accuracy. Objectives: This study aimed to evaluate the predictive performance of protein induced by vitamin K absence or antagonist-II (PIVKA-II) and alpha-fetoprotein (AFP) testing, alone and in combination, for HCC development. Methods: A retrospective cohort study at a single university center included 242 CHB and 181 CHC patients. Data on demographics, clinical status, laboratory parameters, and imaging were collected, with fibrosis and steatosis assessed by FibroScan^®. Serum AFP and PIVKA-II were measured, but measurements of PIVKA-II in patients receiving vitamin K antagonists were excluded from the analysis. HCC diagnosis and staging followed clinical guidelines. Cox regression and ROC analyses identified independent predictors and evaluated biomarker accuracy for HCC detection. Results: HCC incidence was comparable between cohorts (5.0% in CHB vs. 5.5% in CHC). Both AFP and PIVKA-II independently predicted HCC development in multivariate models adjusted for age and sex. The combined biomarker score (AFP × PIVKA-II) showed superior predictive accuracy with hazard ratios of 1.38 (CHB) and 1.36 (CHC). ROC analyses demonstrated high discriminative ability for PIVKA-II (AUC ~0.81) and AFP (AUC ~0.83) in both cohorts. Additional independent predictors were chronic alcohol abuse, cirrhosis, and higher liver stiffness measurements. Specific viral factors such as HBeAg positivity and HCV subgenotype 1b were also associated with increased HCC risk. Conclusions: AFP and PIVKA-II are independent, valuable biomarkers for HCC risk in chronic hepatitis B and C. Combined use improves early detection, aiding timely treatment. These results support adding PIVKA-II to AFP in surveillance, but larger studies are needed to confirm the findings and refine cut-off values. Full article

Background/Objectives: Inflammation plays a key role in the pathophysiology of acute myocardial infarction (AMI). Yet static measures of C-reactive protein (CRP) provide limited prognostic information. CRP velocity (CRPv), which reflects the rate of CRP rise within the first 24 h, may better depict the dynamic inflammatory response. To investigate the prognostic role of CRPv in patients presenting with a first AMI. Methods: Consecutive patients presenting with first AMI were enrolled. CRPv was calculated as the difference between CRP at admission and after 24 ± 8 h, divided by time. A prognostic CRPv cut-off was derived from spline curve analysis to dichotomize the population. Patients were followed up for the primary composite endpoint of cardiovascular death, non-fatal AMI, and hospitalization for heart failure. Results: Among 604 patients, 189 (31.3%) had CRPv ≥ 1.36 mg/L/h and 415 (68.7%) had CRPv < 1.36 mg/L/h. Higher hs-cTnT (adjusted odds ratio [aOR] 2.552, 95% CI, 1.520–4.286; p < 0.001) and NT-proBNP (aOR 2.229, 95% CI, 1.241–4.002; p = 0.007) were independently associated with CRPv ≥ 1.36 mg/L/h. At a median follow-up of 13.8 months, 115 patients (19.0%) reached the primary composite endpoint. High CRPv patients had significantly lower event-free survival rate than low CRPv patients (66.7% vs. 85.5%, log-rank p < 0.001). CRPv independently predicted the primary composite endpoint [adjusted hazard ratio 1.226, 95% CI 1.102–1.364, p < 0.001]. Adding CRPv on top of clinical, echocardiographic, and biochemical risk factors significantly improved model discrimination (p < 0.001), whereas single CRP on admission (p = 0.947) or CRP 24 ± 8 h from admission (p = 0.064) did not. Conclusions: CRPv appears to be a robust predictor of adverse outcomes in first AMI patients, offering incremental prognostic value beyond established clinical and biomarker indices. Its feasibility and low cost support its integration into early clinical risk stratification. Full article

(1) Background: Concomitant mild aortic valve disease is frequently found in patients undergoing rheumatic mitral valve surgery. To date, only a limited number of single-center studies have specifically addressed the untreated baseline aortic valve disease long-term progression and reoperation rate. Thus, we conducted a meta and landmark analysis to systematically review the issue. (2) Methods: This study investigated the long-term prognostic of baseline mild aortic valve disease in patients undergoing rheumatic mitral valve surgery, based on evidence from PubMed, Embase, Cochrane Library, and Web of Science databases. (3) Results: Meta analysis revealed that patients with mild aortic valve disease had a higher risk of disease progression, with a 3.3-fold risk in the 0–5-year follow-up, which jumped to a hazard ratio of 6.42 in longer-term follow-up (5–25 years). Patients with aortic stenosis had an 8.37-fold risk of progression compared with aortic regurgitation and appeared to be poorly related to the time cut-off. Similarly, higher reoperation rates at long-term follow-up were seen in aortic stenosis patients. (4) Conclusions: This study suggests that patients with mild aortic valve disease at baseline have poorer long-term aortic valve-related progression and reoperation rates, especially aortic stenosis. For those with concomitant aortic stenosis, further investigation of the impact of lesion progression is warranted. Full article

To address the limitation that single-index evaluation fails to fully reflect the development performance of reservoirs of different types and at various development stages, a multi-index comprehensive evaluation system featuring the workflow of “index screening–weight determination–model evaluation–strategy guidance” was established. Firstly, the grey correlation analysis method (with a correlation degree threshold set at 0.65) was employed to screen 12 key evaluation indicators, including reservoir physical properties (porosity, permeability) and development dynamics (recovery factor, water cut, well activation rate). Subsequently, the fuzzy analytic hierarchy process (FAHP, for subjective weighting, with the consistency ratio (CR) of expert judgments < 0.1) was coupled with the attribute measurement method (for objective weighting, with information entropy redundancy < 5%) to determine the indicator weights, thereby balancing the influences of subjective experience and objective data. Finally, two evaluation models, namely the fuzzy comprehensive decision-making method and the unascertained measurement method, were constructed to conduct evaluations on 308 reservoirs in the Liaohe Oilfield (covering five major categories: integral medium–high-permeability reservoirs, complex fault-block reservoirs, low-permeability reservoirs, special lithology reservoirs, and thermal recovery heavy oil reservoirs). The results indicate that there are 147 high-efficiency reservoirs categorized as Class I and Class II in total. Although these reservoirs account for 47.7% of the total number, they control 71% of the geological reserves (154,548 × 10⁴ t) and 78% of the annual oil production (738.2 × 10⁴ t) in the oilfield, with an average well activation rate of 65.4% and an average recovery factor of 28.9. Significant quantitative differences are observed in the development characteristics of different reservoir types: Integral medium–high-permeability reservoirs achieve an average recovery factor of 37.6% and an average well activation rate of 74.1% by virtue of their excellent physical properties (permeability mostly > 100 mD), with Block Jin 16 (recovery factor: 56.9%, well activation rate: 86.1%) serving as a typical example. Complex fault-block reservoirs exhibit optimal performance at the stage of “recovery degree > 70%, water cut ≥ 90%”, where 65.6% of the blocks are classified as Class I, and the recovery factor of blocks with a “good” rating (42.3%) is 1.8 times that of blocks with a “poor” rating (23.5%). For low-permeability reservoirs, blocks with a rating below medium grade account for 68% of the geological reserves (8403.2 × 10⁴ t), with an average well activation rate of 64.9%. Specifically, Block Le 208 (permeability < 10 mD) has an annual oil production of only 0.83 × 10⁴ t. Special lithology reservoirs show polarized development performance, as Block Shugu 1 (recovery factor: 32.0%) and Biantai Buried Hill (recovery factor: 20.4%) exhibit significantly different development effects due to variations in fracture–vug development. Among thermal recovery heavy oil reservoirs, ultra-heavy oil reservoirs (e.g., Block Du 84 Guantao, with a recovery factor of 63.1% and a well activation rate of 92%) are developed efficiently via steam flooding, while extra-heavy oil reservoirs (e.g., Block Leng 42, with a recovery factor of 19.6% and a well activation rate of 30%) are constrained by reservoir heterogeneity. This system refines the quantitative classification boundaries for four development levels of water-flooded reservoirs (e.g., for Class I reservoirs in the high water cut stage, the recovery factor is ≥35% and the water cut is ≥90%), as well as the evaluation criteria for different stages (steam huff and puff, steam flooding) of thermal recovery heavy oil reservoirs. It realizes the transition from traditional single-index qualitative evaluation to multi-index quantitative evaluation, and the consistency between the evaluation results and the on-site development adjustment plans reaches 88%, which provides a scientific basis for formulating development strategies for the Liaohe Oilfield and other similar oilfields. Full article

At present, most oilfields in China have entered the late, high-water-cut stage, commonly facing declining single-well productivity and increasingly pronounced reservoir heterogeneity. Prolonged waterflooding has further exacerbated permeability contrast, yielding complex, hard-to-produce residual-oil distributions. Accordingly, the development of efficient enhanced oil recovery (EOR) technologies has become a strategic priority and an urgent research focus in oil and gas field development. Weak gels, typical non-Newtonian fluids, exhibit both viscous and elastic responses, and their distinctive rheology shows broad application potential for crude oil extraction in porous media. Targeting medium–high-permeability reservoirs with high water cut, this study optimized and evaluated a weak gel system. Experimental results demonstrate that the optimized weak gel system achieves remarkable oil displacement performance. The one-dimensional dual-sandpack flooding tests yielded a total recovery of 72.26%, with the weak gel flooding stage contributing an incremental recovery of 14.52%. In the physical three-dimensional model experiments, the total recovery reached 46.12%, of which the weak gel flooding phase accounted for 16.36%. Through one-dimensional sandpack flow experiments and three-dimensional physical model simulations, the oil displacement mechanisms and synergistic effects of the optimized system in heterogeneous reservoirs were systematically elucidated from macro to micro scales. The optimized system demonstrates integrated synergistic performance during flooding, effectively combining mobility control, displacement, and oil-washing mechanisms. Macroscopically, it effectively strips residual oil in high-permeability zones via viscosity enhancement and viscoelastic effects, efficiently blocks high-permeability channels, diverts flow to medium-permeability regions, and enhances macroscopic sweep efficiency. Microscopically, it mobilizes residual oil via normal stress action and a filamentous transport mechanism, improving oil-washing efficiency and increasing ultimate oil recovery. This study demonstrates the technical feasibility and practical effectiveness of the optimized weak gel system for enhancing oil recovery in heterogeneous reservoirs, providing critical technical support for the efficient development of medium–high-permeability reservoirs with high water cut. Full article