Compressibility of lime-treated dredged slurry with high water content.

Lime is widely used for soft ground treatment, rendering the compressibility of lime-treated soil a crucial factor in deformation analysis in engineering applications. This study investigated the compressibility of three remoulded lime-treated slurries with high water content in Southeast China. Sixty groups of oedometer tests were conducted on lime-treated soils with an initial water content of 1 to 3 times the liquid limit and lime contents between 1 and 3%. The oedometer test results were discussed to examine the remoulded yield stress σ y ' of lime-treated slurry. Considering the relationships between σ y ' , the void ratio, lime content, and initial water content were preliminarily discussed and quantitatively established. Research on the normalised compression curve of lime-treated soil revealed that for soil samples containing a lime content of 0-%, the normalised compression curve at σ p ' > σ y ' can be represented by a unique line. Furthermore, the log(1 + e) - log σ v ' compression curve of lime-treated slurry at pre-yield state is analysed, and a prediction method for the modified compression index is proposed.

Portable electrochemical sensing platform based on amidated GO-MOF and PEDOT:PSS for high-efficient detection of ponceau 4R.

A promising electrochemical sensing platform for the detection of ponceau 4R in food has been fabricated based on the carboxylated graphene oxide (GO-COOH), metal-organic framework (MOF) UIO-66-NH2, and poly (3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS). To this end GO-COOH was covalently coupled with UIO-66-NH2 through amide reaction, endowing the material (GO-CONH-UIO-66) unique hierarchical pores and high chemical stability and as a result improving the conductivity of MOF and the dispersion of GO. After the addition of PEDOT:PSS into GO-CONH-UIO-66, the continuity and conductivity of the composite (PEDOT:PSS/GO-CONH-UIO-66) have been further enhanced, due to the high conductivity, favorable film-forming, and hydrophilic properties of PEDOT:PSS. Systematic electrochemical experiments confirm that the PEDOT:PSS/GO-CONH-UIO-66/GCE shows satisfactory electrochemical sensing properties towards the detection of ponceau 4R, with a wide linear detection range of 0.01-30 µM, a low limit of detection of 3.33 nM, and a high sensitivity of 0.606 µA µM-1 cm-2. The PEDOT:PSS/GO-CONH-UIO-66 sensing platform was successfully used to detect ponceau 4R in beverage, and the detection results were compared with  high-performance liquid chromatography. As a result, the PEDOT:PSS/GO-CONH-UIO-66 composite shows a promising application prospect for rapid detection of ponceau 4R in food and will play significant role in food safety detection and supervision.

Maillard Reaction Inspired Microexplosion toward Fast Synthesis of Two-Dimensional Mesoporous Tin Oxides for Efficient Chemiresistive Gas Sensing.

Two-dimensional (2D) mesoporous transition metal oxides are highly desired in various applications, but their fast and low-cost synthesis remains a great challenge. Herein, a Maillard reaction inspired microexplosion approach is applied to rapidly synthesize ultrathin 2D mesoporous tin oxide (mSnO2). During the microexplosion between granular ammonia nitrate with melanoidin at high temperature, the organic species can be carbonized and expanded rapidly due to the instantaneous release of gases, thus producing ultrathin carbonaceous templates with rich functional groups to effectively anchor SnO2 nanoparticles on the surface. The subsequent removal of carbonaceous templates via calcination in air results in the formation of 2D mSnO2 due to the confinement effect of the templates. Pd nanoparticles are controllably deposited on the surface of 2D mSnO2 via in situ reduction, forming ultrathin 2D Pd/mSnO2 nanocomposites with thicknesses of 6-8 nm. Owing to the unique 2D mesoporous structure with rich oxygen defects and highly exposed metal-metal oxide interfaces, 2D Pd/mSnO2 exhibits excellent sensing performance toward acetone with high sensitivity, a short response time, and good selectivity under low working temperature (100 °C). This fast and convenient microexplosion synthesis strategy opens up the possibility of constructing 2D porous functional materials for various applications including high-performance gas sensors.

Manipulating 2D Materials through Strain Engineering.

This review explores the growing interest in 2D layered materials, such as graphene, h-BN, transition metal dichalcogenides (TMDs), and black phosphorus (BP), with a specific focus on recent advances in strain engineering. Both experimental and theoretical results are delved into, highlighting the potential of strain to modulate physical properties, thereby enhancing device performance. Various strain engineering methods are summarized, and the impact of strain on the electrical, optical, magnetic, thermal, and valleytronic properties of 2D materials is thoroughly examined. Finally, the review concludes by addressing potential applications and challenges in utilizing strain engineering for functional devices, offering valuable insights for further research and applications in optoelectronics, thermionics, and spintronics.

Novel biaryloxazolidinone derivatives with broad-spectrum antibacterial activity, favorable drug-like profiles and in vivo efficacy against linezolid-resistant Staphylococcusaureus.

The emergence of multidrug-resistant bacteria along with a declining pipeline of clinically useful antibiotics has led to the urgent need for the development of more effective antibacterial agents to treat drug-resistant bacteria. We previously discovered compound OB-158 with potent antibacterial activity but exhibited poor oral bioavailability. Herein, a systematic structural optimization of OB-158 to improve pharmacokinetic profiles yielded 26 novel biaryloxazolidinone analogues, and their activities against Gram-positive S. aureus, multidrug resistant S. aureus and Enterococcus faecalis were evaluated. Remarkably, compound 8b was identified with potent antibacterial activity against S. aureus (MIC = 0.06 µg/mL), MSSA (MIC = 0.125 µg/mL), MRSA (MIC = 0.06 µg/mL), LRSA (MIC = 0.125 µg/mL) and LREFa (MIC = 0.5 µg/mL). Compound 8b was demonstrated as a promising candidate through druglikeness evaluation including metabolism in microsomes and plasma, Caco-2 cell permeability, plasma protein binding, cytotoxicity, and inhibition of CYP450 and human monoamine oxidase. Notably, compound 8b displayed excellent PK profile with appropriate T1/2 of 1.49 h, high peak plasma concentration (Cmax = 2320 ng/mL), high plasma exposure (AUC0-t = 8310 h ng/mL), and superior oral bioavailability (F = 68.1 %) in Sprague-Dawley rats. Ultimately, in vivo efficacy of compound 8b in a mouse model of LRSA systemic infection was also demonstrated. Taken together, compound 8b represents a promising drug candidate for the treatment of linezolid-resistant Gram-positive bacterial strains infection.

Construction of dual-chiral covalent organic frameworks for enantioselective separation.

Developing novel chiral stationary phases (CSPs) with versatility is of great importance in enantiomer separation. This study fabricated a dual-chiral covalent organic framework (PA-CA COF) via successive post-synthetic modifications. The chiral trans-1,2-cyclohexanediamine (CA) and (D)-penicillamine (PA) groups were periodically aligned within nanochannels of the COF, allowing selective recognition of enantiomers through intermolecular interactions. It can be a versatile high-performance liquid chromatography (HPLC) CSP for separating a wide range of enantiomers, including chiral pharmaceutical intermediates and chiral drugs. With separation performance comparable to commercial chiral columns and even greater versatility, the PA-CA COF@SiO2 column held promise for practical applications. Chiral separation results combined with molecular simulation indicated that the mixed mode of PA and CA resulted in the broad separation capability of PA-CA COF. The introduction of the dual-chiral COFs concept opens up a new avenue for chiral recognition and separation, holding great potential for practical enantiomer separation.

Physiology and transcriptomic analysis revealed the mechanism of silicon promoting cadmium accumulation in Sedum alfredii Hance.

Silicon (Si) effectively promote the yield of many crops, mainly due to its ability to enhance plants resistance to stress. However, how Si helps hyperaccumulators to extract Cadmium (Cd) from soil has remained unclear. In this study, Sedum alfredii Hance (S. alfredii) was used as material to study how exogenous Si affected biomass, Cd accumulation, antioxidation, cell ultrastructure, subcellular distribution and changes in gene expression after Cd exposure. The study has shown that as Si concentration increases (1, 2 mM), the shoot biomass of plants increased by 33.1%-63.6%, the Cd accumulation increased by 31.9%-96.6%, and the chlorophyll, carotenoid content, photosynthetic gas exchange parameters significantly increased. Si reduced Pro and MDA, promoted the concentrations of SOD, CAT and POD to reduce antioxidant stress damage. In addition, Si promoted GSH and PC to chelate Cd in vacuoles, repaired damaged cell ultrastructure, improved the fixation of Cd and cell wall (especially in pectin), and reduced the toxic effects of Cd. Transcriptome analysis found that genes encoding Cd detoxification, Cd absorption and transport were up-regulated by Si supplying, including photosynthetic pathways (PSB, LHCB, PSA), antioxidant defense systems (CAT, APX, CSD, RBOH), cell wall biosynthesis such as pectinesterase (PME), chelation (GST, MT, NAS, GR), Cd absorption (Nramp3, Nramp5, ZNT) and Cd transport (HMA, PCR). Our result revealed the tentative mechanism of Si promotes Cd accumulation and enhances Cd tolerance in S. alfredii, and thereby provides a solid theoretical support for the practical use of Si fertilizer in phytoextraction.

Prevalence of Parkinson's disease among adults aged 45 years and older in China: a cross-sectional study based on the China health and retirement longitudinal study.

BACKGROUND: In recent decades, China has experienced a rapid increase in the number of elderly individuals and life expectancy, as well as industrialization, which is associated with an increased prevalence of Parkinson's disease (PD). However, inconsistent results have recently been reported. Therefore, this study aimed to investigate the prevalence and distribution characteristics of PD among individuals aged 45 years and older. METHODS: Using data from the China Health and Retirement Longitudinal Study (CHARLS), we attempted to estimate the prevalence of PD and its distribution characteristics among 19,034 individuals aged 45 years and older residing in 446 communities/villages within 27 provinces/autonomous regions/municipalities in mainland China. Cases were established based on a doctor's previous diagnosis. Crude and age-adjusted prevalence rates were calculated and stratified by age, sex, area of residence, education level, marital status, and geographic region. Logistic regression models were used to identify risk factors associated with PD. RESULTS: We identified 178 patients with PD among 19,034 residents aged 45 years and older. The crude prevalence was 0.94%, and the age-adjusted prevalence was 0.82% for individuals aged 45 years and older. The prevalence of PD increased with age (P < 0.001). No significant differences were found in terms of sex, area of residence, or education level. Stratified by geographic region, the prevalence of PD was greater in North and Northwest China and lower in southern China (p < 0.001). Multiple regression analyses showed that age was a significant risk factor for PD. CONCLUSION: The prevalence of PD increased with age in the Chinese population.

[Research progress of traditional Chinese medicine intervention for ulcerative colitis based on Notch1 signaling pathway].

Ulcerative colitis(UC) is one of the common gastrointestinal diseases worldwide. In recent years, the incidence of UC has been continuously increasing, seriously threatening the health of people globally. It thus has become an urgent problem that needs to be addressed. There is research evidence that intestinal mucosal barrier dysfunction, including changes in intestinal stem cell secretion lineage, mucosal layer damage, disruption of cell junctions, overactive immune function, and imbalanced gut microbiota, is an important pathogenic factor and molecular basis of UC. The Notch signaling pathway is a highly conserved signaling pathway in eukaryotes during evolution, which transmits signals through cell connections between adjacent cells, affecting a series of processes such as cell proliferation, differentiation, development, migration, and apoptosis. Therefore, the Notch signaling pathway can regulate intestinal stem cells, CD4~+T cells, innate lymphoid cells(ILCs), macrophages(MØ), and intestinal microbiota and thus affect the chemical, physical, immune, and biological mucosal barriers of the intestinal mucosa. Its function is extensive and unique, different from those signaling pathways that mainly focus on anti-inflammatory and antioxidant stress. It can explain the therapeutic effects of traditional Chinese medicine from different perspectives. This article reviewed the role of the Notch1 signaling pathway in the pathogenesis of UC and the relevant literature on the targeted prevention and treatment of UC with traditional Chinese medicine, so as to provide new targets and theoretical support for further research on the effective prevention and treatment of UC.

Developing hydrogels for gene therapy and tissue engineering.

Hydrogels are a class of highly absorbent and easily modified polymer materials suitable for use as slow-release carriers for drugs. Gene therapy is highly specific and can overcome the limitations of traditional tissue engineering techniques and has significant advantages in tissue repair. However, therapeutic genes are often affected by cellular barriers and enzyme sensitivity, and carrier loading of therapeutic genes is essential. Therapeutic gene hydrogels can well overcome these difficulties. Moreover, gene-therapeutic hydrogels have made considerable progress. This review summarizes the recent research on carrier gene hydrogels for the treatment of tissue damage through a summary of the most current research frontiers. We initially introduce the classification of hydrogels and their cross-linking methods, followed by a detailed overview of the types and modifications of therapeutic genes, a detailed discussion on the loading of therapeutic genes in hydrogels and their characterization features, a summary of the design of hydrogels for therapeutic gene release, and an overview of their applications in tissue engineering. Finally, we provide comments and look forward to the shortcomings and future directions of hydrogels for gene therapy. We hope that this article will provide researchers in related fields with more comprehensive and systematic strategies for tissue engineering repair and further promote the development of the field of hydrogels for gene therapy.

A Versatile Synthesis Platform Based on Polymer Cubosomes for a Library of Highly Ordered Nanoporous Metal Oxides Particles.

Polymer cubosomes (PCs) have well-defined inverse bicontinuous cubic mesophases formed by amphiphilic block copolymer bilayers. The open hydrophilic channels, large periods, and robust physical properties of PCs are advantageous to many host-guest interactions and yet not fully exploited, especially in the fields of functional nanomaterials. Here, the self-assembly of poly(ethylene oxide)-block-polystyrene block copolymers is systematically investigated and a series of robust PCs is developed via a cosolvent method. Ordered nanoporous metal oxide particles are obtained by selectively filling the hydrophilic channels of PCs via an impregnation strategy, followed by a two-step thermal treatment. Based on this versatile PC platform, the general synthesis of a library of ordered porous particles with different pore structures 3 ¯ $\bar{3}$ 3 ¯ $\bar{3}$ , tunable large pore size (18-78 nm), high specific surface areas (up to 123.3 m2 g-1 for WO3) and diverse framework compositions, such as transition and non-transition metal oxides, rare earth chloride oxides, perovskite, pyrochlore, and high-entropy metal oxides is demonstrated. As typical materials obtained via this method, ordered porous WO3 particles have the advantages of open continuous structure and semiconducting properties, thus showing superior gas sensing performances toward hydrogen sulfide.

Evaluation of vegetation restoration effectiveness along the Yangtze River shoreline and its response to land use changes.

Assessing the effectiveness of vegetation restoration along the Yangtze River shoreline and exploring its relationship with land use changes are imperative for providing recommendations for sustainable management and environmental protection. However, the impact of vegetation restoration post-implementation of the Yangtze River Conservation Project remains uncertain. In this study, utilizing Sentinel-2 satellite imagery and Dynamic World land use data from pre- (2016) and post- (2022) Yangtze River Conservation Project periods, pixel-based binary models, transition matrices, and geographically weighted regression models were employed to analyze the status and evolution of vegetation coverage along the Yangtze River shoreline. The results indicated that there had been an increase in the area covered by high and high-medium vegetation levels. The proportion of vegetation cover shifting to better was 4201.87 km2 (35.68%). Hotspots of vegetation coverage improvement were predominantly located along the Yangtze River. Moreover, areas witnessing enhanced vegetation coverage experienced notable land use changes, notably the conversion of water to crops (126.93 km2, 22.79%), trees to crops (59.93 km2, 10.76%), and crops to built area (59.93 km2, 10.76%). Notably, the conversion between crops and built area emerged as a significant factor influencing vegetation coverage improvement, with average regression coefficients of 0.68 and 0.50, respectively. These outcomes underscore the significance of this study in guiding ecological environmental protection and sustainable management along the Yangtze River shoreline.

Associations between somatic symptoms and remission of major depressive disorder: A longitudinal study in China.

Previous studies have documented negative associations between somatic symptoms and remission of major depressive disorder (MDD). However, the correlations of specific somatic symptoms with remission remain uncertain. We aimed to explore the associations between specific somatic symptoms and remission focusing on sex differences among patients with MDD. We used data from patients with MDD in the Depression Cohort in China. At baseline, total somatic symptoms were evaluated using the 28-item Somatic Symptoms Inventory and were categorized into pain, autonomic, energy, and central nervous system (CNS) symptoms. To measure remission of MDD, depressive symptoms were evaluated using the Patient Health Questionnaire-9 after 3 months of treatment. We ultimately included 634 patients. Compared with quartile 1 of total somatic symptom scores, the full-adjusted ORs (95% CIs) for remission from quartile 2 to quartile 4 were 0.52 (0.30, 0.90), 0.44 (0.23, 0.83), and 0.36 (0.17, 0.75), respectively (P-value for trend = 0.005). The restricted cubic spline showed no non-linear associations between total somatic symptoms with remission (P-value for non-linear = 0.238). Pain, autonomic, and CNS symptoms showed similar results. Sex-stratified analysis showed that total somatic symptoms, pain symptoms, and autonomic symptoms were negatively correlated with remission in females, whereas CNS symptoms were negatively associated with remission in males. Our findings indicate that specific somatic symptoms exert differential effects on remission of MDD. Therapeutic interventions that target pain, autonomic, and CNS symptoms may increase the probability of remission. Furthermore, interventions for somatic symptoms should be tailored by sex, and females deserve more attention.

P3MOT-decorated metal-porphyrin-based zirconium-MOF for the efficient electrochemical detection of 4-nitrobenzaldehyde.

A novel hybrid composite integrating conductive poly-3-methoxythiophene and PCN-222(Fe) (porphyrin-metal-organic frameworks) was synthesized using an in situ polymerization strategy. Leveraging the large specific area of MOFs and the low electrical resistance of conductive polymers, the modified electrode proved to be a promising candidate for the electrochemical detection of 4-nitrobenzaldehyde. The electrocatalytic response was measured using differential pulse voltammetry techniques and cyclic voltammetry, where the linear concentration range of analyte detection was estimated to be 0-900 µM and the detection limit was 0.233 µM with high selectivity toward the analyte. The sensor demonstrated repeatability and stability, allowing the direct electroanalytical measurement of 4-nitrobenzaldehyde in real samples with reliable recovery. This methodology expands the application of porphyrin MOFs for the electroanalytical sensing of environmental contaminants.

Associations of rumination with suicidal ideation and suicide attempts amongst individuals with major depressive disorder: A 12-month longitudinal study in China.

BACKGROUND: This study aimed to explore the longitudinal associations of rumination with suicidal ideation and suicide attempts in individuals with major depressive disorder (MDD). METHODS: Participants were derived from the Depression Cohort in China study (DCC). Those who completed at least one follow-up visit during the 12 months were included in the analysis. Dimensions of rumination including brooding and reflection were each measured using five items of the Ruminative Responses Scale. Suicidal ideation was assessed using the Beck Scale for Suicide Ideation. Suicide attempts were also assessed and all were analyzed with generalized estimating equations. RESULTS: Our final sample included 532 participants aged 18 to 59 years (mean [SD], 26.91 [6.94] years) consisting of 148 (27.8%) males and 384 (72.2%) females. After adjusting for temporal trend and potential confounders, individuals with higher levels of reflection were more likely to report suicidal ideation (AOR =1.11, 95% CI:1.01-1.22). However, no statistically significant association was found between brooding and suicidal ideation (AOR =1.06, 95% CI:0.96-1.17). Conversely, individuals with higher levels of brooding were more likely to report suicide attempts (AOR =1.13, 95% CI:1.02-1.24), while no statistically significant association was observed between reflection and suicide attempts (AOR =0.91, 95% CI:0.82-1.01). CONCLUSION: Rumination reflects a disturbance in cognitive emotional processing and manifests in different dimensions. Our findings suggest that high levels of reflection and brooding may be associated with a higher likelihood of having suicidal ideation and suicide attempts, respectively. However, it should be interpreted with caution, given that effect sizes are small.

Transcriptome analysis of porcine embryos derived from oocytes vitrified at the germinal vesicle stage.

Vitrification of porcine immature oocytes at the germinal vesicle (GV) stage reduces subsequent embryo yield and changes at the molecular level may occur during embryonic development. Therefore, the present study used porcine parthenogenetic embryos as a model to investigate the effect of GV oocyte vitrification on the transcriptional profiles of the resultant embryos at the 4-cell and blastocyst stages using the Smart-seq2 RNA-seq technique. We identified 743 (420 up-regulated and 323 down-regulated) and 994 (554 up-regulated and 440 down-regulated) differentially expressed genes (DEGs) from 4-cell embryos and blastocysts derived from vitrified GV oocytes, respectively. Functional enrichment analysis of DEGs in 4-cell embryos showed that vitrification of GV oocytes influenced regulatory mechanisms related to transcription regulation, apoptotic process, metabolism and key pathways such as the MAPK signaling pathway. Moreover, DEGs in blastocysts produced from vitrified GV oocytes were enriched in critical biological functions including cell adhesion, cell migration, AMPK signaling pathway, GnRH signaling pathway and so on. In addition, the transcriptomic analysis and quantitative real-time PCR results were consistent. In summary, the present study revealed that the vitrification of porcine GV oocytes could alter gene expression patterns during subsequent embryonic developmental stages, potentially affecting their developmental competence.

PIDNET: Polar Transformation Based Implicit Disentanglement Network for Truncation Artifacts.

The interior problem, a persistent ill-posed challenge in CT imaging, gives rise to truncation artifacts capable of distorting CT values, thereby significantly impacting clinical diagnoses. Traditional methods have long struggled to effectively solve this issue until the advent of supervised models built on deep neural networks. However, supervised models are constrained by the need for paired data, limiting their practical application. Therefore, we propose a simple and efficient unsupervised method based on the Cycle-GAN framework. Introducing an implicit disentanglement strategy, we aim to separate truncation artifacts from content information. The separated artifact features serve as complementary constraints and the source of generating simulated paired data to enhance the training of the sub-network dedicated to removing truncation artifacts. Additionally, we incorporate polar transformation and an innovative constraint tailored specifically for truncation artifact features, further contributing to the effectiveness of our approach. Experiments conducted on multiple datasets demonstrate that our unsupervised network outperforms the traditional Cycle-GAN model significantly. When compared to state-of-the-art supervised models trained on paired datasets, our model achieves comparable visual results and closely aligns with quantitative evaluation metrics.

Machine Learning Supported the Modified Gustafson's Criteria for Dental Age Estimation in Southwest China.

Adult age estimation is one of the most challenging problems in forensic science and physical anthropology. In this study, we aimed to develop and evaluate machine learning (ML) methods based on the modified Gustafson's criteria for dental age estimation. In this retrospective study, a total of 851 orthopantomograms were collected from patients aged 15 to 40 years old. The secondary dentin formation (SE), periodontal recession (PE), and attrition (AT) of four mandibular premolars were analyzed according to the modified Gustafson's criteria. Ten ML models were generated and compared for age estimation. The partial least squares regressor outperformed other models in males with a mean absolute error (MAE) of 4.151 years. The support vector regressor (MAE = 3.806 years) showed good performance in females. The accuracy of ML models is better than the single-tooth model provided in the previous studies (MAE = 4.747 years in males and MAE = 4.957 years in females). The Shapley additive explanations method was used to reveal the importance of the 12 features in ML models and found that AT and PE are the most influential in age estimation. The findings suggest that the modified Gustafson method can be effectively employed for adult age estimation in the southwest Chinese population. Furthermore, this study highlights the potential of machine learning models to assist experts in achieving accurate and interpretable age estimation.

Predicting occult lymph node metastasis in solid-predominantly invasive lung adenocarcinoma across multiple centers using radiomics-deep learning fusion model.

BACKGROUND: In solid-predominantly invasive lung adenocarcinoma (SPILAC), occult lymph node metastasis (OLNM) is pivotal for determining treatment strategies. This study seeks to develop and validate a fusion model combining radiomics and deep learning to predict OLNM preoperatively in SPILAC patients across multiple centers. METHODS: In this study, 1325 cT1a-bN0M0 SPILAC patients from six hospitals were retrospectively analyzed and divided into pathological nodal positive (pN+) and negative (pN-) groups. Three predictive models for OLNM were developed: a radiomics model employing decision trees and support vector machines; a deep learning model using ResNet-18, ResNet-34, ResNet-50, DenseNet-121, and Swin Transformer, initialized randomly or pre-trained on large-scale medical data; and a fusion model integrating both approaches using addition and concatenation techniques. The model performance was evaluated by the area under the receiver operating characteristic (ROC) curve (AUC). RESULTS: All patients were assigned to four groups: training set (n = 470), internal validation set (n = 202), and independent test set 1 (n = 227) and 2 (n = 426). Among the 1325 patients, 478 (36%) had OLNM (pN+). The fusion model, combining radiomics with pre-trained ResNet-18 features via concatenation, outperformed others with an average AUC (aAUC) of 0.754 across validation and test sets, compared to aAUCs of 0.715 for the radiomics model and 0.676 for the deep learning model. CONCLUSION: The radiomics-deep learning fusion model showed promising ability to generalize in predicting OLNM from CT scans, potentially aiding personalized treatment for SPILAC patients across multiple centers.