Development and antimicrobial activity of composite edible films of chitosan and nisin incorporated with perilla essential oil-glycerol monolaurate emulsions.

Aquatic products are highly susceptible to spoilage, and preparing composite edible film with essential oil is an effective solution. In this study, composite edible films were prepared using perilla essential oil (PEO)-glycerol monolaurate emulsions incorporated with chitosan and nisin, and the film formulation was optimized by response surface methodology. These films were applied to ready-to-eat fish balls and evaluated over a period of 12 days. The films with the highest inhibition rate against Staphylococcus aureus were acquired using a polymer composition of 6 µL/mL PEO, 18.4 µg/mL glycerol monolaurate, 14.2 mg/mL chitosan, and 11.0 µg/mL nisin. The fish balls coated with the optimal edible film showed minimal changes in appearance during storage and significantly reduced total bacterial counts and total volatile basic nitrogen compared to the control groups. This work indicated that the composite edible films containing essential oils possess ideal properties as antimicrobial packaging materials for aquatic foods.

[Early identification of acute kidney injury in children with primary nephrotic syndrome]. / å„¿ç«¥åŽŸå‘æ€§è‚¾ç— ç»¼åˆå¾å¹¶å‘æ€¥æ€§è‚¾æŸä¼¤çš„æ—©æœŸè¯†åˆ«.

OBJECTIVES: To investigate the incidence and risk factors for acute kidney injury (AKI) in children with primary nephrotic syndrome (PNS), as well as the role of neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule-1 (KIM-1) in the early identification of AKI in these children. METHODS: A prospective collection of clinical data from children hospitalized with PNS at the Children's Hospital of the Capital Institute of Pediatrics from January 2021 to October 2022 was conducted. The children were divided into two groups based on the presence of AKI: the AKI group (47 cases) and the non-AKI group (169 cases). The risk factors for AKI in children with PNS were identified by multivariate logistic regression analysis. Urinary KIM-1 and NGAL levels were compared between the AKI and non-AKI groups, as well as among the different stages of AKI. RESULTS: The incidence of AKI in children with PNS was 21.8%. Multivariate logistic regression analysis revealed that steroid-resistant nephrotic syndrome, gastrointestinal infections, and heavy proteinuria were independent risk factors for AKI in these children with PNS (P<0.05). Urinary KIM-1 and NGAL levels were higher in the AKI group compared to the non-AKI group (P<0.05), and the urinary NGAL and KIM-1 levels in the AKI stage 2 and stage 3 subgroups were higher than those in the AKI stage 1 subgroup (P<0.017). CONCLUSIONS: KIM-1 and NGAL can serve as biomarkers for the early diagnosis of AKI in children with PNS. Identifying high-risk populations for AKI in children with PNS and strengthening the monitoring of related risk factors is of significant importance.

Image-Level Adaptive Adversarial Ranking for Person Re-identification.

The potential vulnerability of deep neural networks and the complexity of pedestrian images, greatly limits the application of person re-identification techniques in the field of smart security. Current attack methods often focus on generating carefully crafted adversarial samples or only disrupting the metric distances between targets and similar pedestrians. However, both aspects are crucial for evaluating the security of methods adapted for person re-identification tasks. For this reason, we propose an image-level adaptive adversarial ranking method that comprehensively considers two aspects to adapt to changes in pedestrians in the real world and effectively evaluate the robustness of models in adversarial environments. To generate more refined adversarial samples, our image representation enhancement module leverages channel-wise information entropy, assigning varying weights to different channels to produce images with richer information content, along with a generative adversarial network to create adversarial samples. Subsequently, for adaptive perturbation of ranking, the adaptive weight confusion ranking loss is presented to calculate the weights of distances between positive or negative samples and query samples. It endeavors to push positive samples away from query samples and bring negative samples closer, thereby interfering with the ranking of system. Notably, this method requires no additional hyperparameter tuning or extra data training, making it an adaptive attack strategy. Experimental results on large-scale datasets such as Market1501, CUHK03, and DukeMTMC demonstrate the effectiveness of our method in attacking ReID systems.

Local artifacts amplification for deepfakes augmentation.

With the rapid and continuous development of AIGC, It is becoming increasingly difficult to distinguish between real and forged facial images, which calls for efficient forgery detection systems. Although many detection methods have noticed the importance of local artifacts, there has been a lack of in-depth discussion regarding the selection of locations and their effective utilization. Besides, the traditional image augmentation methods that are widely used have limited improvements for forgery detection tasks and require more specialized augmentation methods specifically designed for forgery detection tasks. In this paper, this study proposes Local Artifacts Amplification for Deepfakes Augmentation, which amplifies the local artifacts on the forged faces. Furthermore, this study incorporates prior knowledge about similar facial features into the model. This means that within the facial regions defined in this work, forged features exhibit similar patterns. By aggregating the results from all facial regions, the study can enhance the overall performance of the model. The evaluation experiments conducted in this research, achieving an AUC of 93.40% and an Acc of 87.03% in the challenging WildDeepfake dataset, demonstrate a promising improvement in accuracy compared to traditional image augmentation methods and achieve superior performance on intra-dataset evaluation. The cross-dataset evaluation also showed that the method presented in this study has strong generalization abilities.

A Membrane-Anchoring Self-Assembling Peptide Allows Bioorthogonal Coupling of Type-I AIEgens for Pyroptosis-Induced Cancer Therapy.

Enrichment of photosensitizers (PSs) on cancer cell membranes via bioorthogonal reactions is considered to be a very promising therapeutic modality. However, azide-modified sugars-based metabolic labeling processes usually lack targeting and the labeling speed is relatively slow. Moreover, it has been rarely reported that membrane-anchoring pure type-I PSs can induce cancer cell pyroptosis. Here, we report an alkaline phosphatase (ALP) and cholecystokinin-2 receptor (CCK2R) dual-targeting peptide named DBCO-pYCCK6, which can selectively and rapidly self-assemble on cancer cell membrane, and then bioorthogonal enrich type-I aggregation-induced emission luminogens (AIEgen) PSs (SAIE-N3) on the cell membrane. Upon light irradiation, the membrane-anchoring SAIE-N3 could effectively generate type-I reactive oxygen species (ROS) to induce gasdermin E (GSDME)-mediated pyroptosis. In vivo experiments demonstrated that the bioorthogonal combination strategy of peptide and AIEgen PSs could significantly inhibit tumor growth, which is accompanied by CD8+ cytotoxic T cell infiltration. This work provides a novel self-assembly peptide-mediated bioorthogonal reaction strategy to bridge the supramolecular self-assembly and AIE field through strain-promoted azide-alkyne cycloaddition (SPAAC) and elucidates that pure type-I membrane-anchoring PSs can be used for cancer therapy via GSDME-mediated pyroptosis.

Increased tensile strength induced by the precipitation of nanocrystals for welding joints of Zr-based amorphous alloys.

Zr-based amorphous alloys have attracted intensive attention for applications because of their excellent mechanical property. However, the welding process is inevitable for some special cases, such as the obtain of large size structure parts. It is significant to clarify the influence of introduced welding joints on mechanical properties in Zr-based amorphous alloys. Herein, the increased tensile strength of welding joints in Zr-based amorphous alloys is demonstrated by choosing a suitable initial temperature of Cu cooling fixtures for pulsed laser welding. It is found that an optimized tensile strength is observed when the initial temperature is -20 °C. With the decrease of the initial temperature from 10 to -30 °C, the tensile strength shows a trend of first increasing and then decreasing. Combined with the characterization of microstructures, it can be concluded that the increased tensile strength results from the precipitation of nanocrystals in the heat affected zone. Thus, our results provide a method to improve the mechanical property by controlling the microstructures of the heat affected zone in welding joints of Zr-based amorphous alloys.

Inspector for Face Forgery Detection: Defending Against Adversarial Attacks From Coarse to Fine.

The emergence of face forgery has raised global concerns on social security, thereby facilitating the research on automatic forgery detection. Although current forgery detectors have demonstrated promising performance in determining authenticity, their susceptibility to adversarial perturbations remains insufficiently addressed. Given the nuanced discrepancies between real and fake instances are essential in forgery detection, previous defensive paradigms based on input processing and adversarial training tend to disrupt these discrepancies. For the detectors, the learning difficulty is thus increased, and the natural accuracy is dramatically decreased. To achieve adversarial defense without changing the instances as well as the detectors, a novel defensive paradigm called Inspector is designed specifically for face forgery detectors. Specifically, Inspector defends against adversarial attacks in a coarse-to-fine manner. In the coarse defense stage, adversarial instances with evident perturbations are directly identified and filtered out. Subsequently, in the fine defense stage, the threats from adversarial instances with imperceptible perturbations are further detected and eliminated. Experimental results across different types of face forgery datasets and detectors demonstrate that our method achieves state-of-the-art performances against various types of adversarial perturbations while better preserving natural accuracy. Code is available on https://github.com/xarryon/Inspector.

Enhanced photoelectrochemical water splitting performance of α-Fe2O3 photoanodes through Co-modification with Co single atoms and g-C3N4.

The practical application of α-Fe2O3 in water splitting is hindered by significant charge recombination and slow water oxidation. To address this issue, a CoSAs-g-C3N4/Fe2O3 (CoSAs: cobalt single atoms) photoanode was fabricated in this study through the co-modification of CoSAs and g-C3N4 to enhance photoelectrochemical (PEC) water splitting. The coupling between g-C3N4 and α-Fe2O3 resulted in the formation of a heterojunction, which provided a strong built-in electric field and an additional driving force to mitigate charge recombination. Moreover, g-C3N4 served as a suitable carrier for single atoms, which effectively anchored CoSAs through N/C coordination. The highly dispersed CoSAs provided abundant active sites, which further promoted surface holes extraction and oxidation kinetics, resulting in higher PEC performance and photostability. This study indicates the benefits of these collaborative strategies and provides more efficient designs for solar energy conversion in PEC systems.

Quantitative Assessment and Regulation of Passage and Entrance Attraction Efficiency of Vertical-Slot Fishway on Heishuihe River in Southwest China.

Fish passage facilities are essential for restoring river connectivity and protecting ecosystems, effectively balancing economic and ecological benefits. Systematic and comprehensive monitoring, assessment, and optimized management are therefore crucial. This study quantitatively evaluated the entire upstream migration process of fish from the downstream river to the entrance and exit of the fishway and investigated the upstream movement patterns of fish under various environmental factors. A total of 19 fish species were monitored in the Heishuihe River downstream of the dam, with 15 species reaching the fishway entrance and 12 species successfully passing through it. The entrance attraction and passage rates of the vertical-slot fishway at the Songxin hydropower station were 15.7% and 40.42%, respectively. The best upstream performance was observed in May, with fish demonstrating better upstream timing and speed during nighttime compared to daytime. Specifically, the highest entrance attraction efficiency was recorded at a flow rate of 6-7 m3/s and a temperature of 19-20 °C, while the optimal passage efficiency was observed at a flow rate of 0-0.5 m3/s and a temperature of 17-20 °C. Additionally, a multifactorial Cox proportional hazards regression model was constructed to identify key factors influencing the probability of fishway entrance attraction and successful passage. The model elucidated the impact patterns of these key factors on fish upstream migration, ultimately generating an alignment diagram for prediction and control. This study provides a theoretical foundation and data support for developing optimized operational schedules for fishways. The findings offer a more comprehensive and systematic approach for monitoring and evaluating fish passage facilities, serving as a scientific basis for ecological restoration and fish conservation in this region and similar areas.

GazeForensics: DeepFake detection via gaze-guided spatial inconsistency learning.

DeepFake detection is pivotal in personal privacy and public safety. With the iterative advancement of DeepFake techniques, high-quality forged videos and images are becoming increasingly deceptive. Prior research has seen numerous attempts by scholars to incorporate biometric features into the field of DeepFake detection. However, traditional biometric-based approaches tend to segregate biometric features from general ones and freeze the biometric feature extractor. These approaches resulted in the exclusion of valuable general features, potentially leading to a performance decline and, consequently, a failure to fully exploit the potential of biometric information in assisting DeepFake detection. Moreover, insufficient attention has been dedicated to scrutinizing gaze authenticity within the realm of DeepFake detection in recent years. In this paper, we introduce GazeForensics, an innovative DeepFake detection method that utilizes gaze representation obtained from a 3D gaze estimation model to regularize the corresponding representation within our DeepFake detection model, while concurrently integrating general features to further enhance the performance of our model. Experimental results demonstrate that our proposed GazeForensics method performs admirably in terms of performance and exhibits excellent interpretability.

Tubal replacement of cervix for treatment of type II vaginal atresia complicated with cervical dysplasia: a case report.

Vaginal atresia is a rare obstructive disease of the reproductive tract. It is characterized by the absence or underdevelopment of the vaginal canal and results in various clinical manifestations. Hysterectomy can physically and mentally burden young female patients with a congenital cervix and complete vaginal atresia. This report presents a case of type II vaginal atresia complicated by cervical dysplasia in a female patient >10 years of age. Our team opted to preserve the patient's uterus, innovated a fallopian tube transplantation technique, and performed cervicovaginal reconstruction using natural channels instead of the cervical canal. The patient experienced menarche within the first 2 weeks postoperatively, and follow-up at 6 months revealed no abnormalities.

Genetic Architecture and Analysis Practices of Circulating Metabolites in the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program.

Circulating metabolite levels partly reflect the state of human health and diseases, and can be impacted by genetic determinants. Hundreds of loci associated with circulating metabolites have been identified; however, most findings focus on predominantly European ancestry or single study analyses. Leveraging the rich metabolomics resources generated by the NHLBI Trans-Omics for Precision Medicine (TOPMed) Program, we harmonized and accessibly cataloged 1,729 circulating metabolites among 25,058 ancestrally-diverse samples. We provided recommendations for outlier and imputation handling to process metabolite data, as well as a general analytical framework. We further performed a pooled analysis following our practical recommendations and discovered 1,778 independent loci associated with 667 metabolites. Among 108 novel locus - metabolite pairs, we detected not only novel loci within previously implicated metabolite associated genes, but also novel genes (such as GAB3 and VSIG4 located in the X chromosome) that have putative roles in metabolic regulation. In the sex-stratified analysis, we revealed 85 independent locus-metabolite pairs with evidence of sexual dimorphism, including well-known metabolic genes such as FADS2 , D2HGDH , SUGP1 , UTG2B17 , strongly supporting the importance of exploring sex difference in the human metabolome. Taken together, our study depicted the genetic contribution to circulating metabolite levels, providing additional insight into the understanding of human health.

Rectified Binary Network for Single-Image Super-Resolution.

Binary neural network (BNN) is an effective approach to reduce the memory usage and the computational complexity of full-precision convolutional neural networks (CNNs), which has been widely used in the field of deep learning. However, there are different properties between BNNs and real-valued models, making it difficult to draw on the experience of CNN composition to develop BNN. In this article, we study the application of binary network to the single-image super-resolution (SISR) task in which the network is trained for restoring original high-resolution (HR) images. Generally, the distribution of features in the network for SISR is more complex than those in recognition models for preserving the abundant image information, e.g., texture, color, and details. To enhance the representation ability of BNN, we explore a novel activation-rectified inference (ARI) module that achieves a more complete representation of features by combining observations from different quantitative perspectives. The activations are divided into several parts with different quantification intervals and are inferred independently. This allows the binary activations to retain more image detail and yield finer inference. In addition, we further propose an adaptive approximation estimator (AAE) for gradually learning the accurate gradient estimation interval in each layer to alleviate the optimization difficulty. Experiments conducted on several benchmarks show that our approach is able to learn a binary SISR model with superior performance over the state-of-the-art methods. The code will be released at https://github.com/jwxintt/Rectified-BSR.

Protecting Prostate Cancer Classification From Rectal Artifacts via Targeted Adversarial Training.

Magnetic resonance imaging (MRI)-based deep neural networks (DNN) have been widely developed to perform prostate cancer (PCa) classification. However, in real-world clinical situations, prostate MRIs can be easily impacted by rectal artifacts, which have been found to lead to incorrect PCa classification. Existing DNN-based methods typically do not consider the interference of rectal artifacts on PCa classification, and do not design specific strategy to address this problem. In this study, we proposed a novel Targeted adversarial training with Proprietary Adversarial Samples (TPAS) strategy to defend the PCa classification model against the influence of rectal artifacts. Specifically, based on clinical prior knowledge, we generated proprietary adversarial samples with rectal artifact-pattern adversarial noise, which can severely mislead PCa classification models optimized by the ordinary training strategy. We then jointly exploited the generated proprietary adversarial samples and original samples to train the models. To demonstrate the effectiveness of our strategy, we conducted analytical experiments on multiple PCa classification models. Compared with ordinary training strategy, TPAS can effectively improve the single- and multi-parametric PCa classification at patient, slice and lesion level, and bring substantial gains to recent advanced models. In conclusion, TPAS strategy can be identified as a valuable way to mitigate the influence of rectal artifacts on deep learning models for PCa classification.

Directed differentiation of pancreatic δ cells from human pluripotent stem cells.

Dysfunction of pancreatic δ cells contributes to the etiology of diabetes. Despite their important role, human δ cells are scarce, limiting physiological studies and drug discovery targeting δ cells. To date, no directed δ-cell differentiation method has been established. Here, we demonstrate that fibroblast growth factor (FGF) 7 promotes pancreatic endoderm/progenitor differentiation, whereas FGF2 biases cells towards the pancreatic δ-cell lineage via FGF receptor 1. We develop a differentiation method to generate δ cells from human stem cells by combining FGF2 with FGF7, which synergistically directs pancreatic lineage differentiation and modulates the expression of transcription factors and SST activators during endoderm/endocrine precursor induction. These δ cells display mature RNA profiles and fine secretory granules, secrete somatostatin in response to various stimuli, and suppress insulin secretion from in vitro co-cultured ß cells and mouse ß cells upon transplantation. The generation of human pancreatic δ cells from stem cells in vitro would provide an unprecedented cell source for drug discovery and cell transplantation studies in diabetes.

Contact-electro-catalytic CO2 reduction from ambient air.

Traditional catalytic techniques often encounter obstacles in the search for sustainable solutions for converting CO2 into value-added products because of their high energy consumption and expensive catalysts. Here, we introduce a contact-electro-catalysis approach for CO2 reduction reaction, achieving a CO Faradaic efficiency of 96.24%. The contact-electro-catalysis is driven by a triboelectric nanogenerator consisting of electrospun polyvinylidene fluoride loaded with single Cu atoms-anchored polymeric carbon nitride (Cu-PCN) catalysts and quaternized cellulose nanofibers (CNF). Mechanistic investigation reveals that the single Cu atoms on Cu-PCN can effectively enrich electrons during contact electrification, facilitating electron transfer upon their contact with CO2 adsorbed on quaternized CNF. Furthermore, the strong adsorption of CO2 on quaternized CNF allows efficient CO2 capture at low concentrations, thus enabling the CO2 reduction reaction in the ambient air. Compared to the state-of-the-art air-based CO2 reduction technologies, contact-electro-catalysis achieves a superior CO yield of 33 µmol g-1 h-1. This technique provides a solution for reducing airborne CO2 emissions while advancing chemical sustainability strategy.

A Nanoreactor Based on Metal-Organic Frameworks With Triple Synergistic Therapy for Hepatocellular Carcinoma.

The transformation of monotherapy into multimodal combined targeted therapy to fully exploit synergistic efficacy is of increasing interest in tumor treatment. In this work, a novel nanodrug-carrying platform based on iron-based MOFs, which is loaded with doxorubicin hydrochloride (DOX), dihydroartemisinin (DHA), and glucose oxidase (GOx), and concurrently covalently linked to the photosensitizer 5,10,15,20-tetrakis(4-carboxyphenyl)porphyrin (TCPP) in polydopamine (PDA)-encapsulated MIL-101(Fe) (denoted as MIL-101(Fe)-DOX-DHA@TCPP/GOx@PDA, MDDTG@P), is successfully developed. Upon entering the tumor microenvironment, MDDTG@P catalyzes the hydrogen peroxide (H2O2) into hydroxyl radicals (·OH) and depletes glutathione (GSH); thus, exerting the role of chemodynamic therapy (CDT). The reduced Fe2+ can also activate DHA, further expanding CDT and promoting tumor cell apoptosis. The introduced GOx will rapidly consume glucose and oxygen (O2) in the tumor; while, replenishing H2O2 for Fenton reaction, starving the cancer cells; and thus, realizing starvation and chemodynamic therapy. In addition, the covalent linkage of TCPP endows MDDTG@P with good photodynamic therapeutic (PDT) properties. Therefore, this study develops a nanocarrier platform for triple synergistic chemodynamic/photodynamic/starvation therapy, which has promising applications in the efficient treatment of tumors.

Effect of electromigration on microstructure and properties of CeO2 nanopartical-reinforced Sn58Bi/Cu solder joints.

To mitigate the decrease in mechanical performance of Sn58Bi/Cu solder joints resulting from electromigration-induced damage. The CeO2 nanoparticles were incorporated into Sn58Bi solder by a melt-casting method, and their effects on the microstructure and properties of Sn58Bi/Cu solder joints under electromigration were investigated. The study results demonstrate that the addition of 0.125 ~ 0.5 wt% CeO2 nanoparticles refines the eutectic microstructure of Sn58Bi solder alloy. At an addition amount of 0.5 wt%, the composite solder alloy exhibits the maximum tensile strength of 68.9 MPa, which is 37% higher than that of the base solder. CeO2 nanoparticle-reinforced Sn58Bi solder can achieve excellent solderbility with Cu substrates and the joints can significantly inhibit the growth of the anodic Bi-rich layer, which is responsible for electromigration. With the extension of current stressing time, Bi-rich and Sn-rich layer are respectively formed on the anode and cathode in the joints. The intermetallic compound (IMC) layer grows asymmetrically, transitioning from a fan-shaped morphology to a flattened structure at the anode and to a thickened mountain-like morphology at the cathode. Adding the CeO2 nanoparticles helps to mitigate the decrease in mechanical performance caused by electromigration damage during current application to some extent. Over the current stressing period of 288 ~ 480 h, the fracture position shifts from the anodic IMC/Bi-rich interface to the cathodic Sn-rich/IMC interface. The fracture mechanism transitions from a brittle fracture characterized by plate-like cleavage to a ductile-brittle mixed fracture with fine dimples and cleavage.

Assessing the Post-Activation Performance Enhancement of Upper Limbs in Basketball Athletes: A Sensor-Based Study of Rapid Stretch Compound and Blood Flow Restriction Training.

OBJECTIVE: This study introduces a novel methodology combining rapid stretch compound training with blood flow restriction (BFR) to investigate post activation performance enhancement (PAPE) in basketball players, a field that has been predominantly explored for lower limbs. We aimed to assess the efficacy of this combined approach on upper limb muscle performance in athletes. METHODS: We employed a randomized, self-controlled crossover trial with ten male basketball players. The bench press throw (BPT) served as the primary metric, with players undergoing four interventions post-baseline: (1) STR-plyometric training; (2) BFR-blood flow restriction; (3) COMB-STR integrated with BFR; and (4) CON-control. Innovatively, we utilized an intelligent tracking sensor to precisely measure peak power (PP), peak velocity (PV), mean power (MP), and mean velocity (MV) at 4, 8, and 12 min post-intervention, providing a detailed temporal analysis of PAPE. RESULTS: The COMB intervention demonstrated superior PAPE effects at 4 min, significantly outperforming STR and BFR alone and the control group across all measured indices (p < 0.05). Notably, the COMB group maintained these improvements for PV, PP, and H up to 12 min post-intervention, suggesting a prolonged effect. CONCLUSION: (1) The COMB stimulation has been shown to successfully induce PAPE more effectively than STR and BFR modality alone. (2) It appears that the optimal effects of PAPE are achieved within 4 min of exercising under this COMB. By the 12 min mark, only the COMB group continued to show significant improvements in PV, PP, and H compared to both the baseline and the CON group, while the effects in the STR and BFR groups further diminished. This suggests that although the PAPE effect is maintained over time, its optimal performance may peak at the 4 min mark and then gradually weaken as time progresses.

Multi-Dimensional Fragmentomics Enables Early and Accurate Detection of Colorectal Cancer.

Colorectal cancer (CRC) is frequently diagnosed in advanced stages, highlighting the need for developing approaches for early detection. Liquid biopsy using cell-free DNA (cfDNA) fragmentomics is a promising approach, but the clinical application is hindered by complexity and cost. This study aimed to develop an integrated model using cfDNA fragmentomics for accurate, cost-effective early-stage CRC detection. Plasma cfDNA was extracted and sequenced from a training cohort of 360 participants, including 176 CRC patients and 184 healthy controls. An ensemble stacked model comprising five machine learning models was employed to distinguish CRC patients from healthy controls using five cfDNA fragmentomic features. The model was validated in an independent cohort of 236 participants (117 CRC patients and 119 controls) and a prospective cohort of 242 participants (129 CRC patients and 113 controls). The ensemble stacked model showed remarkable discriminatory power between CRC patients and controls, outperforming all base models and achieving a high area under the ROC curve (AUC) of 0.986 in the validation cohort. It reached 94.88% sensitivity and 98% specificity for detecting CRC in the validation cohort, with sensitivity increasing as cancer progressed. The model also demonstrated consistently high accuracy in within-run and between-run tests and across various conditions in healthy individuals. In the prospective cohort, it achieved 91.47% sensitivity and 95.58% specificity. This integrated model capitalizes on the multiplex nature of cfDNA fragmentomics to achieve high sensitivity and robustness, offering significant promise for early CRC detection and broad patient benefit.