搜索结果：Distilling

Immunotherapy has transformed the therapeutic landscape of advanced cervical cancer, yet clinical benefit remains limited by a highly heterogeneous and immunosuppressive tumor microenvironment. Traditional paradigms, including binary M1/M2 macrophage polarization and models that interpret T-cell dysfunction solely through checkpoint expression, are insufficient to capture the localized intercellular dynamics that drive immune evasion. Recent advances in single-cell and spatial multi-omics have fundamentally reshaped our understanding of this landscape. In this review, we synthesize emerging high-dimensional atlases to reframe macrophage-T-cell crosstalk from simple ligand-receptor interactions into a spatially organized ecological model. We highlight the paradigm shift toward highly resolved myeloid programs, particularly SPP1+ and C1QC+ macrophage states, and discuss how these programs interact with stromal barriers, regulatory T cells, and metabolic checkpoints to restrict, exclude, or functionally constrain effector T cells within suppressive niches. Crucially, we position persistent high-risk human papillomavirus infection not merely as an initiating carcinogenic trigger, but as an upstream and continuous programmer that rewires innate immune sensing, including context-dependent cGAS-STING-related circuits, to stabilize local immune tolerance throughout disease progression. Finally, we propose translational strategies for distilling complex multi-omic atlases into pathology-compatible prognostic and predictive biomarker signatures. Ultimately, by deciphering these spatially organized networks, this review aims to provide actionable translational insights for targeting macrophage vulnerabilities, guiding biomarker-driven combinatorial immunotherapies, and overcoming immune resistance in cervical cancer.

Optical remote sensing object detection faces challenges such as large variations in scale, slender and direction-sensitive targets, complex backgrounds, and limited deployment resources. This paper proposes a lightweight geometrically decoupled student network with a dynamic multi-teacher distillation framework. Based on YOLO11n, the student detector keeps the original classification branch while redesigning the regression branches at different scales. Lightweight regression towers are used for the shallow and deep branches, whereas a geometrically decoupled regression tower is introduced only at the intermediate branch to enhance localization for slender and orientation-sensitive objects with limited extra cost. A geometry-adaptive box loss is further employed to stabilize localization training. For knowledge transfer, three specialized teachers are constructed for semantic classification, geometric regression, and structural topology supervision. A branch-decoupled adaptive weighting strategy dynamically integrates their complementary knowledge for classification and regression distillation. Experiments on DIOR show that the proposed model reduces parameters by 6.8% and GFLOPs by 13.9%, while improving mAP50 by 0.23 percentage points over YOLO11n. Validation on NWPU VHR-10 and deployment tests using PT, ONNX, and TensorRT further demonstrate improved accuracy-efficiency trade-offs and practical inference acceleration.

Generative data augmentation based on diffusion models has emerged as a promising approach for few-shot image classification. Existing methods, such as DA-Fusion, typically follow a "generate-once, use-directly" paradigm, which often suffers from uncontrollable generation quality, unstable semantic consistency, insufficient global diversity, and high sample redundancy. To address these limitations, we propose a two-stage Propose-and-Select framework for controllable data augmentation. This framework curates high-quality synthetic data offline, ensuring that no additional training overhead is introduced to downstream models. For selector optimization, our method eliminates the need for additional human annotations by leveraging the zero-shot prior knowledge of a vision-language model (CLIP) to construct relative-quality pseudo-labels. Furthermore, we develop an adaptive-temperature listwise ranking distillation objective to transfer quality-aware supervision effectively. We also introduce a multi-objective consistency regularization strategy to stabilize training and improve convergence. Under a strictly controlled augmentation budget, where all methods are provided with the same number of synthetic samples, the proposed approach consistently outperforms existing diffusion-based augmentation baselines across both few-shot classification benchmarks, achieving an accuracy of 79.58% on PASCAL VOC and 80.74% on the fine-grained Oxford 102 Flowers dataset. These results demonstrate the effectiveness of the proposed generation-selection paradigm in improving the quality, diversity, and semantic relevance of synthetic samples, thereby enhancing downstream few-shot classification performance.

Salinity stress poses a significant challenge to the productivity of maize, adversely affecting crop growth and development. The present study was conducted to elucidate the potential protective role of phenylalanine against salinity stress by modulating morphological traits, antioxidant defense mechanisms, and ionic balance. The study comprises two factors, including (i) salinity stress: NS = No salinity (control) SS = Salinity stress at 100 mM (sodium chloride), and phenylalanine foliar applications: NS = No spray, DWS = Distilled water spray, PAN10 = 10 mM, PAN20 = 20 mM, PAN30 = 30 mM, and PAN40 = 40 mM spray of phenylalanine. The results revealed that salinity stress caused a significant reduction in the morphological attributes (SFW by 47% and RFW by 39%) and photosynthetic pigments (chlorophyll a by 14% and chlorophyll b by 8%), while the foliar-applied PAN40 ameliorated the stress induce toxicity and improved these morphological attributes by 49% and 43% and photosynthetic pigments by 32% and 47%, respectively as compared to control conditions. Salinity stress increased the production of stress indicators such as MDA by 10% and H2O2 by 19%, while foliar application of PAN40 improved the activities of enzymatic and non-enzymatic antioxidants such as POD by 38%, CAT by 36%, and AsA by 38% and reduced the production of MDA by 10% and H2O2 by 19% under salinity stress. In crux, the foliar-applied PNA (40 mM) showed better performance under stress by improving the morphological and biochemical attributes to cope with the toxic effects of stress. The application of PAN may serve as a scalable, sustainable strategy to impart salinity stress tolerance in maize, improving both agronomic and biochemical resilience.

Rapid nurse decision-making is needed to detect patient deterioration and prevent mortality. Current approaches to support nurses' decisions involve diagnostic data processing and providing a decision with little explanation. Our team aimed to demonstrate the utility of attention architecture to model sequential nurse-patient care actions. Experienced nurses and students completed patient care simulations. Nurse actions were systematically coded and analyzed using our model, consisting of an attention encoder to sequentially process and predict nurse behavior. Performance of our model was compared to recurrent neural networks and long-short term memory models based on accuracy, precision, recall, and F1 score. Behavioral data from 24 nurses (11 experienced nurses and 13 nursing students) were collected during patient care simulations. Nineteen unique types of actions were distilled down to 8 common actions. There were 33 episodes captured (i.e., 33 unique sequences of patient care actions), including a total of 1024 actions (i.e., an average of 31 ± 11 actions). Results showed that the attention model outperformed the other models on all metrics except for precision. Our team demonstrated that machine learning can model sequential nurse actions. These results could be leveraged to provide real-time guidance to support novice nurses' decision-making in the simulated environment.

This study assessed the effect of solvent aging on Vickers microhardness (VHN) and fracture toughness (KIC) of four resin composites: two high-strength flowable materials indicated for all cavity classes (GUF and CBF), a packable nanohybrid composite (XTE), and its flowable counterpart (XTEF). Disk specimens (n = 6 per group) were photocured and stored at 37 °C in distilled water or in a 75%/25% ethanol/water solution for 1 or 30 days. Vickers microhardness was recorded with a 300 g load applied for 15 s. Single-edge-notched beam (SENB) specimens (32 × 6 × 3 mm; n = 6 per group) were loaded in three-point bending for KIC determination after 1 and 30 days of storage in water or ethanol/water. Data were analyzed by three-way ANOVA followed by Tukey's post hoc test (α = 0.05). At 1 day, the four composites separated into distinct VHN groups in the order XTE > GUF > CBF > XTEF (p ≤ 0.05). After 30 d, VHN decreased in all materials, with larger reductions in ethanol/water (17-30% relative to the 1 d water value) than in water alone (5-11%). At 1 day, KIC values for XTE, GUF and CBF formed a single statistical group, all significantly higher than XTEF (p ≤ 0.05). Thirty-day water storage did not affect KIC for any material (p > 0.05), whereas ethanol/water storage reduced KIC by 21-26% in all four composites and produced four distinct material groups (p ≤ 0.05). The high-strength flowable composites showed greater hardness and toughness than a conventional flowable composite but did not match the mechanical performance of the highly filled packable composite. Ethanol/water aging markedly softened the composite surfaces and reduced fracture toughness, whereas prolonged water storage had a smaller effect on hardness and no measurable effect on KIC.

Understanding gene spatial expression and the organization of multicellular systems is vital for disease diagnosis and studying biological processes. However, existing models often struggle to integrate gene expression data with cellular spatial information effectively. Here we introduce SpatialFormer, a hybrid framework combining convolutional networks and transformers to learn single-cell multimodal and multiscale information in the niche context, including expression data and subcellular gene spatial distribution. Pretrained on 700 million cell pairs from 17 million spatially resolved single cells across 71 Xenium slides, SpatialFormer merges gene spatial expression profiles with cell niche information via the pairwise training strategy. Our findings demonstrate that SpatialFormer distills biological signals across various tasks, including single-cell batch correction, cell-type annotation and co-localization detection. The perturbation analysis identified gene pairs essential for the immune cell-cell communication in pulmonary fibrosis, epithelial-myoepithelial co-localization and tumor transition signals in breast cancer. These advancements enhance our understanding of cellular dynamics and offer additional pathways for applications in biomedical research.

A robust and highly sensitive LC-MS/MS method was developed and validated for the quantification of diosgenin in mice plasma in accordance with ICH M10 guidelines. Chromatographic separation was achieved using Kinetex C8 column (50 x 2.1 mm, 5 µm) with 0.1% formic acid in distilled water and methanol as mobile phase at a flow rate of 0.6 mL/min, and the run time was 17 min. Detection was performed in positive ESI+ mode using MRM at m/z 415.10 → 271.20, employing sarsasapogenin as the internal standard. Protein precipitation yielded efficient extraction with minimal matrix interference. The method exhibited excellent linearity over 3.125-1000 ng/mL (r2 = 0.996) with an LLOQ of 3.125 ng/mL. The accuracy during within-day and between-day evaluations ranged from 91.7% to 110.99%, and precision (%CV) remained below 5.14%. Dilution integrity indicated that the analyte can be reliably quantified in biological samples up to 24,000 ng/mL. The stability studies confirmed that the analyte was stable under various conditions. Further, this validated method was applied for pharmacokinetic studies in female mice via oral (10mg/kg) and intramuscular (5mg/kg) administration. Overall, the method is reliable for trace-level quantification and demonstrated its applicability in pharmacokinetic, bioavailability, and formulation studies of diosgenin.

This investigation demonstrates the potential use of coffee parchment (CP) thermally treated at 500 or 1000&#xb0;C (denoted as CP500 or CP1000, respectively) in removing gadolinium (Gd) ions. Their physicochemical characteristics were assessed. In addition, the Gd ion adsorption capacity of the samples was evaluated against certain factors: starting concentration, adsorption operating temperature, reaction time, pH, and the presence of interfering cations. Adsorption capacity was in the following order: CP < CP500 < CP1000. Higher initial concentrations and higher pH values, and lower adsorption temperatures, improved the adsorption of Gd ions. Elemental distribution studies were conducted to understand the adsorption mechanism. Based on the proposed adsorption mechanism, the adsorption of Gd ions on CP1000 may be related to the physicochemical characteristics of the CP1000 surface, specifically its specific surface area. Moreover, the effects of the coexistence of interfering cations on the adsorption of Gd ions were evaluated. Some trivalent cations significantly reduced adsorption compared with mono- and/or divalent cations. Next, the desorption of Gd ions was also assessed using distilled water, a hydrochloric acid (HCl) solution, and a sodium hydroxide solution at different concentrations. The results indicated that the optimal condition for the desorption (recovery) of Gd ions was the use of 100 mmol/L of the HCl solution, with three adsorption-desorption cycles producing favorable results. Finally, CP1000 showed potential use for the removal of Gd-based contrast agents from aqueous phases. Overall, this study suggests that the prepared CP1000 can be used for the adsorption-desorption of Gd ions from water solutions.

Litchi, a fruit of considerable commercial value, is highly susceptible to postharvest deterioration. Chemical preservatives pose potential risks to consumer health and the environment. Therefore, there is an urgent demand for natural preservative alternatives. This study investigates the efficacy of an antimicrobial peptide extract (AMS) as a natural preservative for litchi. Fruit samples were immersed in distilled water (control), sodium hypochlorite, prochloraz, or AMS solutions at varying concentrations (0.5 and 1.0&#x2009;mg&#x2009;mL-1), then stored at room temperature. Our results indicated that 1.0&#x2009;mg&#x2009;mL-1 AMS predominantly maintained the pericarp polyphenol and flavonoid contents during the middle stage of storage and inhibited peroxidase activity, whereas 0.5&#x2009;mg&#x2009;mL-1 AMS exerted similar protective effects on polyphenols and flavonoids during the late stage and suppressed polyphenol oxidase activity. Compared to the control group, 0.5&#x2009;mg&#x2009;mL-1 AMS effectively delayed the reduction in vitamin C content of pulp and the color deterioration of the pericarp, and 1.0&#x2009;mg&#x2009;mL-1 AMS significantly reduced the microbial load on the litchi pericarp, thereby inhibiting microbial-induced decay. The results of correlation analysis and principal component analysis indicate a strong link between microbial load and quality deterioration in litchi. As a biological preservative, AMS may preserve litchi fruits by inhibiting microbial growth, along with other mechanisms, with a different preservation effect on fruit quality compared to traditional chemical preservatives. The results indicate that AMS can be applied as a safe and effective natural preservative for postharvest storage and preservation of litchi. This study contributes to enhancing food safety in litchi and supports the development of green food preservation technologies. &#xa9; 2026 Society of Chemical Industry.

[This corrects the article DOI: 10.3389/fpls.2026.1852592.].

Wearable sensor-based gait analysis has attracted increasing attention for diverse applications in industrial fields, such as healthcare, rehabilitation, sports science, and continuous monitoring for worker injury prevention. Among gait-related signals, ground reaction force (GRF) is a key indicator for understanding human locomotion and is conventionally measured using force plates or instrumented treadmills. However, such systems are costly and limited to controlled laboratory environments, which restrict their use in daily-life settings. Wearable insole sensors provide a more portable alternative for GRF acquisition, but their measurements are often noisy and less reliable, making accurate estimation challenging. Although deep learning models have shown promise in improving GRF estimation from wearable sensor data, high-performing models are often computationally expensive and difficult to deploy on resource-constrained devices. Moreover, since gait signals exhibit phase-specific temporal structures, effective estimation requires the model to develop a semantically meaningful understanding of the gait sequence rather than merely matching signal values. In this regard, human-interpretable semantic information can provide useful guidance for modeling such structured temporal dynamics. To address these issues, we propose a Gait-Semantic Relational Knowledge Distillation (SRKD) framework for GRF estimation using wearable insole sensor data. The proposed method introduces gait-phase-aware semantic guidance into the teacher-student distillation process by decomposing intermediate representations into phase-specific components and aligning them with corresponding textual semantic representations. This phase-aware distillation framework uses textual semantics as anchors to bridge the teacher-student representational gap, enabling the student to effectively learn the structured temporal relational knowledge underlying gait dynamics. We evaluate SRKD under several experimental configurations, including different teacher-student architectures, temporal window lengths, text encoders, and human-refined semantic descriptions. The experimental results show that SRKD generally improved the estimation performance of the evaluated lightweight student models compared with the considered baseline methods. Qualitative analyses also indicate that the predicted GRF signals more closely captured both the overall waveform and phase-dependent temporal variations in the evaluated settings. Furthermore, the results obtained using human-refined descriptions suggest that human-guided semantics can enhance the effectiveness of knowledge distillation by providing more task-relevant semantic guidance. Further validation across broader populations, sensor configurations, and real-world environments is required to assess the generalizability and practical applicability of the proposed framework.

To enhance unmanned ground vehicle (UGV) intelligence in smart cities, disaster rescue, and infrastructure inspection, this paper investigates the collaborative optimization of multimodal fusion end-to-end architectures. Through dynamic alignment of heterogeneous features, a multi-head distillation attention mechanism, and parallel decision-planning, a high-precision, low-latency closed-loop autonomous navigation framework is constructed. A Multi-Head Distillation Attention-based Trajectory Prediction (MDA-TP) method is proposed, combining a BEVFormer-based multimodal fusion perception model with a two-stage progressive knowledge distillation framework. On NuScenes, the method achieves an ADE of 0.88 m, an FDE of 1.32 m (4.34% and 10.81% reductions), and a collision rate of 15.2%, with 42.6 M parameters and 46 ms latency. Ablation shows removing attention distillation increases FDE by 13.6%. For system validation, a multi-sensor UGV platform is built. Through NuScenes online testing and real-world closed-loop validation, the Euclidean deviation remains within 0.5 m. Compared with traditional distillation, speed prediction MSE is reduced by 51.5%, wheel angle RMSE by 58.4%, and route completion improves from 60.99% to 97.26%. The results provide practical support for autonomous driving in smart cities, disaster rescue, and infrastructure inspection.

UAV-borne visual sensors provide high-resolution aerial observations for low-altitude scene understanding, urban monitoring, traffic observation, emergency inspection, and infrastructure assessment. However, semantic perception from UAV visual sensor data remains challenging because aerial images often contain dense small objects, elongated road structures, fragmented boundaries, scale variations caused by flight-altitude changes, oblique viewpoints, and strict onboard or edge computational constraints. To address these challenges, this paper proposes MEKD-UAVSeg, a lightweight semantic perception framework based on conflict-suppressed heterogeneous expert distillation. During training, a Transformer-based semantic expert provides global contextual understanding and region-level class consistency, while a Mamba-based spatial expert provides complementary structural guidance for roads, roofs, boundaries, and other continuous aerial structures. Both experts are used only during training, and the final inference model remains a compact CNN-based segmentation network. In addition, UAV-aware density and hard-region priors are designed to emphasize small-object-dense areas, boundary-sensitive regions, rare classes, and uncertain aerial categories. A conflict-suppressed reliability routing strategy is further developed to reduce inconsistent supervision between heterogeneous experts and selectively transfer reliable knowledge to the student model. Experiments on UAVid and UDD6 demonstrate that the proposed framework achieves a favorable accuracy-efficiency trade-off compared with representative CNN-, Transformer-, Mamba-, and hybrid-based UAV segmentation methods, without introducing expert-induced inference complexity.

Accurate polyp segmentation in colonoscopy supports early detection of colorectal cancer, but compact models under a student-only inference budget tend to lose boundary fidelity. BDKD-Net is a 3.72M-parameter compact student trained under a composite knowledge-distillation loss that combines a response signal, a boundary-probability signal restricted to a static teacher-derived boundary band, and an auxiliary detail-feature alignment; the teacher is used only during training and discarded at inference. In a uniform five-seed re-run on a locked Kvasir-SEG and CVC-ClinicDB development split, the same-architecture scratch student reaches Dev Dice 0.9210 &#xb1; 0.0029 and boundary F1 at 3-pixel tolerance 0.7726 &#xb1; 0.0057, while full BDKD-Net reaches 0.9300 &#xb1; 0.0029 and 0.8008 &#xb1; 0.0094. Boundary-probability distillation is the load-bearing signal: it has the highest mean Dev BF1t3 among the single KD signals and carries the boundary gain at no inference cost, with the full model reaching four-external Dice 0.8225 &#xb1; 0.0071 at 1.76 GFLOPs and 140.9 FPS. Among the directly reproduced baselines, the higher-Dice Polyp-PVT (0.8417 &#xb1; 0.0091) needs 6.8&#xd7; the parameters and 5.7&#xd7; the compute at roughly half the frame rate. BDKD-Net thus delivers a compact, boundary-faithful student that keeps most of the accuracy of much larger models at a fraction of their inference cost.

The proliferation of high-definition video data necessitates highly efficient processing pipelines for real-time edge analytics. However, traditional object detection architectures rely exclusively on pixel-domain inputs, which renders the computationally prohibitive decoding phase a latency bottleneck. In this paper, we propose a novel dual-phase framework designed to achieve fast and efficient object detection directly within the partially decoded compressed-domain data. First, we introduce a partial decoding paradigm featuring the Low-Frequency Spectral Prioritization method on the encoder side. By systematically discarding high-frequency residual coefficients and retaining only a sparse subset of fundamental spatial frequencies, this method dramatically reduces transmission payloads and accelerates the standard decoding process. Second, to recover the structural fidelity lost due to the intentional omission of residual data, we employ a multi-granularity cross-domain knowledge distillation architecture. This strategy aligns global contextual features, foreground boundary attention maps, and final response logits, transferring rich representational capacities from a high-performing pixel-domain teacher network to a lightweight compressed-domain student network. Comprehensive experiments utilizing RetinaNet, FCOS, and GFL object detection networks on the COCO-mini dataset demonstrate the superiority of the proposed framework. By retaining fundamental residual coefficients within the HEVC pipeline, the proposed method reduces average decoding latency while improving the mAP score by +0.96% over the conventional fully decoded pixel-domain baseline on the COCO-mini dataset.

Background: Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by the selective loss of dopaminergic neurons in the substantia nigra. Oxidative stress, neuroinflammation, and &#x3b1;-synuclein aggregation are central pathological features of PD. Zanthoxylum piperitum DC, commonly known as Korean pepper or chopi, is a traditional dietary spice in Eastern Asia and has been reported to possess antioxidant and anti-inflammatory properties. This study investigated the neuroprotective and motor function-enhancing effects of distilled extract of Z. piperitum (deZP) in 1-Methyl-4-phenylpyridinium (MPP+)-treated Caenorhabditis elegans and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced mouse models of PD. Methods: In the C. elegans model, dopaminergic neurotoxicity was induced by MPP+, and deZP was tested at 0.25, 0.5, and 1% (v/v) to evaluate neuronal preservation through GFP-labeled dopaminergic neurons and &#x3b1;-synuclein expression. Concurrently, in the MPTP-induced mouse model, deZP was administered intranasally at a fixed dose of 20 &#x3bc;L/mouse, equivalent to 5 mg/mouse. Motor function was assessed using the rota-rod test, pole test, and grip strength test, while dopaminergic neuronal survival was evaluated by tyrosine hydroxylase (TH) immunostaining. Results: In MPP+-treated C. elegans, deZP significantly restored green fluorescent protein (GFP) fluorescence in dopaminergic neurons and reduced &#x3b1;-synuclein expression, with the most pronounced effects observed at 1% (v/v). In the MPTP-induced mouse model, deZP at this fixed intranasal dose significantly improved motor performance and preserved TH-positive neurons in the substantia nigra. Conclusions: These findings suggest that deZP may represent a promising preclinical candidate for further investigation in PD-related neurodegeneration.

Accurate protein family classification is essential since proteins within the same family share conserved structural domains and biochemical functions that deter mine their biological roles. G protein-coupled receptors (GPCRs) represent one of the largest and most diverse protein families in eukaryotes, serving as targets for ap proximately 35% of FDA-approved drugs. While traditional sequence alignment methods, such as BLAST, provide foundational tools for identifying homologous sequences, they exhibit limited accuracy in distinguishing closely related GPCR families with low sequence homology. Recently, deep learning approaches offer promising accuracy; however, they employ fixed-size classification architectures that force newly discovered protein families into pre-existing categories, preventing the recognition of novel families and limiting scalability as the protein universe expands. In this work, we present a scalable machine learning framework GPCR SLM, that classifies GPCRs across 86 distinct families using a lightweight transformer model optimized through knowledge distillation. Our approach achieved an overall ac curacy of 99%, significantly outperforming BLAST (86.1%) and HMMER (91%), while demonstrating substantial computational efficiency with an average speedup of 33.5&#xd7; compared to large protein language models. These results demonstrate the effectiveness of combining distilled protein language models with flexible classification frameworks for high-resolution functional annotation.

This study evaluated the effects of bioceramic and resin-based root canal sealers and different irrigation protocols on the push-out bond strength (PBS) of fiber posts. Fifty-four freshly extracted mandibular canines were instrumented and irrigated with NaOCl and EDTA. The specimens were first divided into two groups based on the sealer used: a bioceramic sealer (Bioserra) or a resin-based sealer (AH Plus) (n=27 per group). Each group was then randomly subdivided into three subgroups (n=9) based on the irrigation solution used for smear layer removal during post space preparation: distilled water, 17% EDTA, or 0.2% chitosan. Fiber posts were cemented with self-adhesive resin cement. Sections (1.5-mm thick) from the coronal third were subjected to a push-out bond strength test. Data was analyzed using two-way ANOVA and post hoc Bonferroni tests (&#x3b1;=0.05). Both the type of sealer and the irrigation protocol significantly influenced PBS values (P<0.05). The highest PBS was observed in the bioceramic sealer+chitosan group (16.08&#xb1;0.48 MPa), whereas the lowest value was recorded in the bioceramic sealer+distilled water group (11.20&#xb1;0.58 MPa). Root canal sealer type and irrigation protocol during post space preparation significantly affected the bond strength of fiber posts. Chitosan irrigation demonstrated superior performance, suggesting its potential to improve post retention.