Browse Articles

Discover research articles across all indexed journals

Systematic characteristic evaluation of clay-based cementitious material derived from calcium carbide residue and waste tile powder

Scientific Reports Chuanxin Du, Yanguo Sun Jun 30, 2026 DOI: 10.1038/s41598-026-60193-7

Acoustic feature decoupling and pre-trained language model integration for music-to-text cross-modal generation

Scientific Reports Jiaying Yu Jun 30, 2026 DOI: 10.1038/s41598-026-58108-7

A federated learning framework integrating knowledge graphs and Node2Vec for multi-source medical image classification

Scientific Reports Abdullah Ali Alqarni Jun 30, 2026 DOI: 10.1038/s41598-026-60172-y

Abstract Medical image classification in federated healthcare environments is challenged by the difficulty of learning robust and generalized representations from heterogeneous, non-IID data distributed across multiple institutions. Conventional federated learning approaches primarily rely on visual features and often fail to capture structural relationships among medical images, leading to reduced classification performance and limited generalization across diverse clinical settings. This study proposes GraphMedFL, a federated learning framework that integrates knowledge graphs and Node2Vec-based structural embeddings to enhance classification performance in multi-source environments. Local knowledge graphs are constructed from image features to capture inter-sample relationships, and Node2Vec embeddings are fused with feature representations to enrich local model training. A similarity-aware adaptive aggregation strategy is introduced to address non-IID data distributions across clients. Experiments conducted on three breast cancer imaging datasets (BreakHis, CBIS-DDSM, and INbreast) demonstrate that GraphMedFL achieves superior performance compared to locally trained models, with an accuracy of 98.3% and strong precision-recall balance. The proposed GraphMedFL achieves an accuracy of 98.3% and consistently outperforms baseline federated learning and local training approaches across BreakHis, CBIS-DDSM, and INbreast datasets. It improves precision, recall, and F1-score by a measurable margin compared to conventional methods, demonstrating enhanced robustness and generalization in heterogeneous medical imaging environments.

A nurse-led peer support intervention to enhance decision-making for fecal microbiota transplantation in recurrent UTI: a pilot study

Scientific Reports Hongyuan Liu, Junjie Zhi, Zhou Li et al. Jun 30, 2026 DOI: 10.1038/s41598-026-58218-2

Abstract To examine the preliminary effects and feasibility of peer support on decision-making regarding fecal microbiota transplantation (FMT) among patients with recurrent urinary tract infections (rUTIs). This was a prospective, two-arm pilot study conducted from September 2023 to April 2024 in the urology outpatient departments of two tertiary hospitals in China using convenience sampling. Patients were assigned to either a peer support group or a control group. The intervention consisted of weekly WeChat-based interactions between trained peer supporters and patients for four weeks. Decision-making was assessed using validated tools including the Decisional Conflict Scale, Preparation for Decision Making Scale, Decision Self-Efficacy Scale, Choice Predisposition Scale, and Decisional Satisfaction Scale, all derived from the Ottawa Decision Support Framework. Anxiety and depression were measured using the Self-Rating Anxiety Scale (SAS) and Self-Rating Depression Scale (SDS). A total of 24 patients were assigned to a peer support group ( n  = 12) and a control group ( n  = 12). Compared with the control group, the peer support group showed higher FMT-related knowledge, stronger choice predisposition toward FMT, higher decision-making self-efficacy, and lower decisional conflict after the intervention. No significant between-group differences were observed in preparation for decision making, decisional satisfaction, anxiety, or depression. In this small exploratory pilot study, nurse-led peer support showed preliminary potential to improve selected decision-related outcomes among women with rUTIs considering FMT. Rather than confirming efficacy, this study supports the feasibility of a structured nurse-led peer support protocol and provides a basis for future larger-scale research. Given the small sample size, convenience sampling, and single cultural context, these findings should be interpreted cautiously and require confirmation in larger studies. Future studies should also assess whether these preliminary improvements are sustained and translate into actual treatment decisions and long-term decisional satisfaction.

An XR-based multisensory feedback system for real-time sprint technique optimization in track athletes

Scientific Reports Zifu Xu, Ziyu Wang, Gang Qin Jun 30, 2026 DOI: 10.1038/s41598-026-60240-3

Reply to Borst et al.: Acupuncture and the placebo effect

Proceedings of the National Academy of Sciences Lynne Peeples, Gene Russo Jun 30, 2026 DOI: 10.1073/pnas.2618232123

Prediction of the effect of biochar on soil CEC improvement based on machine learning

Scientific Reports Hang Yang, Chenxi Zhao, Hongyu He et al. Jun 30, 2026 DOI: 10.1038/s41598-026-58902-3

Abstract Biochar is an environmentally friendly soil amendment and is widely used for improving soil properties. Especially the Cation Exchange Capacity (CEC) of soil, which is the main criterion for assessing soil nutrients. Therefore, this study proposes a method for predicting the cation exchange capacity of soil, which is of great significance for the precise application of biochar and improving soil amendment efficiency. This study collects and organizes experimental data from published literature on biochar-amended soils to construct a dataset that includes biochar properties (feedstock type, pyrolysis temperature, specific surface area, cation exchange capacity) and soil properties. The dataset is divided into seven groups based on the properties of biochar to investigate the impact of biochar properties on the model’s prediction results. Using four machine learning algorithms—Light Gradient Boosting Machine (LightGBM), Deep Neural Network (DNN), Categorical gradient Boosting (CatBoost), and Random Forest (RF)—a predictive model for soil CEC after biochar was established. The results show that the CatBoost model performed best, with a coefficient of determination (R 2 ) of 0.963, a Mean Absolute Error (MAE) of 1.346, and a Root Mean Square Error (RMSE) of 2.238, indicating that it is effective in predicting soil CEC after the addition of biochar. Shapley Additive Explanations (SHAP) analysis and Partial Dependence Plot (PDP) results indicate that the pyrolysis temperature of biochar promotes the predicted values of soil CEC, while biochar with high CEC reduces the predicted values of soil CEC. The reason for this counterintuitive result may be that biochar with a high CEC competes for cations in the soil solution. Choosing biochar produced at high pyrolysis temperatures and with a specific surface area (SSA) below 50 m 2 /g can achieve a good improvement effect within the studied conditions. This study develops a promising model for predicting soil CEC, which can better optimize actual soil improvement, and provides valuable insights into the mechanism of the impact of biochar on soil CEC.

Adverse childhood experiences and problematic gambling among young adults: a moderated serial mediation analysis

Scientific Reports İbrahim Erdoğan Yayla, Samet Makas, Kübra Dombak et al. Jun 30, 2026 DOI: 10.1038/s41598-026-59889-7

Functional classification systems and their association with botulinum toxin type a use in pediatric cerebral palsy

Scientific Reports Duygu Kurtuluş, Gokhan Basar, Ozlem Kaleoglu Jun 30, 2026 DOI: 10.1038/s41598-026-57266-y

Comparison of personality domains derived from personality ratings and behavioral observations in Japanese macaques (Macaca fuscata)

Scientific Reports Kosho Katayama, Masayuki Nakamichi, Kazunori Yamada Jun 30, 2026 DOI: 10.1038/s41598-026-52385-y

Abstract Primate personalities are a phenomenon of widespread interest. Researchers have attempted to measure primate personalities using two main methods: personality ratings and behavioral observations. However, a lack of consensus exists on whether the same personality domains can be measured using different methods. In our study, we aimed to explore the personality structure of 32 free-ranging Japanese macaques ( Macaca fuscata ) using personality ratings and behavioral observations. A principal component analysis of personality ratings assessed via questionnaires (the Hominoid Personality Questionnaire) revealed three components: Dominance PR , Anxiety PR , and Friendliness PR . Behavioral observations across a wide range of measures in daily situations also revealed two Behavioral Components we term Sociability BC and Grooming BC . The behavioral domains captured individual tendencies in sociability, particularly with respect to the extent of interaction with kin. Comparison of the domains identified by the two methods showed that no overlapping domains were found. Furthermore, subject attributes such as rank and age were associated with Dominance PR and Friendliness PR . This may be attributable to the lack of contextual information in the observational data. The concept of “personality coherence,” whereby behavior changes according to situational demands, may be important for understanding the association between personality ratings and behavioral observations.

Pesticide Contamination in Indian Agricultural and Residential Areas: A Comparative Assessment of Health Risks and Environmental Impacts

Indian Journal of Forensic Medicine and Pathology Riya Raj C.A., Nandini Katare, Don Caeiro et al. Jun 30, 2026 DOI: 10.21088/ijfmp.0974.3383.19226.14

The extensive use of pesticides in India, particularly in agriculturally intensive regions, has raised significant concerns regarding environmental contamination and associated human health risks. This review synthesizes recent evidence on pesticide residue occurrence in water, milk, vegetables, and soil, with special emphasis on high-risk agricultural districts of Chhattisgarh and comparable regions. Data were compiled from government databases, peer-reviewed literature, and empirical field-based analytical studies employing standardized protocols (APHA, AOAC) and advanced instrumental techniques such as GC–MS, GC–MS/MS, HPLC-PDA, AAS, ICP–MS, and HPTLC. Findings indicate widespread detection of organophosphates, organochlorines, and pyrethroids—including carbaryl, chlorpyrifos, malathion, diazinon, deltamethrin, quinalphos, and 4,4′- DDT—in environmental and food matrices. Several studies reported exceedance of acceptable daily intake (ADI) and acute reference dose (ARfD) values, particularly among children, highlighting elevated vulnerability. Health risk assessments utilizing Hazard Quotient (HQ), Hazard Index (HI), Target Hazard Quotient (THQ), and Lifetime Cancer Risk (LCR) models revealed potential noncarcinogenic and carcinogenic risks in areas proximal to pesticide application sites.These compounds exert toxicity primarily through mechanisms such as acetylcholinesterase inhibition, endocrine disruption, oxidative stress induction, and bioaccumulation, contributing to both acute and chronic health effects. Spatial analyses further demonstrated contamination hotspots near agricultural spraying zones, while seasonal variation influenced physicochemical water quality and pollutant persistence. Despite regulatory efforts, critical research gaps remain, including limited longitudinal monitoring, inadequate evaluation of cumulative and mixture toxicity, insufficient exploration of non-agricultural contamination sources, and weak implementation strategies for sustainable alternatives.The review underscores the urgent need for integrated surveillance systems, farmer education on safe pesticide handling, improved regulatory enforcement, and adoption of sustainable pest management strategies. Strengthening evidence-based policymaking is essential to safeguard environmental integrity, public health, and long-term agricultural sustainability.

Three-body abrasive wear behavior of nano-hydroxyapatite reinforced carbon–epoxy hybrid composites with experimental and ANN-based analysis

Scientific Reports G. S. Divya, B. E. Naveena, K. Santhosh et al. Jun 30, 2026 DOI: 10.1038/s41598-026-59671-9

Abstract This study investigates the three-body abrasive wear behavior of nano-Hydroxyapatite (nHAP) reinforced Carbon-Epoxy (CE) composites, using an integrated experimental and machine learning approach. Taguchi Design of Experiments (DoE), regression modeling, and machine learning prediction were used to analyze and predict wear behavior. Laminates containing 0.5, 1.5, and 3 wt% nHAP were fabricated using the vacuum bagging technique and tested using dry sand–rubber wheel standards. The morphology size elemental composition crystalline phase structure of nHAP was characterized by scanning electron microscopy (SEM), Energy-Dispersive X-ray Analysis (EDAX), X-ray diffraction (XRD). The measured density values increased progressively from 1.485 g/cm³ for 0.5 wt% nHAP to 1.515 g/cm³ for 3 wt% nHAP, which was in close agreement with the theoretical density range of 1.505–1.545 g/cm³ confirming uniform filler dispersion. Barcol hardness increased with an increase in nHAP content from 74.4 at 0.5 wt% to 75.6 at 1.5 wt% and 78 0.8 at 3 wt% nHAP filled composites. Taguchi L 27 orthogonal array analysis found that filler content is the most significant parameter with a contribution of 52.34%, followed by abrading distance (36.32%), applied load (7.03%), and abrasive size (3.04%). The lowest experimental Specific wear rate (Ks) was observed in 3 wt% with 0.714 10 − 11 m 3 /Nm. Wear prediction equations were developed using regression modeling with R² values greater than 98%. An Artificial Neural Network (ANN) having two hidden layers, achieved R² values of 0 0.9987 (training) and 0 0.9988 (testing). Strong model accuracy was established through residual histograms, scatter plots, and parity plots confirming nHAP reinforced CE composites as promising materials for abrasion resistance. Confirmation test was carried out using Signal to noise ratio, and the result was validated with worn surface morphology.

Neurocysticercosis as an Unrecognized Cause of Sudden Death: A CaseBased Study

Indian Journal of Forensic Medicine and Pathology Mohamed Izran, Ganesh Rajahan, Sneha S. et al. Jun 30, 2026 DOI: 10.21088/ijfmp.0974.3383.19226.17

Neurocysticercosis (NCC) is a leading global cause of adult-onset epilepsy and the most common parasitic infection of the central nervous system. It is caused by the larval stage of the pork tapeworm, Taenia solium, which encysts in the brain after the ingestion of eggs. While often presenting with chronic symptoms, NCC can lead to sudden, unexpected death through acute neurological or cardiac complications. Forensic awareness and a high index of suspicion are critical for identifying subclinical infections in at-risk populations. This case is based on a 43-year-old male who had multiple episodes of seizure for which he was brought to a tertiary care hospital, where he was declared brought dead. This case highlights the significance of conducting a comprehensive medicolegal death investigation, which involves interviews with relatives and a complete forensic autopsy to establish the cause and manner of death.

Optimized steganographic embedding guided by snake algorithm and fusion-aware attention maps

Scientific Reports Ahmed Aljughaiman, Rana Alrawashdeh Jun 30, 2026 DOI: 10.1038/s41598-026-59840-w

Forensic Age Estimation using CBCT-Derived Mandibular Morphometrics: A Comparative Study of Regression and Machine-Learning Models

Indian Journal of Forensic Medicine and Pathology Asheerbad Swain, Kamala K.A., Suraiya Khan et al. Jun 30, 2026 DOI: 10.21088/ijfmp.0974.3383.19226.3

Background: Accurate age estimation in adolescents and adults remains challenging in forensic practice once dental development is complete. Cone-beam computed tomography (CBCT) enables three-dimensional evaluation of skeletal structures and may improve age estimation without additional radiation exposure. Aim: To develop and internally validate a CBCT-based multivariate regression model for chronological age estimation using mandibular morphometrics and to compare its performance with a machine-learning approach. Materials and Methods: This retrospective study analyzed 150 CBCT scans of individuals aged 10–70 years. Mandibles were segmented using ITK-SNAP software, and standardized three-dimensional morphometric measurements were obtained. The dataset was randomly divided into a training set (n = 105) and a testing set (n = 45). Pearson’s correlation analysis and stepwise multivariate linear regression were used to develop the regression model. A Random Forest regression model was trained for comparison. Model performance was assessed using the coefficient of determination (R²), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Agreement between predicted and chronological age was evaluated using Bland–Altman analysis and intraclass correlation coefficient (ICC). Results: Chronological age showed a strong positive correlation with the gonial angle and a strong negative correlation with the ramus height–to–body length ratio. The final regression model retained gonial angle, bicondylar width, and ramus height–to–body length ratio as significant predictors. In the testing dataset, the regression model demonstrated excellent predictive accuracy (R² = 0.881; RMSE = 6.09 years; MAE = 5.89 years), minimal bias (−0.49 years), and excellent agreement (ICC = 0.90). The Random Forest model showed reasonable performance but did not outperform the regression model. Conclusion: CBCT-derived mandibular morphometrics enable accurate, noninvasive forensic age estimation. The regression model demonstrated superior reliability and interpretability compared with machine-learning, supporting its clinical and medico-legal applicability. Key findings 1. A regression-based model using CBCT-derived mandibular morphometrics achieved high predictive accuracy and agreement for forensic age estimation, outperforming the machine-learning approach in internal validation. 2. Mandibular shape and angular parameters, particularly gonial angle and ramus height–to–body length ratio, were more reliable indicators of chronological age than isolated linear mandibular measurements.

Bioactive metabolites from endophytic Fusarium equiseti isolated from Hyoscyamus muticus mitigate cadmium toxicity in biological models

Scientific Reports Hagar E. Mohammed, Mervat G. Hassan, Ahmed A. Hamed et al. Jun 30, 2026 DOI: 10.1038/s41598-026-59146-x

Abstract Cadmium is a hazardous metal that induces oxidative stress, hepato-renal dysfunction, and metabolic disturbances. This study investigated the protective potential of bioactive secondary metabolites extracted from endophytic Fusarium equiseti (FE) against cadmium-induced toxicity. Gas chromatography- Mass Spectrometry (GC–MS) profiling revealed 17 compounds dominated by esters (67%) and fatty acids (26%). The major constituents were 9,12-Octadecadienoic acid (Z,Z) (10.21%), and Oleic acid (8.13%). In vitro -FE exhibited dose dependent antioxidant activity (IC50 = 145.91µ g/ml in DPPH assay) and broad-spectrum antimicrobial efficacy, producing inhibition zones up to 46 mm against E. coli . Anti- inflammatory evaluation showed hemolysis inhibition rising from 3.1% (100 µg/mL) to 55.9% (1000 µg/ml). Furthermore, the MTT assay revealed that the FE exhibited moderate cytotoxicity against HeLa and PC3 (IC50 values of 194.15 µg/ml and 162.88 µg/ml, respectively) cancer cell lines, while showing a lower toxicity toward normal WI38 cells (IC50 of 237.26 µg/ml; Selectivity Index up to 1.46). These metabolites are suggested to be the key contributors to the extract’s antioxidant and antimicrobial activities. In vivo , twenty Wistar rats were randomly assigned to four sets (n = 5 per set): (1) Control; (2) CdCl 2 , receiving a single IP dose of 3 mg/kg; (3) FE (100 mg/ kg; OP; for 14 constitutive days; and (4) FE + CdCl 2. CdCl 2 significantly elevated WBCs, lymphocyte, MDA, ALT, AST, ALP, creatinine, uric acid, total lipids, TG, CH, and LDL- C while reducing bilirubin, albumin, HDL-C, SOD, and GSH. Pretreatment with FE markedly restored antioxidant markers (SOD and GSH), reduced MDA, normalized lipid and liver profiles, and improved hepato-renal histopathology. These findings show that FE metabolites exert antioxidant, antimicrobial, anti-inflammatory, and cytoprotective effects, mitigating cadmium- induced oxidative damage. FE represents a promising natural source for developing hepato-renal protective agents against heavy metal toxicity.

A Questionnaire-Based Cross Sectional Study on the Prevalence of Substance Abuse among Age 17 to 25 Years in the Chengalpattu Population with Medico Legal Perspectives

Indian Journal of Forensic Medicine and Pathology Unnamalai R.M., Dipayan Deb Barman, Vivek Valliappan Jun 30, 2026 DOI: 10.21088/ijfmp.0974.3383.19226.11

Background: Substance abuse among young individuals represents a growing public health concern with significant social, psychological, and medico-legal implications. The age group of 17–25 years constitutes a vulnerable transitional phase during which initiation of substance use may result in dependence, health deterioration, academic impairment, and legal consequences.¹ Aim: To determine the prevalence and patterns of substance abuse among individuals aged 17–25 years in the Chengalpattu population and to assess associated medico-legal perspectives. Methodology: A questionnaire-based cross-sectional study was conducted among participants aged 17–25 years residing in Chengalpattu district. Data on socio-demographic characteristics, substance use patterns, reasons for initiation, frequency of use, features suggestive of dependence, and associated behavioral and medico-legal issues were collected using a structured questionnaire. Descriptive and inferential statistical analyses were performed. Results: The overall prevalence of substance use was 9.8%. Alcohol was the most commonly used substance, followed by tobacco and cannabis. A proportion of substance users exhibited features suggestive of dependence. Substance use showed associations with family-related issues, health problems, sleep deprivation, and medico-legal concerns, including exposure to violence. Conclusion: The study demonstrates a notable prevalence of substance abuse among youth in the Chengalpattu population, influenced by socio-demographic and psychosocial factors. The observed medico-legal implications highlight the need for targeted preventive strategies, youth-focused counseling, and coordinated public health and legal interventions to reduce substance-related harm.

Next-generation antenna beamforming via caterpillar fungus optimization for enhanced wireless communication

Scientific Reports Samar I. Farghaly, Mona Gafar, Abdullah M. Shaheen et al. Jun 30, 2026 DOI: 10.1038/s41598-026-55840-y

Abstract Beamforming has emerged as an essential enabling technique for beyond 5G and future 6G systems because it improves spectral efficiency. However, optimizing antenna weights in beamforming is a highly nonlinear and multidimensional problem that traditional approaches struggle to solve. To address this, we offer a unique beamforming strategy based on the Caterpillar Fungus Optimization (CFO) algorithm, which strikes an optimal balance between exploration and exploitation, making it ideal for large-scale antenna systems. The CFO is inspired by the rare lifecycle of caterpillar fungus considering its soil exploration and parasitic behaviors. Its unique blend of wave-like and spiral search strategies, dual parasitism operators, and hybrid noise-handling makes it stand out among bio-inspired algorithms, enabling high accuracy and robustness in complex engineering optimization problems. The proposed scheme has two goals: first, to reduce the number of active antenna elements, thereby improving energy efficiency and reducing system complexity; and second, to suppress side lobe levels (SLL), which mitigate interference and improve communication performance. To assess its efficacy, the CFO-based method is compared to five established algorithms: Artificial Rabbits Optimizer (ARO), Whale Shark Optimization (WSO), Grey Wolf Optimizer (GWO), Particle Swarm Optimization (PSO), and Boomerang Aerodynamic Ellipse (BAE). According to simulation data, CFO maintains beamwidth deviations within 1% to 2% of the standard reference while achieving an average error reduction of up to 99.7% when compared to PSO and WSO. Furthermore, CFO outperforms all benchmark algorithms in terms of accuracy, and computing efficiency, delivering the lowest SLL deviations and the most steady convergence behavior. This paper provides a simulation-based beamforming optimization framework that employs the metaheuristic Caterpillar Fungus Optimization (CFO) algorithm for antenna array synthesis in beyond 5G and future 6G wireless systems.

Assessment of Knowledge on Sexual Assault Forensic Examination among Nurses in India: A Cross-Sectional Study

Indian Journal of Forensic Medicine and Pathology Renjulal Yesodharan, Tessy Treesa Jose, Vinod C. Nayak Jun 30, 2026 DOI: 10.21088/ijfmp.0974.3383.19226.13

Background: Sexual violence is a major global public health and human rights issue that affects women and children in various societies. Survivors of sexual assault need immediate medical care and forensic examination, where healthcare professionals play a very crucial role. Nurses are the first point of contact in most of the health care settings to give immediate and prompt care. Nurses help to identify injuries, collect, preserve, and pack evidence, document findings and maintain chain of custody. However, limited training in forensic aspects may impact their ability to perform this task effectively. Objectives: This study aimed to evaluate the knowledge level regarding sexual assault examination among registered nurses and final-year nursing students Methods: A quantitative cross-sectional study was conducted with 276 participants from hospitals and nursing colleges in Dakshina Kannada and Udupi districts. Data were collected using socio-demographic proforma and a 38 item knowledge questionnaire on sexual assault nurse examination (KQSANE). Descriptive statistics summarize knowledge levels. Spearman’s correlation and generalized linear model regression were used to explore the relationships between knowledge domains and socio-demographic variables. Results: The findings indicate that overall knowledge about sexual assault examination was inadequate, with most participants demonstrating poor knowledge. Among the knowledge areas, legal and ethical knowledge domain received the highest scores. The domain forensic examination and imaging received the lowest scores. Significant positive correlation was found between the overall knowledge scores and domains of knowledge. Conclusion: The study points out significant gaps in knowledge about forensic sexual assault examination among nurses and nursing students. Improving forensic nursing education through better curriculum integration and in-service education and training is important to prepare nurses for enhancing the survivorcentered care and ensuring proper evidence management.

The bio-adsorptive treatment of detergent: performance and mechanisms

Scientific Reports Aya H. Saqr, Abd El-Latif Hesham, Mohamed I. Attia et al. Jun 30, 2026 DOI: 10.1038/s41598-026-59039-z

Abstract This study develops an optimized hybrid system to treat sodium dodecyl sulfate (SDS)-contaminated wastewater that synergistically integrates microbial biodegradation with physical adsorption within a Response Surface Methodology (RSM) framework. A highly effective SDS-degrading bacterium, Serratia plymuthica strain BSU-AH-03, was isolated from hydrocarbon-contaminated soil. Using a Box-Behnken Design (BBD), the optimal biodegradation conditions (pH 7.9, 20 °C, 300 mg L⁻¹ SDS) yielded a degradation efficiency of 90.46% (predicted 91.27%, R² = 0.981). Subsequently, natural anthracite coal was employed as an adsorbent to polish the effluent. The anthracite exhibited a high surface area (890.9 m² g⁻¹) and a heterogeneous micro-mesoporous structure. Batch adsorption experiments achieved a maximum SDS uptake capacity of 158.7 mg g⁻¹ and a near-complete removal efficiency of 99.58% at 318 K and pH 7. The adsorption process was endothermic (ΔH° = 58.4 kJ mol⁻¹), spontaneous (ΔG° from – 5.15 to -10.65 kJ mol⁻¹), and followed pseudo-second-order kinetics and the Langmuir isotherm, indicating chemisorption as the dominant mechanism. Mechanistic analysis revealed that SDS adsorption involves intra-particle pore diffusion, hydrophobic interactions, electrostatic forces, and hydrogen bonding, culminating in interfacial hemi-micellar aggregation. The synergistic combination of tailored biodegradation and advanced adsorption provides a highly efficient, statistically optimized strategy for the complete remediation of surfactant-laden industrial effluents.