Browse Articles

Discover research articles across all indexed journals

Protocol to establish standards for the elements infection prevention and control programs and competencies for infection control professionals in Australian residential aged care homes

PLoS ONE Ramon Z. Shaban, Kate Curtis, Margaret Fry et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0319108

The COVID-19 pandemic elucidated the importance of infection prevention and control (IPC) in residential aged care homes (RACHs), both on the health and wellbeing of its residents, and the staff and clinicians who care for them. While considerable efforts have been made in Australia to improve IPC both during and in the aftermath of the COVID-19 pandemic, many of these resources remain reliant on evidence originating from hospitals and acute healthcare settings. This research aims to establish the core minimum components that will populate standards for IPC programs and governance (Stream A) and for professional practice and competencies (Stream B) in RACHs. This research will be completed using a sequential three-phase design. In Phase 1, two integrative literature reviews will be completed to synthesise the elements of current global IPC programs and professional competencies in RACHs. In Phase 2, a qualitative analysis of IPC programs and professional practice in Australian RACHs using a combination of surveys and interviews will be completed. Finally, in Phase 3, an e-Delphi will be conducted to collate expert opinion and generate consensus on the minimum components of the IPC program and professionals who administer them in RACHs. The final standards and resources will be collaboratively designed with aged care partners, industry leaders, professional bodies and key Australian health policymakers. These standards seek to empower IPC and aged care staff, not only by ensuring that they are well-equipped with the knowledge and skills to implement effective IPC programs themselves, but also that the organisation is adequately prepared to provide the resources and governance systems.

Algorithm for pixel-level concrete pavement crack segmentation based on an improved U-Net model

Scientific Reports Zixuan Zhang, Yike He, Di Hu et al. Feb 24, 2025 DOI: 10.1038/s41598-025-91352-x

Abstract Cracks that occur in concrete surfaces are numerous and diverse, and different cracks will affect road safety in different degrees. Accurately identifying pavement cracks is crucial for assessing road conditions and formulating maintenance strategies. This study improves the original U-shaped convolutional network (U-Net) model through the introduction of two innovations, thereby modifying its structure, reducing the number of parameters, enhancing its ability to distinguish between background and cracks, and improving its speed and accuracy in crack detection tasks. Additionally, datasets with different exposure levels and noise conditions are used to train the network, broadening its predictive ability. A custom dataset of 960 road crack images was added to the public dataset to train and evaluate the model. The test results demonstrate that the proposed U-Net-FML model achieves high accuracy and detection speed in complex environments, with MIoU, F1 score, precision, and recall values of 76.4%, 74.2%, 84.2%, and 66.4%, respectively, significantly surpassing those of the other models. Among the seven comparison models, U-Net-FML has the strongest overall performance, highlighting its engineering value for precise detection and efficient analysis of cracks.

Longitudinal healthcare use after pediatric brain injury: A population-based birth cohort study

PLoS ONE Vincy Chan, Clarissa Serafine Wirianto, Robert Balogh et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0316165

Background Traumatic brain injury is a chronic disease with lifelong consequences. In children, it can affect developmental milestones. Longitudinal data on brain injury and long-term healthcare use is limited, with lack of clarity on social determinants of health and its effects on healthcare use. This study explores rates of healthcare use, from birth, and up to 10 years after a childhood traumatic brain injury-related healthcare visit. Methods and findings This study uses a population-based birth cohort of individuals born between April 1, 2002 and March 31, 2020 from Ontario, Canada. A case cohort (TBI cohort) was created using a sample of individuals who had at least one traumatic brain injury-related healthcare visit between the ages of 0 and 4 years, inclusive (n = 26,988). Controls were generated from a sample of individuals who did not have any traumatic brain injury-related healthcare visit during the study period (n = 193,253 for emergency department visits and hospitalizations, and n = 19,313 for primary care physician visits). The primary outcome is rates of primary care physician visits, emergency department visits, and hospitalizations for each year prior to and up to 10 years after the index traumatic brain injury-related healthcare visit, calculated using standard life table methods. Rates and 95% confidence intervals were further calculated and stratified by rurality of residence, and the following Ontario Marginalization Index metrics: neighbourhood income quintile and neighbourhood racialized and newcomer populations. Rates of healthcare use remained consistently higher in the TBI cohort compared to controls both prior to and after the index TBI-related healthcare visit. Rates also varied across social determinants of health. Overall, rates were higher in males compared to females across all healthcare settings. Rates of primary care physician visits were higher among those living in urban (vs. rural) settings. However, rates of emergency department visits were higher among those living in rural (vs. urban) settings. Rates of emergency department visits and hospitalizations were higher among those living in the lowest (vs. highest) income quintile neighbourhoods. Rates of primary care physician visits were higher among those living in areas with the most (vs. least) racialized and newcomer populations. However, rates of emergency department and hospitalizations were higher among those living in areas with the least (vs. most) racialized and newcomer populations. This study is limited to change in rates of healthcare use over time and does not quantify the magnitude of these changes. Conclusions Research on longitudinal healthcare use is needed to explore the causes of sustained and increased healthcare use post-injury, to inform opportunities for targeted health and social care interventions. Findings also suggest that a lifespan perspective is critical to understand how early life events can impact post-injury outcome.

Relative improvement in language vs. motor functions with reperfusion therapies for large vessel occlusion

Scientific Reports Zaka Ahmed, Vivek Yedavalli, Wladimir Sarmiento Gonzalez et al. Feb 24, 2025 DOI: 10.1038/s41598-025-90871-x

Abstract When weighing potential risks versus benefits of reperfusion therapy, the functions likely to recover if blood flow can be restored should be considered. Because deep and motor areas often infarct relatively early in acute stroke, we hypothesized that reperfusion therapies improve predominantly cortical functions more than motor function. In this retrospective review of a prospectively collected database of patients with acute stroke due to large vessel occlusion, we evaluated percent improvement (mean change in score/maximum score) for different items of the National Institutes of Health Score Scale with and without endovascular thrombectomy (EVT), and/or intravenous thrombolysis. In total, 290 patients met the inclusion criteria. For all outcome measures there were significant effects of treatment group (p < 0.0001 for all), with the greatest change in the EVT + thrombolysis group, then EVT only group, followed by thrombolysis only, followed by no intervention. Differences between EVT + thrombolysis and EVT only were not significant (p = 0.30 to 0.79 across outcomes). For patients with aphasia and/or right sided weakness before treatment, the percent change in language was significantly greater than the percent change in weakness (29.8% vs. 12.7%; t(93) = 5.3;p < 0.0001). Greater percent improvement in language was observed in all treatment groups (p = 0.0003 to 0.03 across treatment groups). After acute ischemic stroke due to LVO, improvements occur in all neurological functions with intervention. However, gains in language are even greater than gains in motor function in the same patients. Few patients had neglect before treatment, but of those who did, the majority improved, and most (92.8%) with EVT.

Research on three-dimensional path planning of unmanned aerial vehicle based on improved Whale Optimization Algorithm

PLoS ONE HaoCheng Wang, ZeXian Hao, Yu Zhang Feb 24, 2025 DOI: 10.1371/journal.pone.0316836

Addressing the insufficient optimization performance in drone 3D path planning and the issues of inadequate optimization precision and tendency to fall into local optima in the existing Whale Optimization Algorithm (WOA), this paper proposes a drone 3D path planning method based on an improved Whale Optimization Algorithm (CSRD-WOA). Firstly, to enhance the search efficiency and fitness accuracy of the Whale Algorithm, the Cuckoo Search and Random Differential Strategy were introduced and compared with the traditional Particle Swarm Optimization algorithm, Whale Algorithm, and Cuckoo Search Algorithm. Experimental results demonstrate that the CSRD-WOA algorithm improves global search capabilities and prevents premature convergence, significantly enhancing optimization precision and convergence speed. Secondly, applying the CSRD-WOA algorithm to drone 3D path planning issues, the simulation results show that the CSRD-WOA algorithm can effectively manage path planning in complex terrains, showcasing its application potential in drone path planning.

Microfluidic fabrication of rattle shaped biopolymer microcapsules via sequential phase separation in oil droplets

Scientific Reports Takaichi Watanabe, Yuko Sakai, Kurumi Mori et al. Feb 24, 2025 DOI: 10.1038/s41598-025-91550-7

Influencing consumer perceptions in green tourism: criteria and strategies for effective destination branding

PLoS ONE Tianchang Chen, Gao Liu, Xin Sui et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0319254

As the global shift towards sustainable tourism intensifies, the need for effective destination branding strategies becomes increasingly important. This study investigates and prioritizes these strategies to enhance consumer perceptions of sustainable tourism destinations, with a focus on Guilin. The research examines five key criteria, twenty sub-criteria, and nine strategic approaches essential for destination branding. The primary objective is to shape consumer perceptions and promote green tourism marketing in the region. Through the application of Fuzzy Analytical Hierarchy Process (AHP) and Fuzzy Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), key criteria and strategies are identified and evaluated. Results from the fuzzy AHP highlight communication and brand transparency (GM3) as the most critical criterion, stressing the necessity of transparent communication channels. Environmental sustainability (GM1) ranks as the second most important criterion, emphasizing the importance of eco-friendly practices and the preservation of Guilin’s natural environment. Stakeholder collaboration and partnerships (GM4) is identified as the third key criterion. The fuzzy TOPSIS analysis reveals the most effective strategies, with strategic collaborations and partnerships (STM9), eco-friendly infrastructure development (STM1), and energy efficiency measures (STM4) emerging as the top approaches. These strategies have significant potential to influence consumer perceptions positively towards green tourism marketing.

The impact of an amino acid-humus preparation on lawn boning value

Scientific Reports Adam Radkowski, Iwona Radkowska, Karen Khachatryan et al. Feb 24, 2025 DOI: 10.1038/s41598-025-90862-y

Abstract The application of biostimulants in fertilisation has been shown to enhance plant metabolism, stimulate physiological processes, and mitigate adverse environmental and pathogenic stresses. This study, conducted at the Agricultural University of Krakow (2021–2023), evaluated the efficacy of a foliar amino acid preparation enriched with humic acids (L-Amino + ® Humus) on turf quality. Four experimental variants were established: an untreated control (Variant I) and three treated plots sprayed with the biostimulant at 1.0 (Variant II), 2.0 (Variant III), and 3.0 L ha−1 (Variant IV). Higher application rates (Variants III and IV) significantly improved aesthetic turf quality and reduced fungal disease incidence. Compared to the control, Variant IV (3.0 L ha−1) decreased snow mould (Microdochium nivale) infections by 8% and brown spot (Rhizoctonia solani) prevalence by 12%. Variant III (2.0 L ha−1) also demonstrated positive effects, with 6% and 8% reductions in snow mould and brown spot, respectively. Furthermore, the Normalized Difference Vegetation Index (NDVI) and leaf greenness index (SPAD) chlorophyll meter values increased by 6% and 8%, respectively, in Variant IV relative to the control, indicating enhanced leaf greenness and photosynthetic efficiency. Nutrient analysis revealed that Variants III and IV significantly elevated macro- and micronutrient concentrations (e.g., nitrogen, potassium, iron) in plant biomass compared to the control. In contrast, Variant II (1.0 L ha−1) showed no statistically significant improvements in nutrient content or disease resistance. These findings suggest that L-Amino + ® Humus at doses of 2.0–3.0 L ha−1 enhances turf quality, bolsters resistance to fungal pathogens, and improves nutrient assimilation. Lower doses (1.0 L ha−1), however, were ineffective under the tested conditions. The study underscores the dose-dependent efficacy of biostimulants in sustainable turf management.

Chemical nucleases are a robust alternative for RNase H cleavage of human ribosomal RNA

PLoS ONE Hagen Wesseling, Dennis Krug, Marvin Wehrheim et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0318697

Besides the abundant ribosomal and transfer RNA transcripts (rRNA and tRNA, respectively), there are tens of thousands of long coding (mRNA) and non-coding transcripts (lncRNA) within each cell whose modification profiles have not yet been elucidated. One reason for this is that most mRNAs and lncRNAs are low abundant and their purification prior to direct modification analysis by mass spectrometry (LC-MS) is highly challenging. State-of-the-art mRNA purification protocols are based on poly(A) enrichment with subsequent rRNA depletion using either magnetic pulldown assays or RNase H. While these methods are well suited for RNA sequencing, where residual rRNA content can be acceptable, LC-MS analysis of mRNA requires samples with less than 1% rRNA and higher yields, making the existing methods close to unaffordable. Thus, a new principle for low-cost (pre)mRNAs and lncRNAs preparation from total RNA will be beneficial for LC-MS analysis but also sequencing approaches. Here, we show that the use of ARRR (artificial ribosomal RNA remover: conjugates of standard DNA probes and a small-molecule chemical nuclease) is suitable for rRNA cleavage. In addition, ARRR has a higher target specificity compared to E. coli RNase H using regular DNA probes and only limited off-target RNA degradation is observed with ARRR. In summary, we present a promising tool with high potential to remove overly abundant rRNA, which might be used for enrichment of lncRNAs and (pre)mRNAs for downstream sequencing and MS-based analysis.

Analysis of trends in cancer mortality and the years of life lost in six provinces in northwest China from 2013 to 2021

Scientific Reports Haitao Bi, Ning Wang, Xiaolin Yu et al. Feb 24, 2025 DOI: 10.1038/s41598-025-88941-1

Comparative efficacy and safety of dupilumab versus newly approved biologics and JAKi in pediatric atopic dermatitis: A systematic review and network meta-analysis

PLoS ONE Qiwei Liao, Hanwen Pan, Yixin Guo et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0319400

Background The newly approved biologics and Janus kinase inhibitors (JAKi) for pediatric atopic dermatitis (AD) offer additional options for clinical treatment. However, the efficacy and safety differences compared to the first approved biologic, dupilumab, remain unclear. Therefore, a network meta-analysis was conducted to evaluate these differences and identify potentially superior agents. Methods This systematic review was PROSPERO-registered (CRD42024583658). Randomized controlled trials involving pediatric patients (<18 years old) published in PubMed, Embase, Web of Science, and the Cochrane Library up to October 27, 2024 were searched and screened. RevMan software was utilized for quality assessment, and  meta-analysis was performed using R version 4.4.1. Efficacy measures included the Investigator’s Global Assessment (IGA), the Numeric Rating Scale for Itch (NRS), and the Eczema Area and Severity Index (EASI). The results of these measures were expressed as odds ratios (OR), while treatment rankings of different interventions were determined using the P-score. Result This study included 11 trials involving 7 agents and 2,352 pediatric patients. The results indicated that dupilumab (300 mg) showed better outcomes than placebo in IGA-0/1 (OR =  4.68, 95% CI: 2.53–8.63), NRS-4 (OR =  6.75, 95% CI: 3.85–11.86), and all EASI outcomes. Tralokinumab may be the most effective option for alleviating pruritus (P-score for NRS-4, 0.8447). Upadacitinib (30 mg) performed best in IGA-0/1 (P-score, 0.9414), EASI-90 (P-score, 0.9926), and EASI-75 (P-score, 0.9707). Dupilumab (300 mg) had a higher risk of nasopharyngitis compared to placebo (OR = 2.15, 95%CI: 1.04–4.43). Compared to both placebo and dupilumab (300 mg), adverse event rates were higher with upadacitinib (15 mg and 30 mg), and upper respiratory tract infection risk was elevated with baricitinib (2 mg and 4 mg) and tralokinumab (300 mg). Conclusion The efficacy of dupilumab for pediatric AD remains substantial, while other agents including upadacitinib, delgocitinib, and tralokinumab also present certain advantages. Future clinical trials may necessitate further evaluation of safety concerns.

Glaucocalyxin A alleviates ulcerative colitis by inhibiting PI3K/AKT/mTOR signaling

Scientific Reports Tongtong Zhou, Yujing Ye, Weijie Chen et al. Feb 24, 2025 DOI: 10.1038/s41598-025-91358-5

Revisiting CPSF30-mediated alternative polyadenylation in Arabidopsis thaliana

PLoS ONE Guijie Hao, Lichun Zhou, Huazhen Liu et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0319180

Alternative polyadenylation (APA) is an important contributor to the regulation of gene expression in plants. One subunit of the complex that cleaves and polyadenylates mRNAs in the nucleus, CPSF30 (for the 30 kD subunit of the mammalian Cleavage and Polyadenylation Specificity Factor), has been implicated in a wide-ranging network of regulatory events. CPSF30 plays roles in root development, flowering time, and response to biotic and abiotic stresses. CPSF30 also is a conduit that links cellular signaling and RNA modification with alternative RNA processing events and transcriptional dynamics. While much is known about CPSF30 and its roles in plants, questions remain regarding the connections between CPSF30-mediated APA and the downstream events that lead to specific phenotypic outcomes. To address these, we conducted a detailed analysis of poly(A) site usage in the CPSF30 mutant. Our results corroborate earlier reports that link CPSF30 with a distinctive cis element (AAUAAA) that is present 10-30 nts upstream of some, but not all, plant pre-mRNAs. Interestingly, our results reveal a distinctive shift in poly(A) site in mutants deficient in CPSF30, resulting in cleavage and polyadenylation at the location of motifs similar to AAUAAA. Importantly, CPSF30-associated APA had at best a small impact on mRNA functionality. These results necessitate the formulation of new hypotheses for mechanisms by which CPSF30-mediated APA influences physiological processes.

Genetic diversity analysis of Phyllanthus acidus Skeels of north-east India: Insights from multivariate analysis

Scientific Reports T. K. Hazarika, Linthoingambi Ningombam, Panthor Debbarma et al. Feb 24, 2025 DOI: 10.1038/s41598-025-86719-z

Enrichment of rare CFTR variants in Finnish patients with congenital chloride diarrhea

PLoS ONE Satu Wedenoja, Jarmo Ritari, Jukka Partanen et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0318249

Objective The autosomal recessive disease congenital chloride diarrhea (CLD), caused by loss-of-function mutations in the solute carrier family 26 member 3 (SLC26A3) gene, shows association with inflammatory bowel disease (IBD). However, it is unclear whether IBD risk is associated with genetic or immune signatures. SLC26A3 interacts with several ion transporters linked to intestinal inflammation, such as cystic fibrosis transmembrane conductance regulator (CFTR) and solute carrier family 9 member 3 (SLC9A3) causing congenital sodium diarrhea. We hypothesized that other epithelial channels affecting intestinal salt balance might modulate CLD phenotype or IBD risk. Materials and methods We analyzed 495 gene variants within 33 ion transporters among 28 patients with CLD and 44,443 population controls. Results We found three intronic variants at or near the CFTR locus (rs17132543, rs2283054 and rs76622533) showing statistically significant (P < 1.42x10-5) associations with CLD. Conclusions These data demonstrate enrichment of rare variants at the CFTR locus in chromosomes harboring the Finnish founder mutation for CLD.

Mindfulness-based stress reduction as perceived by individuals with pathological mental fatigue after an acquired brain injury

Scientific Reports Gustaf Glavå, Birgitta Johansson Feb 24, 2025 DOI: 10.1038/s41598-025-90452-y

Abstract After acquired brain injury (ABI), some individuals suffer from long-term fatigue and emotional distress, which affects their work ability and daily life. Mindfulness-based stress reduction (MBSR) has shown promising results in quantitative studies as a complementary treatment for pathological mental fatigue (PMF) after ABI. This study aims to explore how people suffering from lasting PMF after ABI experience MBSR in relation to their PMF, with the intention of better meeting the rehabilitation needs of those affected. Seven individuals (mean age 42 years) who had suffered from long-lasting PMF after ABI took part in the study. None of them had resumed work due to their PMF, but all had recovered from neurological impairments. They were interviewed after completing an MBSR course. Thematic analysis of the participants’ perspectives suggested that the MBSR program provided them with coping techniques for living with PMF. They gained a deeper understanding of their condition becoming more self-compassionate, and the treatment provided them with a forum for meeting and sharing experiences with others with similar problems. The qualitative results strengthen the evidence for MBSR as a feasible psychoeducative complementary treatment for PMF after ABI.

Effectiveness of a school-based physical activity intervention on overweight and obesity among children and adolescents in Pakistan

PLoS ONE Moazzam Tanveer, Ejaz Asghar, Georgian Badicu et al. Feb 24, 2025 DOI: 10.1371/journal.pone.0317534

Background Childhood obesity poses a significant public health challenge, yet effective school-based physical activity (PA) interventions remain scarce, especially in Pakistan. There is a lack of data assessing the impact of such interventions on obesity and related health outcomes in Pakistani school children. Methods This study aimed to design and implement a school-based intervention targeting multiple levels of the socio-ecological model to increase physical activity and reduce the prevalence of overweight and obesity in Pakistani youth. Conducted from October 2022 to January 2023 in Lahore, Pakistan, the 12-week, non-randomized controlled trial involved 1,200 students from eight schools, with four schools (n = 570) in the intervention group and four (n = 630) in the control group. Primary outcomes included changes in body mass index (BMI), waist circumference, and the prevalence of overweight/obesity, measured through anthropometric assessments. Secondary outcomes involved alterations in moderate-to-vigorous physical activity duration. Descriptive statistics, Chi-square tests, general linear mixed models, and repeated measures ANOVA were used for analysis. Results The intervention showed significant improvements across various socio-ecological levels. Intrapersonal factors saw a reduction in fast food consumption from 11.9% to 7.9% (F(1,1198) = 90.39, p < 0.001; η² = 0.074) and an increase in physical activity frequency from 11.9% to 39.6% (F(1,1198) = 465.25, p < 0.001; η² = 0.028). Screen time decreased from 27.0% to 7.4% (F(1,1198) = 219.83, p = 0.015; η² = 0.15), and normal sleep duration increased from 44.6% to 71.8% (F(1,1198) = 242.73, p < 0.001; η² = 0.16). At the interpersonal level, parental involvement in encouraging sports and providing financial support for sports activities significantly increased. School-level factors also showed positive changes, including improved sports facilities and equipment access. Community-level factors revealed increased opportunities for physical activity and a more supportive community environment. The intervention group’s BMI change (−0.06 ± 0.07 kg·m²) significantly differed from the control group’s (0.19 ± 0.09 kg·m²). Conclusions This study demonstrates the effectiveness of a multi-level intervention in boosting physical activity and addressing obesity among Pakistani school-aged children, supporting the implementation of similar school-based interventions.

Mesenchymal stem cells modulate breast cancer progression through their secretome by downregulating ten-eleven translocation 1

Scientific Reports Romina Motamed, Keyvan Jabbari, Mahboubeh Sheikhbahaei et al. Feb 24, 2025 DOI: 10.1038/s41598-025-91314-3

Optimizing depression detection in clinical doctor-patient interviews using a multi-instance learning framework

Scientific Reports Xu Zhang, Chenlong Li, Weisi Chen et al. Feb 24, 2025 DOI: 10.1038/s41598-025-90117-w

Abstract In recent years, the number of people suffering from depression has gradually increased, and early detection is of great significance for the well-being of the public. However, the current methods for detecting depression are relatively limited, typically relying on the self-rating depression scale (SDS) and interviews. These methods are influenced by subjective or environmental factors. To improve the objectivity and efficiency of diagnosis, deep learning techniques have been applied to the field of automatic depression detection (ADD), providing a more accurate and objective approach. During interviews, transcribed interview data is one of the most commonly used modalities in ADD. However, previous studies have only utilized response texts or selected question–answer pairs, resulting in information redundancy and loss. This paper is the first to apply the multiple instance learning (MIL) framework to the field of textual interview data, aiming to overcome issues of inadequate text representation and ineffective information extraction in long texts. In the MIL framework, each instance undergoes an independent feature extraction process, ensuring that the local features of each instance are fully captured. This not only enhances the overall text representation capability but also alleviates the issue of sample imbalance in the dataset. Additionally, this paper improves upon previous aggregation strategies by introducing two hyper-parameters to accommodate the uncertainties in the field of text sentiment. An ensemble model of MT5 and RoBERTa (referred to as multi-MTRB) was constructed to extract features from each instance and output confidence scores indicating the presence of depressive information in the instances. Due to the unique design of the MIL framework, the proposed method is highly interpretable and is able to identify specific sentences that identify people from depressed patients, while introducing LIME techniques to provide more in-depth interpretation of negative instance sentences. This provides a promising approach for depression detection in the context of text interview data patterns. We evaluated the proposed method on DAIC-WOZ and E-DAIC datasets with excellent results. The F1 score is 0.88 on the DAIC-WOZ dataset and 0.86 on the E-DAIC dataset.

CTDNN-Spoof: compact tiny deep learning architecture for detection and multi-label classification of GPS spoofing attacks in small UAVs

Scientific Reports Ahmad Almadhor, Jamel Baili, Shtwai Alsubai et al. Feb 24, 2025 DOI: 10.1038/s41598-025-90809-3