Displaying publications 281 - 300 of 934 in total

Abstract:
Sort:
  1. Using artificial intelligence methods for systematic review in health sciences: A systematic review
    Blaizot A, Veettil SK, Saidoung P, Moreno-Garcia CF, Wiratunga N, Aceves-Martins M, et al.
    Res Synth Methods, 2022 May;13(3):353-362.
    PMID: 35174972 DOI: 10.1002/jrsm.1553
    The exponential increase in published articles makes a thorough and expedient review of literature increasingly challenging. This review delineated automated tools and platforms that employ artificial intelligence (AI) approaches and evaluated the reported benefits and challenges in using such methods. A search was conducted in 4 databases (Medline, Embase, CDSR, and Epistemonikos) up to April 2021 for systematic reviews and other related reviews implementing AI methods. To be included, the review must use any form of AI method, including machine learning, deep learning, neural network, or any other applications used to enable the full or semi-autonomous performance of one or more stages in the development of evidence synthesis. Twelve reviews were included, using nine different tools to implement 15 different AI methods. Eleven methods were used in the screening stages of the review (73%). The rest were divided: two in data extraction (13%) and two in risk of bias assessment (13%). The ambiguous benefits of the data extractions, combined with the reported advantages from 10 reviews, indicating that AI platforms have taken hold with varying success in evidence synthesis. However, the results are qualified by the reliance on the self-reporting of the review authors. Extensive human validation still appears required at this stage in implementing AI methods, though further evaluation is required to define the overall contribution of such platforms in enhancing efficiency and quality in evidence synthesis.
    Matched MeSH terms: Machine Learning
  2. Fulltext A biomarker discovery of acute myocardial infarction using feature selection and machine learning
    Mohd Faizal AS, Hon WY, Thevarajah TM, Khor SM, Chang SW
    Med Biol Eng Comput, 2023 Oct;61(10):2527-2541.
    PMID: 37199891 DOI: 10.1007/s11517-023-02841-y
    Acute myocardial infarction (AMI) or heart attack is a significant global health threat and one of the leading causes of death. The evolution of machine learning has greatly revamped the risk stratification and death prediction of AMI. In this study, an integrated feature selection and machine learning approach was used to identify potential biomarkers for early detection and treatment of AMI. First, feature selection was conducted and evaluated before all classification tasks with machine learning. Full classification models (using all 62 features) and reduced classification models (using various feature selection methods ranging from 5 to 30 features) were built and evaluated using six machine learning classification algorithms. The results showed that the reduced models performed generally better (mean AUPRC via random forest (RF) algorithm for recursive feature elimination (RFE) method ranges from 0.8048 to 0.8260, while for random forest importance (RFI) method, it ranges from 0.8301 to 0.8505) than the full models (mean AUPRC via RF: 0.8044). The most notable finding of this study was the identification of a five-feature model that included cardiac troponin I, HDL cholesterol, HbA1c, anion gap, and albumin, which had achieved comparable results (mean AUPRC via RF: 0.8462) as to the models that containing more features. These five features were proven by the previous studies as significant risk factors for AMI or cardiovascular disease and could be used as potential biomarkers to predict the prognosis of AMI patients. From the medical point of view, fewer features for diagnosis or prognosis could reduce the cost and time of a patient as lesser clinical and pathological tests are needed.
    Matched MeSH terms: Machine Learning
  3. Fulltext Machine Learning-Based Performance Comparison to Diagnose Anterior Cruciate Ligament Tears
    Awan MJ, Mohd Rahim MS, Salim N, Rehman A, Nobanee H
    J Healthc Eng, 2022;2022:2550120.
    PMID: 35444781 DOI: 10.1155/2022/2550120
    In recent times, knee joint pains have become severe enough to make daily tasks difficult. Knee osteoarthritis is a type of arthritis and a leading cause of disability worldwide. The middle of the knee contains a vital portion, the anterior cruciate ligament (ACL). It is necessary to diagnose the ACL ruptured tears early to avoid surgery. The study aimed to perform a comparative analysis of machine learning models to identify the condition of three ACL tears. In contrast to previous studies, this study also considers imbalanced data distributions as machine learning techniques struggle to deal with this problem. The paper applied and analyzed four machine learning classification models, namely, random forest (RF), categorical boosting (Cat Boost), light gradient boosting machines (LGBM), and highly randomized classifier (ETC) on the balanced, structured dataset of ACL. After oversampling a hyperparameter adjustment, the above four models have achieved an average accuracy of 95.72%, 94.98%, 94.98%, and 98.26%. There are 2070 observations and eight features in the collection of three diagnosis ACL classes after oversampling. The area under curve value was approximately 0.998, respectively. Experiments were performed using twelve machine learning algorithms with imbalanced and balanced datasets. However, the accuracy of the imbalanced dataset has remained under 76% for all twelve models. After oversampling, the proposed model may contribute to the investigation of ACL tears on magnetic resonance imaging and other knee ligaments efficiently and automatically without involving radiologists.
    Matched MeSH terms: Machine Learning
  4. Fulltext Common institutional investors and the quality of management earnings forecasts-Empirical and machine learning evidences
    Yang S, Li X, Jiang Z, Xiao M
    PLoS One, 2023;18(10):e0290126.
    PMID: 37844110 DOI: 10.1371/journal.pone.0290126
    Based on the data of the Chinese A-share listed firms in China Shanghai and Shenzhen Stock Exchange from 2014 to 2021, this article explores the relationship between common institutional investors and the quality of management earnings forecasts. The study used the multiple linear regression model and empirically found that common institutional investors positively impact the precision of earnings forecasts. This article also uses graph neural networks to predict the precision of earnings forecasts. Our findings have shown that common institutional investors form external supervision over restricting management to release a wide width of earnings forecasts, which helps to improve the risk warning function of earnings forecasts and promote the sustainable development of information disclosure from management in the Chinese capital market. One of the marginal contributions of this paper is that it enriches the literature related to the economic consequences of common institutional shareholding. Then, the neural network method used to predict the quality of management forecasts enhances the research method of institutional investors and the behavior of management earnings forecasts. Thirdly, this paper calls for strengthening information sharing and circulation among institutional investors to reduce information asymmetry between investors and management.
    Matched MeSH terms: Machine Learning
  5. Special care dentistry education during the COVID-19 pandemic: The impact of online peer-assisted learning
    Mohd Suria TYI, Omar AF, Wan Mokhtar I, Rahman ANAA, Kamaruddin AA, Ahmad MS
    Spec Care Dentist, 2023;43(6):848-855.
    PMID: 37013967 DOI: 10.1111/scd.12857
    OBJECTIVES: This study aims to analyze the impact and students' perceptions of online peer-assisted learning (OPL), developed as an alternative and innovative approach to Special Care Dentistry (SCD) training during the COVID-19 pandemic. Online peer-assisted learning (OPL) is an alternative pedagogical approach that combines online education and peer-assisted teaching.

    METHODS: The OPL session was conducted by two postgraduate students in SCD (as teachers), to final year undergraduate dental students (as learners) (n = 90), supervised by two specialists in SCD-related areas (as supervisors). Vetted online pre- and post-intervention quizzes were conducted before and after the session, respectively, followed by an online validated feedback survey of the students' learning experiences. Meanwhile, a reflective session was conducted between the postgraduate students and supervisors to explore their perceptions of OPL. Quantitative data was analyzed via paired t-test (significance level, P 

    Matched MeSH terms: Learning
  6. Fulltext Analysis of the Association between Teachers' Classroom Teaching Behaviors and Students' Knowledge Acceptance Based on Psychological Data Analysis
    Yao T, Yang X
    Occup Ther Int, 2022;2022:2661398.
    PMID: 35814354 DOI: 10.1155/2022/2661398
    This paper adopts the method of psychological data analysis to conduct in-depth research and analysis on the correlation between teachers' classroom teaching behaviors and students' knowledge acceptance. Firstly, this paper proposes a health factor prediction model, which is specifically divided into clustering and then classification model and a clustering and classification synthesis model. The classroom learning process is coded, sampled, and quantified to obtain data on students' learning behaviors, and a visualization system based on classroom students' learning behaviors is designed and developed to record and analyze students' behaviors in the classroom learning process and grasp students' classroom learning. These two models use algorithms to fine-grained divide the dataset from the perspective of subject users and mental health factors, respectively, and then use decision tree algorithms to classify and predict the mental health factor information by the subject user base information. Second, based on the collected datasets, we designed comparison experiments to validate the clustering-then-classification model and the integrated clustering-classification model and selected the optimal model for comparison. Teachers should increase effective praise and encouragement behaviors; teachers should increase meaningful teacher-student interaction behaviors; teachers should be proficient in teaching media technology to reduce unnecessary time wastage. Strategies to enhance teachers' TPACK include enriching teachers' knowledge base of CK, TK, and PK; developing teachers' integration thinking; and enriching teachers' types of activities for integrating technology.
    Matched MeSH terms: Learning
  7. Fulltext Academics' continuance intention to use learning technologies during COVID-19 and beyond
    Vasuthevan K, Vaithilingam S, Ng JWJ
    PLoS One, 2024;19(1):e0295746.
    PMID: 38166113 DOI: 10.1371/journal.pone.0295746
    The COVID-19 pandemic has revolutionized the teaching pedagogy in higher education as universities are forecasted to increase investments in learning technology infrastructure to transition away from traditional teaching methods. Therefore, it is crucial to investigate whether academics intend to continually integrate learning technologies as part of a permanent pedagogical change beyond the COVID-19 pandemic. Drawing upon the Unified Theory of Acceptance and Use of Technology (UTAUT), and Expectation Confirmation Model (ECM), this study examines the salient determinants influencing the continuance intention of academics to use learning technologies in their teaching pedagogy during and after COVID-19. Primary data collected from a private university was analyzed using the partial least squares structural equation modelling technique (PLS-SEM). The findings revealed two sequential mediating relationships which serve as the mechanism linking the relationship between facilitating conditions and their continuance intention to use learning technologies during and beyond the COVID-19 pandemic.
    Matched MeSH terms: Learning
  8. A method to improve the prediction performance of cancer-gene association by screening negative training samples through gene network data
    Xu M, Abdullah NA, Md Sabri AQ
    Comput Biol Chem, 2024 Feb;108:107997.
    PMID: 38154318 DOI: 10.1016/j.compbiolchem.2023.107997
    This work focuses on data sampling in cancer-gene association prediction. Currently, researchers are using machine learning methods to predict genes that are more likely to produce cancer-causing mutations. To improve the performance of machine learning models, methods have been proposed, one of which is to improve the quality of the training data. Existing methods focus mainly on positive data, i.e. cancer driver genes, for screening selection. This paper proposes a low-cancer-related gene screening method based on gene network and graph theory algorithms to improve the negative samples selection. Genetic data with low cancer correlation is used as negative training samples. After experimental verification, using the negative samples screened by this method to train the cancer gene classification model can improve prediction performance. The biggest advantage of this method is that it can be easily combined with other methods that focus on enhancing the quality of positive training samples. It has been demonstrated that significant improvement is achieved by combining this method with three state-of-the-arts cancer gene prediction methods.
    Matched MeSH terms: Machine Learning
  9. An ensemble of bioinformatics and machine learning approaches to identify shared breast cancer biomarkers among diverse populations
    Sultan G, Zubair S
    Comput Biol Chem, 2024 Feb;108:107999.
    PMID: 38070457 DOI: 10.1016/j.compbiolchem.2023.107999
    Breast cancer continues to be a prominent cause for substantial loss of life among women globally. Despite established treatment approaches, the rising prevalence of breast cancer is a concerning trend regardless of geographical location. This highlights the need to identify common key genes and explore their biological significance across diverse populations. Our research centered on establishing a correlation between common key genes identified in breast cancer patients. While previous studies have reported many of the genes independently, our study delved into the unexplored realm of their mutual interactions, that may establish a foundational network contributing to breast cancer development. Machine learning algorithms were employed for sample classification and key gene selection. The best performance model further selected the candidate genes through expression pattern recognition. Subsequently, the genes common in all the breast cancer patients from India, China, Czech Republic, Germany, Malaysia and Saudi Arabia were selected for further study. We found that among ten classifiers, Catboost exhibited superior performance with an average accuracy of 92%. Functional enrichment analysis and pathway analysis revealed that calcium signaling pathway, regulation of actin cytoskeleton pathway and other cancer-associated pathways were highly enriched with our identified genes. Notably, we observed that these genes regulate each other, forming a complex network. Additionally, we identified PALMD gene as a novel potential biomarker for breast cancer progression. Our study revealed key gene modules forming a complex network that were consistently expressed in different populations, affirming their critical role and biological significance in breast cancer. The identified genes hold promise as prospective biomarkers of breast cancer prognosis irrespective of country of origin or ethnicity. Future investigations will expand upon these genes in a larger population and validate their biological functions through in vivo analysis.
    Matched MeSH terms: Machine Learning
  10. Fulltext Discovering Knee Osteoarthritis Imaging Features for Diagnosis and Prognosis: Review of Manual Imaging Grading and Machine Learning Approaches
    Teoh YX, Lai KW, Usman J, Goh SL, Mohafez H, Hasikin K, et al.
    J Healthc Eng, 2022;2022:4138666.
    PMID: 35222885 DOI: 10.1155/2022/4138666
    Knee osteoarthritis (OA) is a deliberating joint disorder characterized by cartilage loss that can be captured by imaging modalities and translated into imaging features. Observing imaging features is a well-known objective assessment for knee OA disorder. However, the variety of imaging features is rarely discussed. This study reviews knee OA imaging features with respect to different imaging modalities for traditional OA diagnosis and updates recent image-based machine learning approaches for knee OA diagnosis and prognosis. Although most studies recognized X-ray as standard imaging option for knee OA diagnosis, the imaging features are limited to bony changes and less sensitive to short-term OA changes. Researchers have recommended the usage of MRI to study the hidden OA-related radiomic features in soft tissues and bony structures. Furthermore, ultrasound imaging features should be explored to make it more feasible for point-of-care diagnosis. Traditional knee OA diagnosis mainly relies on manual interpretation of medical images based on the Kellgren-Lawrence (KL) grading scheme, but this approach is consistently prone to human resource and time constraints and less effective for OA prevention. Recent studies revealed the capability of machine learning approaches in automating knee OA diagnosis and prognosis, through three major tasks: knee joint localization (detection and segmentation), classification of OA severity, and prediction of disease progression. AI-aided diagnostic models improved the quality of knee OA diagnosis significantly in terms of time taken, reproducibility, and accuracy. Prognostic ability was demonstrated by several prediction models in terms of estimating possible OA onset, OA deterioration, progressive pain, progressive structural change, progressive structural change with pain, and time to total knee replacement (TKR) incidence. Despite research gaps, machine learning techniques still manifest huge potential to work on demanding tasks such as early knee OA detection and estimation of future disease events, as well as fundamental tasks such as discovering the new imaging features and establishment of novel OA status measure. Continuous machine learning model enhancement may favour the discovery of new OA treatment in future.
    Matched MeSH terms: Machine Learning
  11. Fulltext Deep Ensemble Fake News Detection Model Using Sequential Deep Learning Technique
    Ali AM, Ghaleb FA, Al-Rimy BAS, Alsolami FJ, Khan AI
    Sensors (Basel), 2022 Sep 15;22(18).
    PMID: 36146319 DOI: 10.3390/s22186970
    Recently, fake news has been widely spread through the Internet due to the increased use of social media for communication. Fake news has become a significant concern due to its harmful impact on individual attitudes and the community's behavior. Researchers and social media service providers have commonly utilized artificial intelligence techniques in the recent few years to rein in fake news propagation. However, fake news detection is challenging due to the use of political language and the high linguistic similarities between real and fake news. In addition, most news sentences are short, therefore finding valuable representative features that machine learning classifiers can use to distinguish between fake and authentic news is difficult because both false and legitimate news have comparable language traits. Existing fake news solutions suffer from low detection performance due to improper representation and model design. This study aims at improving the detection accuracy by proposing a deep ensemble fake news detection model using the sequential deep learning technique. The proposed model was constructed in three phases. In the first phase, features were extracted from news contents, preprocessed using natural language processing techniques, enriched using n-gram, and represented using the term frequency-inverse term frequency technique. In the second phase, an ensemble model based on deep learning was constructed as follows. Multiple binary classifiers were trained using sequential deep learning networks to extract the representative hidden features that could accurately classify news types. In the third phase, a multi-class classifier was constructed based on multilayer perceptron (MLP) and trained using the features extracted from the aggregated outputs of the deep learning-based binary classifiers for final classification. The two popular and well-known datasets (LIAR and ISOT) were used with different classifiers to benchmark the proposed model. Compared with the state-of-the-art models, which use deep contextualized representation with convolutional neural network (CNN), the proposed model shows significant improvements (2.41%) in the overall performance in terms of the F1score for the LIAR dataset, which is more challenging than other datasets. Meanwhile, the proposed model achieves 100% accuracy with ISOT. The study demonstrates that traditional features extracted from news content with proper model design outperform the existing models that were constructed based on text embedding techniques.
    Matched MeSH terms: Machine Learning
  12. The role of chunking in drawing Rey complex figure
    Obaidellah UH, Cheng PC
    Percept Mot Skills, 2015 Apr;120(2):535-55.
    PMID: 25706345 DOI: 10.2466/24.PMS.120v17x6
    The study investigated the effects of chunking and perceptual patterns that guide the drawings of Rey complex figure. Ten adult participants (M age=22.2 yr., SD=4.1) reproduced a single stimulus in four drawing modes including delayed recall, tracing, copying, and immediate recall across 10 sessions producing a total of 400 trials. It was hypothesized that the effect of chunking is most obvious in the free recall tasks than in the tracing or copying tasks. Measures such as pauses, patterns of drawings, and transitions among patterns of drawings suggested that participants used chunking to aid rapid learning of the diagram. The analysis of the participants' sequence of chunk production further revealed that they used a spatial schema to organize the chunks. Findings from this study provide additional evidence to support prior studies that claim graphical information is hierarchically organized.
    Matched MeSH terms: Learning/physiology*
  13. Feature extraction and classification for EEG signals using wavelet transform and machine learning techniques
    Amin HU, Malik AS, Ahmad RF, Badruddin N, Kamel N, Hussain M, et al.
    Australas Phys Eng Sci Med, 2015 Mar;38(1):139-49.
    PMID: 25649845 DOI: 10.1007/s13246-015-0333-x
    This paper describes a discrete wavelet transform-based feature extraction scheme for the classification of EEG signals. In this scheme, the discrete wavelet transform is applied on EEG signals and the relative wavelet energy is calculated in terms of detailed coefficients and the approximation coefficients of the last decomposition level. The extracted relative wavelet energy features are passed to classifiers for the classification purpose. The EEG dataset employed for the validation of the proposed method consisted of two classes: (1) the EEG signals recorded during the complex cognitive task--Raven's advance progressive metric test and (2) the EEG signals recorded in rest condition--eyes open. The performance of four different classifiers was evaluated with four performance measures, i.e., accuracy, sensitivity, specificity and precision values. The accuracy was achieved above 98 % by the support vector machine, multi-layer perceptron and the K-nearest neighbor classifiers with approximation (A4) and detailed coefficients (D4), which represent the frequency range of 0.53-3.06 and 3.06-6.12 Hz, respectively. The findings of this study demonstrated that the proposed feature extraction approach has the potential to classify the EEG signals recorded during a complex cognitive task by achieving a high accuracy rate.
    Matched MeSH terms: Machine Learning*
  14. Evaluation of attentional demands during motor learning: validity of a dual-task probe paradigm
    Goh HT, Gordon J, Sullivan KJ, Winstein CJ
    J Mot Behav, 2014;46(2):95-105.
    PMID: 24447033 DOI: 10.1080/00222895.2013.868337
    The aim of this study was to examine the validity of a 2-choice audio-vocal reaction time (RT) probe task for measuring the changes in attentional demand during practice and learning of a discrete motor task. Twenty participants practiced the motor task across 3 days and were probed with the RT task during either the preparation or execution phase of the primary task. As practice progressed, participants improved in the primary task performance and shortened the RTs to the probe task. This indicated that less attention was required to plan and execute the movement and suggested that the RT probe task was a sensitive and valid tool to measure changes in attentional demands across practice. The authors implemented several additional experimental controls to address possible confounders including unintentional learning of the probe task, primary-secondary task trade-off effects, and compliance with task priority instructions. These experimental controls further ensured the validity of the probe paradigm and interpretability of the dual-task cost findings. Our experimental methods provided confirmatory evidence for the validity of the 2-choice RT task as a means to assess attentional demands during motor learning.
    Matched MeSH terms: Learning/physiology*
  15. Human memory retention and recall processes. A review of EEG and fMRI studies
    Amin H, Malik AS
    Neurosciences (Riyadh), 2013 Oct;18(4):330-44.
    PMID: 24141456
    Human memory is an important concept in cognitive psychology and neuroscience. Our brain is actively engaged in functions of learning and memorization. Generally, human memory has been classified into 2 groups: short-term/working memory, and long-term memory. Using different memory paradigms and brain mapping techniques, psychologists and neuroscientists have identified 3 memory processes: encoding, retention, and recall. These processes have been studied using EEG and functional MRI (fMRI) in cognitive and neuroscience research. This study reviews previous research reported for human memory processes, particularly brain behavior in memory retention and recall processes with the use of EEG and fMRI. We discuss issues and challenges related to memory research with EEG and fMRI techniques.
    Matched MeSH terms: Learning/physiology
  16. The use of jigsaw learning technique in teaching medical students prescribing skills
    Sim SM, Foong CC, Tan CH, Lai PS, Chua SS, Mohazmi M
    Med Teach, 2014 Feb;36(2):182.
    PMID: 24156275 DOI: 10.3109/0142159X.2013.848977
    Matched MeSH terms: Learning*
  17. Fulltext Appreciation of learning environment and development of higher-order learning skills in a problem-based learning medical curriculum
    Mala-Maung, Abdullah A, Abas ZW
    Med J Malaysia, 2011 Dec;66(5):435-9.
    PMID: 22390096 MyJurnal
    This cross-sectional study determined the appreciation of the learning environment and development of higher-order learning skills among students attending the Medical Curriculum at the International Medical University, Malaysia which provides traditional and e-learning resources with an emphasis on problem based learning (PBL) and self-directed learning. Of the 708 participants, the majority preferred traditional to e-resources. Students who highly appreciated PBL demonstrated a higher appreciation of e-resources. Appreciation of PBL is positively and significantly correlated with higher-order learning skills, reflecting the inculcation of self-directed learning traits. Implementers must be sensitive to the progress of learners adapting to the higher education environment and innovations, and to address limitations as relevant.
    Matched MeSH terms: Problem-Based Learning*
  18. Fulltext Children with learning disabilities in the paediatric clinic, Hospital Tuanku Ja'afar Seremban: an overview
    Mariana AM, Wong SL
    Med J Malaysia, 2011 Dec;66(5):487-90.
    PMID: 22390107 MyJurnal
    The aim of the study was to document the prevalence of learning disability among the children attending the Paediatric Clinic in Hospital Tuanku Ja'afar Seremban. The demographic distribution of these patients; the age of detection of the problem; the associated medical conditions and types of intervention received by these patients were documented. Patients who were between the ages of five to twelve years were included in the study. Learning disability was divided into three categories: speech and articulation problems, academic skills disorder and other categories which included developmental delay. Children with cerebral palsy were excluded from the study. Out of 1320 patients screened, 355 were found to have learning disorders. Majority were Malays, with the male to female ratio of 1.9:1. Most of the patients stayed in Seremban. The learning problem was most commonly detected at the age of 4 years and below. The commonest type of learning disorder was developmental delay, followed by academic skills disorder, speech and academic skills problems and speech disorders. Problems that were detected early were speech problems and developmental delay. Majority of the children had associated medical conditions. Most of the patients received some form of intervention but 11.3% did not attend any intervention program at all. A strategy should be formulated and implemented to help this group of children.

    Study site: Paediatric Clinic in Hospital Tuanku Ja'afar Seremban
    Matched MeSH terms: Learning Disorders/epidemiology*
  19. Training students to learn in a problem-based learning programme
    Azer SA
    Med Educ, 2011 May;45(5):510.
    PMID: 21486331 DOI: 10.1111/j.1365-2923.2011.03952.x
    Matched MeSH terms: Problem-Based Learning/methods*
  20. Fulltext Enhancing active learning in microbiology through case based learning: experiences from an Indian medical school
    Ciraj AM, Vinod P, Ramnarayan K
    Indian J Pathol Microbiol, 2010 Oct-Dec;53(4):729-33.
    PMID: 21045402 DOI: 10.4103/0377-4929.72058
    Case-based learning (CBL) is an interactive student-centered exploration of real life situations. This paper describes the use of CBL as an educational strategy for promoting active learning in microbiology.
    Matched MeSH terms: Problem-Based Learning/methods*
Related Terms
Filters
Contact Us

Please provide feedback to Administrator ([email protected])

External Links