Nested ensemble selection: An effective hybrid feature selection method.

It has been shown that while feature selection algorithms are able to distinguish between relevant and irrelevant features, they fail to differentiate between relevant and redundant and correlated features. To address this issue, we propose a highly effective approach, called Nested Ensemble Selection (NES), that is based on a combination of filter and wrapper methods. The proposed feature selection algorithm differs from the existing filter-wrapper hybrid methods in its simplicity and efficiency as well as precision. The new algorithm is able to separate the relevant variables from the irrelevant as well as the redundant and correlated features. Furthermore, we provide a robust heuristic for identifying the optimal number of selected features which remains one of the greatest challenges in feature selection. Numerical experiments on synthetic and real-life data demonstrate the effectiveness of the proposed method. The NES algorithm achieves perfect precision on the synthetic data and near optimal accuracy on the real-life data. The proposed method is compared against several popular algorithms including mRMR, Boruta, genetic, recursive feature elimination, Lasso, and Elastic Net. The results show that NES significantly outperforms the benchmarks algorithms especially on multi-class datasets.

Suspicious Behavior Detection with Temporal Feature Extraction and Time-Series Classification for Shoplifting Crime Prevention.

The rise in crime rates in many parts of the world, coupled with advancements in computer vision, has increased the need for automated crime detection services. To address this issue, we propose a new approach for detecting suspicious behavior as a means of preventing shoplifting. Existing methods are based on the use of convolutional neural networks that rely on extracting spatial features from pixel values. In contrast, our proposed method employs object detection based on YOLOv5 with Deep Sort to track people through a video, using the resulting bounding box coordinates as temporal features. The extracted temporal features are then modeled as a time-series classification problem. The proposed method was tested on the popular UCF Crime dataset, and benchmarked against the current state-of-the-art robust temporal feature magnitude (RTFM) method, which relies on the Inflated 3D ConvNet (I3D) preprocessing method. Our results demonstrate an impressive 8.45-fold increase in detection inference speed compared to the state-of-the-art RTFM, along with an F1 score of 92%,outperforming RTFM by 3%. Furthermore, our method achieved these results without requiring expensive data augmentation or image feature extraction.

Development and Psychometric Evaluation of the Antibiotic Knowledge and Consumption Tool (AKCT).

Knowledge of antibiotics and awareness of microbial resistance are essential for appropriate antibiotic consumption. This study aimed to develop and validate a measure of antibiotic knowledge and consumption (AKCT) and to make it available in the Arabic language and context. The tool was developed and applied on individuals ≥ 18 years, with mastery of Arabic or English. Exploratory factor analysis using principal-component analysis tested the psychometric properties of the items. AKCT scores were compared with the Infectious Numeracy Test (INT) scores to establish convergent validity. Cronbach's α > 0.7 measured reliability. Three hundred-eighty-six participants completed the questionnaire, achieving a 95.3% response rate. Five components were retained after factor analysis: Side-effects and resistance, Access to antibiotics, Recovery after use, Antibiotics use indications, and Body response. Cronbach's α = 0.85. The mean ± SD of AKCT = 9.82 ± 3.85 (range = 7−20); lowest scores were related to "Side-effects and resistance" (2.32 ± 2.00, max = 7) and "Antibiotic use indications" (1.61 ± 1.29, max = 5). Scores on the AKCT and INT positively correlated. The AKCT is a valuable, valid, and reliable tool developed for measurement of antibiotic knowledge and consumption behaviors to identify specific areas needing improvements; hence, targeted interventions are devised.

Modeling of the In Vitro Release Kinetics of Sonosensitive Targeted Liposomes.

Targeted liposomes triggered by ultrasound are a promising drug delivery system as they potentially improve the clinical outcomes of chemotherapy while reducing associated side effects. In this work, a comprehensive model fitting was performed for a large dataset of liposomal release profiles with seven targeting moieties (albumin, cRGD, estrone, hyaluronic acid, Herceptin, lactobionic acid, and transferrin) in addition to the control liposomes under ultrasound release protocols. Two levels of ultrasound frequencies were tested: low frequency (20 kHz) at 6.2, 9, and 10 mW/cm2 as well as high frequencies (1.07 MHz and 3 MHz) at 10.5 and 173 W/cm2. At a low frequency, Hixson-Crowell, Korsmeyer-Peppas, Gompertz, Weibull, and Lu-Hagen showed good fits to our release profiles at all three power densities. At high frequencies, the former three models reflected the best fit. These models will aid in predicting drug release profiles for future in vitro studies.

The relationship between diabetes distress, medication taking, glycaemic control and self-management.

Background Diabetes distress (DD) has broad-ranging effects on type 2 diabetes (T2DM) management and outcomes. DD research is scarce among ethnic minority groups, particularly Arabic-speaking immigrant communities. To improve outcomes for these vulnerable groups, healthcare providers, including pharmacists, need to understand modifiable predictors of DD. Aim To assess and compare DD and its association with medication-taking behaviours, glycaemic control, self-management, and psychosocial factors among first-generation Arabic-speaking immigrants and English-speaking patients of Anglo-Celtic background with diabetes, and determine DD predictors. Setting Various healthcare settings in Australia. Method A multicentre cross-sectional study was conducted. Adults with T2DM completed a survey comprised of validated tools. Glycated haemoglobin, blood pressure, and lipid profile were gathered from medical records. Multiple linear regression models were computed to assess the DD predictors. Main outcome measure Diabetes distress level. Results Data was analysed for 696 participants: 56.3% Arabic-speaking immigrants and 43.7% English-speaking patients. Compared with English-speaking patients, Arabic-speaking immigrants had higher DD, lower medication adherence, worse self-management and glycaemic control, and poorer health and clinical profile. The regression analysis demonstrated that higher DD in Arabic-speaking immigrants was associated with cost-related medication underuse and lower adherence to exercise, younger age, lower education level, unemployment, lower self-efficacy, and inadequate glycaemic control. Whereas among English-speaking patients, higher DD was associated with both cost- and non-cost-related underuse of medication and lower dietary adherence. Conclusion Results provided new insights to guide healthcare providers on reducing the apparent excess burden of DD among Arabic-speaking immigrants and potentially improve medication adherence, glycaemic control, and self-management.

Machine learning based approach to exam cheating detection.

The COVID-19 pandemic has impelled the majority of schools and universities around the world to switch to remote teaching. One of the greatest challenges in online education is preserving the academic integrity of student assessments. The lack of direct supervision by instructors during final examinations poses a significant risk of academic misconduct. In this paper, we propose a new approach to detecting potential cases of cheating on the final exam using machine learning techniques. We treat the issue of identifying the potential cases of cheating as an outlier detection problem. We use students' continuous assessment results to identify abnormal scores on the final exam. However, unlike a standard outlier detection task in machine learning, the student assessment data requires us to consider its sequential nature. We address this issue by applying recurrent neural networks together with anomaly detection algorithms. Numerical experiments on a range of datasets show that the proposed method achieves a remarkably high level of accuracy in detecting cases of cheating on the exam. We believe that the proposed method would be an effective tool for academics and administrators interested in preserving the academic integrity of course assessments.

A Bootstrap Framework for Aggregating within and between Feature Selection Methods.

In the past decade, big data has become increasingly prevalent in a large number of applications. As a result, datasets suffering from noise and redundancy issues have necessitated the use of feature selection across multiple domains. However, a common concern in feature selection is that different approaches can give very different results when applied to similar datasets. Aggregating the results of different selection methods helps to resolve this concern and control the diversity of selected feature subsets. In this work, we implemented a general framework for the ensemble of multiple feature selection methods. Based on diversified datasets generated from the original set of observations, we aggregated the importance scores generated by multiple feature selection techniques using two methods: the Within Aggregation Method (WAM), which refers to aggregating importance scores within a single feature selection; and the Between Aggregation Method (BAM), which refers to aggregating importance scores between multiple feature selection methods. We applied the proposed framework on 13 real datasets with diverse performances and characteristics. The experimental evaluation showed that WAM provides an effective tool for determining the best feature selection method for a given dataset. WAM has also shown greater stability than BAM in terms of identifying important features. The computational demands of the two methods appeared to be comparable. The results of this work suggest that by applying both WAM and BAM, practitioners can gain a deeper understanding of the feature selection process.

Patient expectations and willingness to use primary care pharmacy services in the United Arab Emirates.

OBJECTIVES: No previous studies have investigated patients' expectations of, or interest in, using primary care pharmacy services in the UAE. The study investigated primary care services that patients would use if they were provided through community pharmacies. METHODS: A questionnaire was developed in English and translated to Arabic. Participants were asked to rate their likelihood of using future primary care services on a three-point Likert-type scale (very likely-unlikely). Services included screening for disease, monitoring of disease control, health advice and referral, lifestyle and preventive care, supply of printed information, counselling on medication use and side effects, patient record keeping, and pharmacist intervention in chronic disease. The questionnaire was distributed to patrons of public places by hand to eligible participants: ≥21 years, taking at least one scheduled prescription medication and having adequate Arabic or English proficiency. KEY FINDINGS: Areas of most interest were as follows: the pharmacist explaining how to use medications (Median = 3, interquartile range (IQR) = 1), the pharmacist advising on side effects of medications (Median = 3, IQR = 1), receiving advice on how to use devices (Median = 3, IQR = 1) and receiving printed information about medications (Median = 3, IQR = 1). Participants were least supportive of pharmacists keeping medication records (Median = 2, IQR = 1) and intervening in chronic disease management (Median = 2, IQR = 1). Participants anticipated a caring professional service by pharmacists and a pharmacy layout that allows private communication. CONCLUSIONS: Participants were interested in accessing enhanced primary care services by community pharmacists. Participants were more supportive of receiving information about medications, help in self-management and monitoring, and advice about disease prevention than management of chronic disease.

Assessing patient satisfaction with community pharmacy in the UAE using a newly-validated tool.

BACKGROUND: Patient satisfaction has become an integral component of the quality of healthcare services. It has been used for the purpose of performance assessment, reimbursement, and quality management of health service delivery. It has been suggested that patient satisfaction could be a predictor of health-related behavior. OBJECTIVES: To develop and validate a tool for use within the Arabic context to assess patient satisfaction. To assess patient satisfaction with current community pharmacy services in the UAE using the validated tool. METHODS: A systematic process was used to develop an assessment tool that could be used within the Arabic context and establish its validity and reliability. Survey participants assessed their satisfaction with the services based on a 5-point Likert-type scale: Poor = 1, Fair = 2, Good = 3, Very good = 4, Excellent = 5. The anonymous questionnaire was distributed over a 5-month period to eligible participants in public places such as malls and shopping markets, in various emirates across the UAE. Those who were 21 years or older, taking at least one scheduled (regular) medication and having adequate Arabic or English language proficiency were included. RESULTS: The instrument comprised four dimensions: Information, Relationship, Accessibility and Availability. Participants required more information about medications and self-management (Mean = 2.49 ± 1.19). Measures of competence, i.e., care, interest, time, confidence and trust, could also be improved (Mean = 3.05 ± 1.07). Accessibility scores measuring physical, geographical and financial items were lowest (Mean = 2.80 ± 1.33). Overall scores on availability of medications indicated relative satisfaction with this dimension (Mean = 3.51 ± 0.7). CONCLUSIONS: This study is the first to use a patient satisfaction tool specifically developed for the Arabic context. Patient satisfaction scores in all dimensions were significantly lower than published data, suggesting patients have unmet expectations of community pharmacy services in the UAE. Stakeholders could utilize this information to help in the design and delivery of improved services that could lead to increased demand.

Community pharmacy services in the United Arab Emirates.

OBJECTIVES: To identify the type and frequency of services provided through community pharmacies in the United Arab Emirates (UAE). METHODS: A survey was conducted using an anonymous questionnaire distributed by hand to 700 community pharmacies. Items included information about the pharmacists and pharmacies, type of products sold, type and extent of enhanced services provided and perceived barriers to providing these services. KEY FINDINGS: Most pharmacies provided a wide range of medicinal and non-medicinal products. The frequency with which services were provided was assessed on a scale of 1 (never) to 5 (always). Enhanced professional services were not provided to a large extent in most pharmacies. Fewer than one-third (29%) reported they always supplied printed information to patients (mean = 3.37, 95% confidence interval = 3.23-3.52); fewer than one-third (28%) counselled patients on a regular basis (3.25, 3.09-3.40); nearly two-thirds (62%) reported monitoring patients' adherence to therapy at least sometimes (2.96, 2.81-3.10). Most pharmacies (92%) in the UAE did not routinely keep patient records (2.09, 1.96-2.32). While just over a quarter of respondents claimed that they always reported medication errors (27%) and adverse drug reactions (28%), these activities were not often performed in around 40% of pharmacies. CONCLUSIONS: This is the first study to explore the type and extent of professional services provided through community pharmacies in the UAE and provides baseline data critical to inform the development of strategies to improve the quality of community pharmacy services.

Community pharmacy in the United Arab Emirates: characteristics and workforce issues.

OBJECTIVES: To determine the characteristics and workforce issues of community pharmacy practice in the United Arab Emirates (UAE). METHODS: Data collection was by anonymous cross-sectional survey. Questionnaires were distributed by hand to 700 community pharmacies to collect information about the participating pharmacists, pharmacy characteristics, the types of products and professional pharmacy services available to patients, and the barriers to offering professional services. KEY FINDINGS: A total of 344 pharmacists (49%) responded. Most were male (64%), had been in practice for less than 10 years (mean = 9.3, 95% confidence interval (CI) = 8.4-10.0) and were trained in India (35%) or Egypt (15%). The pharmacies were open for business 7 days/week (mean = 6.8, 95% CI = 6.7-8.8) with an average working day of 13 h (mean = 12.9, 95% CI = 12.7-13.2) and were mostly owned by independent non-pharmacists (70%). The pharmacies employed on average 2.6 full-time-equivalent (FTE) pharmacists (95% CI = 2.3-2.8) with 74% employing 1.8 FTE pharmacy assistants (95% CI = 1.7-2.0) and 47% employing trainee pharmacists (mean = 1.8 FTE, 95% CI = 1.6-2.0). Around three-quarters of the pharmacies dispensed fewer than 100 prescriptions (75%) and responded to fewer than 100 requests for over-the-counter medicines (69%) per day. Most pharmacists encountered limited immediate access to up-to-date resources. CONCLUSIONS: This is the first study to explore the characteristics of community pharmacy practice in the UAE. The study provides baseline data which are critical to inform the development of strategies to improve the quality of community pharmacy services in the UAE.

Assessment of the current practice of antibiotic skin testing in a tertiary hospital in United Arab Emirates.

INTRODUCTION: Skin testing can be a useful diagnostic tool to identify patients who are allergic to penicillin. Procedures for skin testing in the United Arab Emirates have not been standardized. The aim of this study was to examine the current practice of antibiotic skin testing in a tertiary hospital in the United Arab Emirates (UAE). METHODOLOGY: This was a prospective cross-sectional study conducted in Al Qassimi Hospital, in which the medical records of all patients who were prescribed antibiotics over an eight-week period were screened to evaluate patients' history, indication for performing the test, results, and documentation of findings. RESULTS: During the study period 357 patients received parenteral antibiotics, of which 238 had one skin test, 21 had two skin tests, and one patient had four skin tests. Skin testing was performed without regard for patient history. Documentation of both positive and negative results was poor. There was no standard technique for skin testing used within the institution, and significant variations were noted between wards. In most cases the techniques used deviated from recommended procedures in the medical literature. CONCLUSIONS: Standardized guidelines for antibiotic skin testing should be established and implemented as soon as possible using recommended international guidelines.

Intragroup differences in risk factors for breastfeeding outcomes in a multicultural community.

A sample of 221 women who delivered at Al Qassimi Hospital, Sharjah, United Arab Emirates, were included in this prospective study to identify breastfeeding patterns at day 1, 1 month, and 6 months postpartum. The exclusive breastfeeding rate was 76.5% on day 1, 48.4% at 1 month, and 13.3% at 6 months. At 6 months, 16.1% had stopped breastfeeding. Simple and multivariable binary logistic regression analyses were used to identify factors associated with better breastfeeding outcomes. Nationality significantly affected exclusive breastfeeding at day 1 and 1 month. Pethidine use was associated with lower levels of exclusive breastfeeding at 1 month. Education was the most significant determinant of breastfeeding behavior at 6 months. Effects of the interrelationships between factors were examined and shown to influence breastfeeding outcomes in different population subgroups. The findings of this study suggest that strategies to improve breastfeeding should focus on risk factors specific to the population subgroup.