Birjand University of Medical Sciences

Metaheuristic algorithms have become a widely adopted approach for addressing feature selection problems in high-dimensional datasets. Among these methods, Particle Swarm Optimization (PSO) has received attention due to its simple structure, efficient search capability, and adaptability to different optimization scenarios. As a result, numerous PSO-based feature selection methods have been proposed in recent years, each introducing various modifications to improve search performance and subset quality. Despite this rapid development, a structured analysis that highlights the strengths, limitations, and practical implications of these approaches remains necessary. This survey provides a systematic examination of prominent PSO-based feature selection algorithms reported in the literature. The reviewed methods are analyzed and compared with respect to several important aspects, including search behavior, strategies used to balance exploration and exploitation, design of fitness functions for evaluating feature subsets, and commonly used evaluation criteria such as classification accuracy, dimensionality reduction rate, and computational cost. The analysis highlights the main limitations of PSO-based feature selection, including a tendency to premature convergence, sensitivity to parameter settings, and scalability issues in high-dimensional environments. Based on these observations, several open research challenges are identified and potential directions for future work are outlined in order to improve the applicability of PSO-driven feature selection methods.

Introduction: Social distancing and wearing a face mask are highly recommended to mitigate the transmission of coronavirus disease 2019 (COVID-19). However, the success of these strategies relies on individuals’ adherence and public compliance. This study was conducted to assess the level of belief in social distancing and face mask practices in communities in low- and middle-income countries (LMICs) and to identify their possible determinants. Methods: A cross-sectional study was conducted in ten LMICs countries in Asia, Africa, and South America from February to May 2021. A questionnaire was used to assess the belief, practice, and their plausible determinants. Identification of the associated determinants was performed using a logistic regression model. Results: Our data revealed that only 62.6% and 66.9% of the participants had good beliefs in social distancing and good face mask practices, respectively. Residing in the Americas, having a healthcare-related job, knowing people in immediate social environment who are or have been infected and exposure to information of COVID-19 cases on social media or TV were factors significantly associated with good belief in social distancing. Residing country, gender, monthly household income, type of job and exposure to information of COVID-19 cases were significantly associated with face mask wearing practice. Conclusion: The proportion of participants having good beliefs in social distancing and good face mask practices is relatively low (<75%). Hence, sustained health campaigns regarding social distancing benefits and face mask-wearing practices during COVID-19 are critical in LMICs.

Background/Objectives: Cervical cancer remains a major cause of morbidity and mortality among women worldwide, marked by stark geographic and socioeconomic disparities. Preventable via HPV vaccination and screening, progress toward elimination varies widely across and within countries. This narrative review synthesizes the epidemiology, including incidence, mortality, survival, and stage distribution, as well as risk factors and the coverage/equity of HPV screening and vaccination programs. Methods: Comprehensive searches were performed in PubMed, Web of Science, Scopus, and Google Scholar (no date restrictions; English only). Included were original epidemiological studies, systematic reviews, meta-analyses, and international reports on burden, risk factors, or prevention indicators. Data were qualitatively synthesized into three themes: epidemiological patterns, risk factors, and screening/prevention programs. Results: Persistent high-risk HPV infection causes nearly all cervical cancers, predominantly HPV16/18, with regional variation in other types. Strong co-factors include HIV immunosuppression, early sexual debut, multiple partners, high parity, long-term oral contraceptive use, and smoking. Inequalities in incidence, late diagnosis, and survival are driven by socioeconomic disadvantages, low education, rural residence, and poor health system access. Screening ranges from cytology/VIA to primary HPV testing, but coverage is low and inequitable in high-burden settings. HPV vaccination has expanded yet faces major gaps in low- and middle-income countries. Conclusions: Cervical cancer burden concentrates in low-resource and marginalized populations. Global elimination demands accelerated, equitable scale-up of HPV vaccination and screening, alongside health system strengthening and barrier reduction.

Large language models (LLMs) have recently gained prominence in healthcare content provision due to their numerous advantages. Despite these benefits, LLMs exhibit notable limitations in this domain. This study aimed to systematically identify the limitations of LLMs in provision of healthcare content. This study was a systematic review conducted in September 2025, including articles published in English between 2018 and 2025. Searches were performed in PubMed, Scopus, and the Cochrane Database of Systematic Reviews. Two independent evaluators screened the references and assessed quality of the selected studies using the Authority, Accuracy, Coverage, Objectivity, Date, and Significance (AACODS) checklist. Data were analyzed using Boyatzis's qualitative thematic approach with an inductive methodology, applying the input-process-output (IPO) model as the analytical framework. A total of 81 studies were included in the final analysis. The included studies were predominantly of high quality and demonstrated minimal risk of bias. The thematic analysis identified key themes: data limitations, dependence on input and prompt quality, accessibility issues, model design and architecture constraints, interaction challenges, response quality and comprehensiveness, and ethical, safety, and regulatory concerns. The study identified multiple limitations of LLMs in healthcare, with output issues being most common. In this regard, the most frequently cited limitation was the accuracy gap. However, these output issues were mainly resulted from flaws in input data, emphasizing the crucial role of input quality. The study also proposed strategies to address these challenges.

Chronic pain is highly prevalent among older adults and has been shown to be associated with differences in cognitive function. While pain intensity reflects the severity of pain, pain interference assesses the extent to which pain disrupts daily activities. Distinguishing between these dimensions of pain and their associations with cognitive function may improve understanding of how pain relates to cognitive health in older populations. This cross-sectional study was conducted using baseline data from the Birjand Longitudinal Aging Study (BLAS). Pain was assessed using the Brief Pain Inventory (BPI), with pain severity defined as the mean of items 3–6 and pain interference assessed using item 9. Cognitive impairment was determined based on the combination results of the Six Item Cognitive Impairment Test (6-CIT), the Abbreviated Mental Test Score (AMTS), and the Category Fluency Test (CFT). Multiple logistic regression models were employed to examine associations between pain measures and cognitive impairment, adjusting for potential confounders. Among 1,343 participants (mean age: 69.73 ± 7.53 years; 51.82