Rowzatul Zannat, Abdullah Al Shafi, Abdul Muntakim
Increased access to reliable health information is essential for non-English-speaking populations, yet resources in Bangla for disease prediction remain limited. This study addresses this gap by developing a comprehensive Bangla symptoms-disease dataset containing 758 unique symptom-disease relationships spanning 85 diseases. To ensure transparency and reproducibility, we also make our dataset publicly available. The dataset enables the prediction of diseases based on Bangla symptom inputs, supporting healthcare accessibility for Bengali-speaking populations. Using this dataset, we evaluated multiple machine learning models to predict diseases based on symptoms provided in Bangla and analyzed their performance on our dataset. Both soft and hard voting ensemble approaches combining top-performing models achieved 98\% accuracy, demonstrating superior robustness and generalization. Our work establishes a foundational resource for disease prediction in Bangla, paving the way for future advancements in localized health informatics and diagnostic tools. This contribution aims to enhance equitable access to health information for Bangla-speaking communities, particularly for early disease detection and healthcare interventions.
Background: Nurses perform many daily care tasks that expose them to work-related musculoskeletal disorders (WMSDs). Many studies have reported a high prevalence worldwide. Analyses by continent have provided a better understanding of the WMSD occurrence, but none have yet been conducted among African nurses. The aim was to conduct a systematic review analysis with meta-analysis of the overall WMSD prevalence and the prevalence by body area among nurses in Africa. Methods: The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) method was used to present the results in the form of a systematic review analysis with meta-analysis. PubMed/Medline, ScienceDirect, Google Scholar, Mendeley, and Science.gov were explored between August 20 and 29, 2025 to identify studies that investigated the overall and body area WMSD prevalence among African nurses of any specialty without a date limit. Studies were included if they were cross sectional survey assessing the WMSD prevalence among nurses of any specialty or department working in Africa. Any study that was not a peer-reviewed cross-sectional survey published in English, that did not involve African nurses, or that did not report, or sufficiently detail data on the prevalence was excluded. The quality of each article included was assessed using the cross-sectional study assessment tool (AXIS). A meta-analysis with quantification of heterogeneity (Cochran’s Q test and I2 statistic) was conducted. Based on these parameters, a fixed or random effects model was selected to estimate the prevalence. Forest plots were used to summarize the overall, neck, upper back, lower back, shoulder, elbow, wrist, hip, knee, and ankle WMSD prevalence. Results: Nineteen cross-sectional studies were selected from the 4,305 identified studies, involving 4,670 African nurses from 10 countries. A significant heterogeneity was highlighted between studies (Cochran’s Q test and I2 statistic). Lower back [59.5%, 95% confidence interval (CI): 52.8–66.2%, 4,670 participants], neck (35.4%, 95% CI: 28.0–42.8%, 4,670 participants), and knee (34.4%, 95% CI: 27.2–41.6%, 4,601 participants) were the most exposed areas. The overall WMSD prevalence was pooled at 74.6% (95% CI: 67.0−82.3%, 4,266 nurses). Discussion: Comparison of these results with the literature showed that African nurses were less affected than those on other continents. However, the data were highly heterogeneous. Due to the numerous risk factors associated with nursing work, it is necessary to continue research projects and educational activities, as well as the development of health policies aimed at improving quality of life at work, specifically by expanding the investigation using subgroup analysis.
Maddison N. Salois, Saiphone Webb, Isaiah A. Proctor
et al.
ABSTRACT Ankyloblepharon‐ectodermal defects‐cleft lip/palate (AEC) is a disorder caused by autosomal‐dominant mutations in the TP63 gene. AEC is characterised by the presence of severe and painful skin erosions that can take years to heal. Current treatment options for these devastating lesions are limited, highlighting the need for new therapeutic strategies. We previously generated keratinocytes from patient‐derived induced pluripotent stem cells (iPSC‐K) and identified defects in several cell adhesion complexes, including desmosomes, hemidesmosomes and focal adhesions. In the present study, we developed a complementary in vitro model using NTERT keratinocytes transduced with lentiviral constructs expressing AEC‐related TP63 mutations (N‐AEC). This model allows for the large‐scale production of disease‐relevant material, overcoming the limitations of iPSC‐derived keratinocytes, which have the characteristics of primary keratinocytes, including limited cell doublings and lifespan. We demonstrate that N‐AEC keratinocytes exhibit key defects observed in AEC iPSC‐K and AEC patient skin, including downregulation of cell adhesion proteins. In addition, 3D epidermal equivalents generated from these cells replicate pathological features seen in AEC patient skin, such as intra‐epidermal cysts, reduced desmosomal protein expression and altered expression of differentiation markers. Our N‐AEC model provides a valuable tool for investigating the mechanisms underlying skin fragility in AEC and other genetic skin disorders and advances the potential for novel therapeutic development.
Routine regulatory requirements for large comparative efficacy trials (CETs) to support marketing approval of monoclonal antibody (mAb) biosimilars have been the focus of extensive debate in the last few years. This review examines the mounting evidence, accumulated over the past decade, focusing on relevant literature and data published in the European Product Assessment Reports (EPARs) for the fifteen anti-TNFα biosimilars approved to date. The potential for residual uncertainties that may require resolution through CETs following comparative physico-chemical, in-vitro potency, and single dose studies in healthy subjects is examined. It is noted that structural and physicochemical differences between biosimilars and reference products are detectable using modern analytical methods at levels well below those that could impact clinical outcomes, and that in vitro potency testing is fully capable of revealing clinically relevant differences. Additionally, comparative pharmacokinetic studies in healthy participants provide a sensitive assessment of potential differences in drug exposure and immunogenicity. The added value of CETs is further questioned in the light of the fact that anti-TNFα’s display a flat dose-response relationship, meaning that unlike cell-based assays, CETs have limited sensitivity to detect potency differences. Initial concerns about extrapolating data from rheumatoid arthritis studies to support marketing approval of other indications such as inflammatory bowel disease for which not all anti-TNFα’s are effective have been alleviated by post-approval studies. Additionally, as of the end of 2024, no cases where clinical efficacy data were necessary to resolve residual quality concerns have arisen following regulatory assessment of 56 mAb biosimilars and fusion proteins. CETs add significant cost and delay to the development of biosimilars and the time is now ripe to re-examine the need for these CET’s and for further evolution in regulatory thinking.
Valeriia Husak, Olena Povelychenko, Valentyna Maltseva
et al.
Abstract Background In contemporary regenerative medicine, platelet concentrates (PCs) are actively used as a promising method to support the regenerative process. However, the lack of standardized preparation protocols and the selection of PC types limits their broad clinical implementation, particularly in patients with non-unions or bone defects. Therefore, the aim of this study was to determine the concentration of platelets, leukocytes, and growth factors such as vascular endothelial growth factor (VEGF-A), transforming growth factor beta (TGF-β1), platelet-derived growth factor (PDGF-BB) in three types of PC, such as platelet-rich plasma (PRP), leukocyte- and platelet-rich plasma (L-PRP), platelet-rich fibrin (PRF) using a single donor model in patients with large bone defects after combat trauma compared to healthy individuals. Methods Blood for PRP, L-PRP and PRF was collected from 30 participants. 19 healthy volunteers and 11 patients with long bone defects after combat injuries. For the production of three types of PC, 15 ml of blood was taken from each participant. The cellular composition was determined using an automated hematological analyzer. The concentration of growth factors VEGF-A, TGF-β1, PDGF-BB was determined by ELISA. Results Differences in cellular composition and growth factor concentration between PC types were identified in all study participants. The concentration of platelets in PCs was distributed as follows: L-PRP > PRP > PRF; however, this did not affect the concentration of growth factors. The concentration of growth factors in PCs from patients with bone defects did not differ from that of healthy individuals. In patients with bone defects, it was not possible to achieve an enrichment of leukocyte concentration in L-PRP compared to the baseline level of whole blood; however, this parameter did not differ from that of healthy individuals. Conclusions Growth factor concentrations were similar in patients and healthy individuals, but patients had differences in L-PRP leukocyte enrichment and lower platelet recovery in PRF. This study highlights the need to consider platelet concentrate characteristics when selecting products for regenerative therapy. Clinical trial number Not applicable.
Physical contact or proximity is often a necessary condition for the spread of infectious diseases. Common destinations, typically referred to as hubs or points of interest, are arguably the most effective spots for the type of disease spread via airborne transmission. In this work, we model the locations of individuals (agents) and common destinations (hubs) by random spatial point processes in $\mathbb{R}^d$ and focus on disease propagation through agents visiting common hubs. The probability of an agent visiting a hub depends on their distance through a connection function $f$. The system is represented by a random bipartite geometric (RBG) graph. We study the degrees and percolation of the RBG graph for general connection functions. We show that the critical density of hubs for percolation is dictated by the support of the connection function $f$, which reveals the critical role of long-distance travel (or its restrictions) in disease spreading.
We introduce LLM x MapReduce-V3, a hierarchically modular agent system designed for long-form survey generation. Building on the prior work, LLM x MapReduce-V2, this version incorporates a multi-agent architecture where individual functional components, such as skeleton initialization, digest construction, and skeleton refinement, are implemented as independent model-context-protocol (MCP) servers. These atomic servers can be aggregated into higher-level servers, creating a hierarchically structured system. A high-level planner agent dynamically orchestrates the workflow by selecting appropriate modules based on their MCP tool descriptions and the execution history. This modular decomposition facilitates human-in-the-loop intervention, affording users greater control and customization over the research process. Through a multi-turn interaction, the system precisely captures the intended research perspectives to generate a comprehensive skeleton, which is then developed into an in-depth survey. Human evaluations demonstrate that our system surpasses representative baselines in both content depth and length, highlighting the strength of MCP-based modular planning.
This editorial, “Sarcopenia: a hidden comorbidity of established rheumatoid arthritis” emphasizes the critical role of addressing comorbidities in rheumatoid arthritis (RA) management, focusing particularly on the clinical impact of sarcopenia. The first section highlights how advances in treating immune-mediated rheumatic diseases have improved RA management but also underscore the increasing necessity to integrate comorbidity management to enhance patient outcomes. The second part focused into sarcopenia as a significant yet overlooked comorbidity in RA, discussing its prevalence, impact on life quality, and the complexities of its diagnosis and management. The editorial advocates for a multidisciplinary approach involving rheumatologists, nurses, and primary care physicians to effectively tackle this issue. A call to action from scientific societies is suggested to raise awareness among healthcare professionals about sarcopenia, aiming to improve care for RA patients.
Determining the reference base of anthropometric parameters on a sample of elite athletes is one of the foundations of further research and forming a clearer picture of each sport and sports discipline. In this study, the aim was to describe the anthropometric and somatotype profiles of elite Finn class sailors and to determine the differences in the measured parameters between sailors at different levels of general competitive success. The subject sample included 57 Finn class sailors who competed at the open Finn European Championship. A set of 25 anthropometric variables were applied. The sailors were divided into three groups according to their level of general competitive success using World Sailing Rankings. Finn sailors had higher average values in almost all morphological characteristics when compared to the sailors in other Olympic classes. Considering the average values of somatotype categories, we determined that Finn sailors fit the <i>endomorphic mesomorph</i> somatotype category (3.94 ± 1.19 − 5.50 ± 1.19 − 1.63 ± 0.74). Significant differences were observed between more-successful, medium, and less-successful sailors in the variables of <i>age</i>, <i>body mass</i>, <i>muscle mass</i>, <i>arm muscle mass</i>, and <i>endomorphy rating</i>. These results indicate the possibility of selection processes and/or adaptation to sailing occurring in the Finn class. The anthropometric characteristics of Finn sailors compared to sailors in Olympic classes further “support” the Finn class being called the “heavy dinghy” male class. This study on anthropometric parameters, determined via a sample of top Finn sailors, may be of great help to coaches and young sailors when deciding on the selection of an adult sailing class.
Rare diseases, despite their low individual incidence, collectively impact around 300 million people worldwide due to the vast number of diseases. The involvement of multiple organs and systems, and the shortage of specialized doctors with relevant experience, make diagnosing and treating rare diseases more challenging than common diseases. Recently, agents powered by large language models (LLMs) have demonstrated notable applications across various domains. In the medical field, some agent methods have outperformed direct prompts in question-answering tasks from medical examinations. However, current agent frameworks are not well-adapted to real-world clinical scenarios, especially those involving the complex demands of rare diseases. To bridge this gap, we introduce RareAgents, the first LLM-driven multi-disciplinary team decision-support tool designed specifically for the complex clinical context of rare diseases. RareAgents integrates advanced Multidisciplinary Team (MDT) coordination, memory mechanisms, and medical tools utilization, leveraging Llama-3.1-8B/70B as the base model. Experimental results show that RareAgents outperforms state-of-the-art domain-specific models, GPT-4o, and current agent frameworks in diagnosis and treatment for rare diseases. Furthermore, we contribute a novel rare disease dataset, MIMIC-IV-Ext-Rare, to facilitate further research in this field.
Existing plant disease classification models have achieved remarkable performance in recognizing in-laboratory diseased images. However, their performance often significantly degrades in classifying in-the-wild images. Furthermore, we observed that in-the-wild plant images may exhibit similar appearances across various diseases (i.e., small inter-class discrepancy) while the same diseases may look quite different (i.e., large intra-class variance). Motivated by this observation, we propose an in-the-wild multimodal plant disease recognition dataset that contains the largest number of disease classes but also text-based descriptions for each disease. Particularly, the newly provided text descriptions are introduced to provide rich information in textual modality and facilitate in-the-wild disease classification with small inter-class discrepancy and large intra-class variance issues. Therefore, our proposed dataset can be regarded as an ideal testbed for evaluating disease recognition methods in the real world. In addition, we further present a strong yet versatile baseline that models text descriptions and visual data through multiple prototypes for a given class. By fusing the contributions of multimodal prototypes in classification, our baseline can effectively address the small inter-class discrepancy and large intra-class variance issues. Remarkably, our baseline model can not only classify diseases but also recognize diseases in few-shot or training-free scenarios. Extensive benchmarking results demonstrate that our proposed in-the-wild multimodal dataset sets many new challenges to the plant disease recognition task and there is a large space to improve for future works.
Lung diseases have become a prevalent problem throughout the United States, affecting over 34 million people. Accurate and timely diagnosis of the different types of lung diseases is critical, and Artificial Intelligence (AI) methods could speed up these processes. A dual-stage vision transformer is built throughout this research by integrating a Vision Transformer (ViT) and a Swin Transformer to classify 14 different lung diseases from X-ray scans of patients with these diseases. The proposed model achieved an accuracy of 92.06% on a label-level when making predictions on an unseen testing subset of the dataset after data preprocessing and training the neural network. The model showed promise for accurately classifying lung diseases and diagnosing patients who suffer from these harmful diseases.
Heart disease is a serious worldwide health issue because it claims the lives of many people who might have been treated if the disease had been identified earlier. The leading cause of death in the world is cardiovascular disease, usually referred to as heart disease. Creating reliable, effective, and precise predictions for these diseases is one of the biggest issues facing the medical world today. Although there are tools for predicting heart diseases, they are either expensive or challenging to apply for determining a patient's risk. The best classifier for foretelling and spotting heart disease was the aim of this research. This experiment examined a range of machine learning approaches, including Logistic Regression, K-Nearest Neighbor, Support Vector Machine, and Artificial Neural Networks, to determine which machine learning algorithm was most effective at predicting heart diseases. One of the most often utilized data sets for this purpose, the UCI heart disease repository provided the data set for this study. The K-Nearest Neighbor technique was shown to be the most effective machine learning algorithm for determining whether a patient has heart disease. It will be beneficial to conduct further studies on the application of additional machine learning algorithms for heart disease prediction.
Seyed Ali Alamdaran, Mohadeseh Taheri-nezhad, Ahmad Nouri
et al.
Abstract Background Septic arthritis is an important differential diagnosis of hip joint pain. Joint aspiration analysis is a necessary diagnostic measure for septic arthritis. In order to reduce the need for joint aspiration, we compared the combination of ultrasound findings and laboratory findings to separate septic arthritis from reactive arthritis. Methods Children aged < 14 years who were referred to Akbar pediatric hospital in 2020–2022 with hip pain or limping were included in this longitudinal study. Participants underwent ultrasound examinations of the hip and blood samples were obtained from them. After confirming an effusion, dependent on patient status and clinical diagnosis, one of the following approaches was recommended; the close follow-up, or the ultrasound-guided aspiration of the hip joint effusion, and or arthrotomy. The various ultrasound and laboratory were documented. Data were analyzed and P < 0.001 being considered statistically significant. Results Overall, 115 patients with a mean age of 3.43 ± 5.76 years, 46 of whom were girls, were studied. The final diagnosis in 23 cases (20.0%) was septic arthritis and 92 (80.0%) had reactive arthritis. C-reactive protein (CRP) and The erythrocyte sedimentation rate (ESR) unlike aspirate volume, effusion volume measured on ultrasound, capsule thickness, total thickness, and recorded capsule-to-effusion ratio were significantly higher in patients with septic arthritis (P < 0.001). There was a significant agreement between the volume of measured fluid in the anterior recess and the volume of aspirated fluid (2.5 times, P < 0.001). Septic arthritis was not observed in any of the patients with effusion volume in anterior recess less than 0.5 cc and ESR less than 40 mm/hr or CRP less than 15 mg/L. Conclusion Since septic arthritis was not observed in any of the patients with effusion volume < 0.5 cc and normal inflammatory factors (ESR or CRP), conservative management and close follow-up can be recommended in these patients instead of joint fluid aspiration.
Pediatrics, Diseases of the musculoskeletal system