Large expert-curated database for benchmarking document similarity detection in biomedical literature search

MedCPT: Contrastive Pre-trained Transformers with large-scale PubMed search logs for zero-shot biomedical information retrieval

MOTIVATION

Information retrieval (IR) is essential in biomedical knowledge acquisition and clinical decision support. While recent progress has shown that language model encoders perform better semantic retrieval, training such models requires abundant query-article annotations that are difficult to obtain in biomedicine. As a result, most biomedical IR systems only conduct lexical matching. In response, we introduce MedCPT, a first-of-its-kind Contrastively Pre-trained Transformer model for zero-shot semantic IR in biomedicine.

RESULTS

To train MedCPT, we collected an unprecedented scale of 255 million user click logs from PubMed. With such data, we use contrastive learning to train a pair of closely integrated retriever and re-ranker. Experimental results show that MedCPT sets new state-of-the-art performance on six biomedical IR tasks, outperforming various baselines including much larger models, such as GPT-3-sized cpt-text-XL. In addition, MedCPT also generates better biomedical article and sentence representations for semantic evaluations. As such, MedCPT can be readily applied to various real-world biomedical IR tasks.

AVAILABILITY AND IMPLEMENTATION

The MedCPT code and model are available at https://github.com/ncbi/MedCPT.

Epidemiological Impact of Metabolic Syndrome in Overweight and Obese European Children and Adolescents: A Systematic Literature Review

The prevalence of overweight and obesity is continuously increasing, both in the adult and pediatric populations, posing a substantial challenge to public health. Understanding the epidemiological burden of metabolic syndrome (MetS) among children, particularly regarding its complications and long-term effects in adulthood, is crucial for identifying effective preventive measures and enhancing the clinical care of obese children. Therefore, by searching two databases, a systematic review was conducted in order to evaluate studies that specifically addressed the epidemiological MetS impact among overweight/obese European children and adolescents. Overall, 15 studies were considered. The epidemiological data concerning the MetS impact were contingent on the diagnostic criteria used and varied across countries, resulting in a prevalence range of 1.44% to 55.8%. Spanish studies were the most numerous (34%), revealing a country prevalence rate ranging from 2.5% to 19.6%. Males (prevalence range: 1.4-55.8%) and subjects with overweight/obesity (prevalence range: 12.9-55.8%) were mainly affected. Obesity emerged as the main risk factor in the MetS development and the consequent onset of cardiovascular complications and diabetes. Knowing the MetS burden and its risk factors could improve their prevention, detection, and treatment, and guide the development of targeted public health interventions to appropriately address the health needs of younger patients.

Peristomal Skin Complications in Ileostomy and Colostomy Patients: What We Need to Know from a Public Health Perspective

BACKGROUND

Peristomal skin complications (PSCs) are the most common skin problems seen after ostomy surgery. They have a considerable impact on a patient's quality of life and contribute to a higher cost of care.

METHODS

A systematic review was conducted, querying three databases. The analysis was performed on international studies focused on the clinical-epidemiological burden of PSCs in adult patients with ileostomy/colostomy.

RESULTS

Overall, 23 studies were considered. The main diseases associated with ostomy surgery were rectal, colon and gynecological cancers, inflammatory bowel diseases, diverticulitis, bowel obstruction and intestinal perforation. Erythema, papules, skin erosions, ulcers and vesicles were the most common PSCs for patients with an ostomy (or stoma). A PSCs incidence ranging from 36.3% to 73.4% was described. Skin complications increased length of stay (LOS) and rates of readmission within 120 days of surgery.

CONCLUSIONS

PSCs data are still limited. A knowledge of their burden is essential to support health personnel and decision-makers in identifying the most appropriate responses to patients' needs. Proper management of these complications plays a fundamental role in improving the patient's quality of life. A multidisciplinary approach, as well as increased patient education and their empowerment, are priority measures to be implemented to foster a value-based healthcare.

Realization of Intelligent Scoring System of Taekwondo Protective Gear under the Application of Neural Network BP Model

This article presents the novel design of the technology acceptance model to create a predictive model of whether or not people will accept Taekwondo safety gear. A survey containing 28 items was completed by 200 collegiate Taekwondo practitioners associated with the Taekwondo Association. A significance level of 0.05 was employed for the correlation and structural equation modeling analyses. The perceived usefulness of the proposed system is practical compared to the existing system, which is significantly influenced by perceived quality. Perceived ease of use and perceived usefulness were also unaffected by visual beauty. Wearability had a substantial impact on perceived ease of use but significantly negatively impacted perceived usefulness. Perceived ease of use and perceived usefulness were not affected by the functionality of the proposed system. Perceived usefulness is significantly influenced by perceived ease of use, and the acceptance intention is affected by perceived usefulness which also affects the system's performance. These findings imply that increasing the device's perceived quality and wearability will increase its acceptance. This research shows an adequate verification model to validate the desired range of signals to accept Taekwondo electronic protective devices.

A comparative evaluation of biomedical similar article recommendation

BACKGROUND

Biomedical sciences, with their focus on human health and disease, have attracted unprecedented attention in the 21st century. The proliferation of biomedical sciences has also led to a large number of scientific articles being produced, which makes it difficult for biomedical researchers to find relevant articles and hinders the dissemination of valuable discoveries. To bridge this gap, the research community has initiated the article recommendation task, with the aim of recommending articles to biomedical researchers automatically based on their research interests. Over the past two decades, many recommendation methods have been developed. However, an algorithm-level comparison and rigorous evaluation of the most important methods on a shared dataset is still lacking.

METHOD

In this study, we first investigate 15 methods for automated article recommendation in the biomedical domain. We then conduct an empirical evaluation of the 15 methods, including six term-based methods, two word embedding methods, three sentence embedding methods, two document embedding methods, and two BERT-based methods. These methods are evaluated in two scenarios: article-oriented recommenders and user-oriented recommenders, with two publicly available datasets: TREC 2005 Genomics and RELISH, respectively.

RESULTS

Our experimental results show that the text representation models BERT and BioSenVec outperform many existing recommendation methods (e.g., BM25, PMRA, XPRC) and web-based recommendation systems (e.g., MScanner, MedlineRanker, BioReader) on both datasets regarding most of the evaluation metrics, and fine-tuning can improve the performance of the BERT-based methods.

CONCLUSIONS

Our comparison study is useful for researchers and practitioners in selecting the best modeling strategies for building article recommendation systems in the biomedical domain. The code and datasets are publicly available.

Stabilometric Biofeedback Training in Cognitive and Affective Function Improvement. Contribution of the Russian Scientific School. Part II

This review is the second part of the critical analysis of recent papers of Russian and other authors devoted to the study of the stabilometric parameters in postural control biofeedback training and rehabilitation, associated with psychological functions. The review presents the studies of postural control features in chronic pain syndrome, chronic fatigue syndrome, Parkinson’s disease, multiple sclerosis, and depression. The leading role of Russian researchers in the development and application of stabilometric biofeedback in the training of optimal functioning, rehabilitation, and correction of neurological disorders is noted. The paradigm of stabilometric biofeedback training of the cognitive and affective functions is offered.

Psychophysiological Indicators of Postural Control. Contribution of the Russian Scientific School. Part I

This article aimed to systematically review the published results of studies of psychophysiological mechanisms of posture maintenance and identify the key factors that influence the effectiveness of postural control. The recommendations of "Preferred Reporting Elements for Systematic Reviews and Meta-Analyzes" (PRISMA) were followed for the review. The results were classified, taking into account the target psychophysiological mechanisms and factors affecting postural control. The article presents the theoretical and empirical results of the Russian scientific school of research on the role of support afferentation in the sensorimotor mechanisms of cognitive and postural functions. Due to the limited number of randomized studies found, it was impossible to make meta-analytic comparisons, so the literature analysis was carried out only qualitatively. Meanwhile, our systematic review provides promising information about possible relationships between stabilometric and psychological indicators of postural control, which have theoretical significance and application in the correction and training of posture control. However, more thorough research is needed to overcome the methodological shortcomings that we have encountered in our qualitative analysis.

Real-World Data Regarding Long-Term Administration of Natalizumab from a Neurology Department along Literature Review

BACKGROUND

Natalizumab is a humanized monoclonal antibody with high efficacy and an acceptable safety profile used in the treatment of patients with multiple sclerosis (MS).

OBJECTIVES

Our aim was to report data regarding long-term administration of Natalizumab in patients with relapsing-remitting multiple sclerosis (RRMS) from our clinic.

METHODS

A retrospective observational study was performed including RRMS patients who underwent treatment with ≥ 24 Natalizumab infusions. We analyzed the EDSS values, the relapse rate and the rate and type of adverse events related to Natalizumab administration.

RESULTS

51 subjects were included with a predominance of women (62.74%), an average age of 40.43±1.49 years, a mean disease duration of 9.86±0.7 years and mean number of Natalizumab infusions of 45.58±2.74. An increased number of patients (80.39%) were relapse-free and there was observed a mild reduction of the mean EDSS value following Natalizumab initiation in patients who had not been treated with other disease modifying therapies anteriorly. Among the encountered adverse events we registered: increased liver transaminases (13.72%), local infections (7.84%) and dysmenorrhea in one patient. The rate of severe adverse events was 3.92 and there were registered no cases of Progressive Multifocal Leukoencephalopathy (PML).

CONCLUSIONS

Natalizumab proves to be effective, has an adequate safety profile and can be administered with good tolerability for a rather extended period of time, provided that the patients are closely monitored.