The medical literature contains a wealth of valuable medical knowledge. At present, the research on extraction of entity relationship in medical literature has made great progress, but with the exponential increase in the number of medical literature, the annotation of medical text has become a big problem. In order to solve the problem of manual annotation time such as consuming and heavy workload, a remote monitoring annotation method is proposed, but this method will introduce a lot of noise. In this paper, a novel neural network structure based on convolutional neural network is proposed, which can solve a large number of noise problems. The model can use the multi-window convolutional neural network to automatically extract sentence features. After the sentence vectors are obtained, the sentences that are effective to the real relationship are selected through the attention mechanism. In particular, an entity type (ET) embedding method is proposed for relationship classification by adding entity type characteristics. The attention mechanism at sentence level is proposed for relation extraction in allusion to the unavoidable labeling errors in training texts. We conducted an experiment using 968 medical references on diabetes, and the results showed that compared with the baseline model, the present model achieved good results in the medical literature, and F1-score reached 93.15%. Finally, the extracted 11 types of relationships were stored as triples, and these triples were used to create a medical map of complex relationships with 33 347 nodes and 43 686 relationship edges. Experimental results show that the algorithm used in this paper is superior to the optimal reference system for relationship extraction.
Testing Treatments is a book to help the public understand how to validate the efficacy of testing treatments and the possible bias and error in clinical trial, as well as to call for help to promote good study thus to improve the quality of health care. No matter for the first or the second edition, this book is very popular around the world, and its second edition has been translated into more than ten languages. To help the readers understand the content of the book, we established a website (www.testingtreatments.org) and other sibling sites in different languages. The website not only provided the full-texts to download, but also collected various popular science resources (videos, audios and cartoons) to help the readers assimilate more knowledge. The editors of all the different language websites have established an TTi Editorial Alliance to share experience and provide each other with mutual support, thus to promote health professionals, patients and public around the world to use reliable research to inform their health decisions.
ObjectiveBased on the method of bibliometrics, to visually analyze the current research status of DNA polymerase epsilon catalytic/DNA polymerase delta catalytic subunit (POLE/POLD1) mutations in colorectal cancer, and further to explore its current hot spots and to look forward to future development trends.MethodsUsed the Web of Science database as the data source to retrieve the published related literatures in the 30 years from the establishment of the database to September 2020, and used VOSviewer 1.6.11 and CiteSpace 5.7 software to analyze the the distribution of research countries/institutions, authors and keywords in the included published literatures. And to perform cluster analysis, time evolution analysis and Burst analysis on keywords, draw corresponding visual maps and analyzed them.ResultsA total of 299 published articles were included. Bibliometric analysis showed that the application of POLE/POLD1 mutations in colorectal cancer had developed steadily in recent years. After 2012, the number of published studies had increased significantly; the journals with the most articles were Familial Cancer and Proceedings of the National Academy of Sciences of the United States of America. The author with the highest number of articles was Tomlinson I. The analysis of countries and regions showed that the United States was the main country for research in this field and had contributed the most to the amount of publications. After analyzed the academic institutions/universities where the researchers were located, it could be seen that the University of Oxford was the research university/institution with the most publications. Keyword cluster analysis and time evolution analysis showed that the research hotspots in this field focus on the impact of this gene mutation on immunotherapy. Burst analysis showed that the current published research in this field started two years after publication and continues to be highly cited.ConclusionsPOLE/POLD1 mutation has undoubtedly become a hot area of current research in colorectal cancer research. Among them, tumor immunotherapy is the focus of research in this field.
ObjectiveTo analyze the status and influencing factors of awareness of diabetic retinopathy (DR) knowledge in diabetics in Funing County, Jiangsu Province. MethodsThis is a cross-sectional study. In 2021, a total of 2 125 residents with diabetes in Funing County, Jiangsu Province were recruited using cluster random sampling methods. General examinations were performed and ophthalmological evaluation were conducted including visual acuity in daily life, slit lamp microscope with preset lens, digital non-mydriatic fundus photography. Clinical assessment and classification of DR were according to the international standard clinical classifications of diabetic retinopathy in 2019. DR was classified into no obvious DR, mild non-proliferative diabetic retinopathy (NPDR), moderate NPDR, severe NPDR and proliferative diabetic retinopathy (PDR). Field questionnaire surveys were conducted to show the awareness rate of DR knowledge. Then Chi-square test and Logistic regression analysis were used to investigate influencing factors. ResultsAmong 2 125 subjects, 762 (35.86%, 762/2 125) residents were male and 1 363 (64.14%, 1 363/2 125) residents were female. A total of 444 (20.89%, 444/2 125) subjects were diagnosed as DR. Among which, 438 (20.61%, 438/2 125) subjects were NPDR, 6 (0.28%, 6/2 125) subjects were PDR. A total of 419 (19.72%, 419/2 125) subjects were diagnosed as cataract. The topic awareness rates were all below 45.00%. The overall population awareness rate was 19.45% (4 959/25 500). Awareness rate was correlated with age, visual acuity in daily life, education level, family history of diabetes, diabetic retinopathy grades and the type of treatment for diabetes (P＜0.001). Logistic regression analysis showed that the knowledge rate of prevention and treatment of DR Decreased significantly in patients aged 60-79 compared with those aged under 50 (P＜0.05). Compared with those with visual acuity＜0.1 in daily life, the knowledge rate of DR prevention and treatment in those with visual acuity≥0.1 was significantly decreased (P＜0.05). While primary school to college education, with family history of diabetes, with DR, oral hypoglycemic agents and/or insulin treatment were associated with higher awareness rate (P＜0.05). ConclusionsThe DR knowledge level is low among diabetics in Funing County, Jiangsu Province in 2021. Age, visual acuity in daily life, education level, family history of diabetes, with DR and receiving drug treatments are main influencing factors for diabetic’s awareness.
ObjectiveTo analyze the hotspots and frontiers of oral lichen planus research by bibliometric methods.MethodsWe searched Web of Science Core Collection database to obtain studies on oral lichen planus from inception to January 1st, 2020. After data extraction, Excel 2016 and CiteSpace software were used to carry out descriptive and visual analysis.ResultsA total of 3 105 articles and reviews were included, and the annual publication volume showed a steady growth trend. The research hotspot terms of oral lichen planus were cancer, lesion, and management of the disease. Moreover, pathogenesis, potentially malignant disorder, classification, and diagnosis were defined as novel research frontiers.ConclusionsThrough the bibliometric method, the research hotspots and frontiers of oral lichen planus are displayed intuitively, which provides references for future research.
This article illustrates the development, current status and future prospects of knowledge translation. Its importance and necessity are introduced and some measurements or approaches to promote knowledge translation are discussed.
Methods To explore the level of delirium knowledge of geriatric nurses in Sichuan province and analyze the factors, so as to provide the basis for systematic and targeted knowledge training on delirium and clinical management. Methods Using the self-designed “the Questionnaire of Elderly Delirium Knowledge”, geriatric nurses from 22 hospitals in Sichuan province were investigated through a convenient sampling method from September 2018 to February 2019. Results A total of 475 geriatric nurses were investigated. The average delirium knowledge score of the 475 geriatric nurses was 69.51±12.42. Multiple linear regression analysis showed that the main factors affecting the score of delirium-related knowledge were the education of nurses (P=0.037), technical title (P<0.001), years of working in the geriatric department (P=0.001), and the level of working hospital (P=0.001). Conclusions The level of delirium knowledge of geriatric nurses is low and can not meet the needs of clinical work. Nursing managers should carry out delirium knowledge training according to the different characteristics of nurses.
目的 了解社区人群对脑卒中危险因素及症状的知晓现状。 方法 随机抽样调查1 208名居民及112名医务人员的人口学特征、对危险因素及预警信号的知晓现状及其影响因素、信息来源及需求情况。 结果 90.1%、100.0%的居民及医务人员可辨识高血压是卒中的危险因素，而社区居民及医务人员对年龄、糖尿病、吸烟等危险因素的认识比例分别为65.0% 和85.0%，且对危险因素的控制策略缺乏了解。87.4%、100.0%社区居民及医务人员将一侧肢体的活动障碍作为卒中的第一大预警信号，其次为头晕、步态不稳、头痛、言语困难、视物模糊。卒中知识得分的单因素及多因素分析示：大学文化、已婚且在职的居民及高学历的医务人员对卒中信息的了解程度高。卒中信息的获取途径依次为电视、社区医生/讲堂、报纸、杂志、网络。 结论 西部城市社区人群卒中知识知晓率低，开展针对低学历医务者的卒中培训及低学历、独居及退休人员的居民讲堂是改善现状的必要途径。同时也为政府建立有效的院前早期识别及快速转诊技术提供了依据。Objective To assess the baseline knowledge of risk factors and warning signs of cerebral apoplexy in communities. Methods A total of 1 208 inhabitants and 112 medical personnel were selected by systematic sampling. The questionnaire included social-demographic data, knowledge of cerebral apoplexy risk and warning signs and influencing factors, the sources and requirement of information about cerebral apoplexy. Results Hypertension was a risk factor in 90.1% of residents and 100.0% of medical personnel. Age, diabetes and smoking were identified as the risk factor in 65.0% of medical personnel and 85.0% of residents. Medical therapy of risk factors was insufficient. The most common warning signs of cerebral apoplexy was hemiplegia, the following were vertigo, ataxia, headache, aphasia and double vision. Stepwise multiple regression analyses showed that residents who had higher educational background, spousal, workers and community worker with higher educational background had higher knowledge scores of cerebral apoplexy. The main sources of information about cerebral apoplexy were television, doctors, newspaper, magazine and network. Conclusions At present, the urban community residents in west China are lacking in knowledge about cerebral apoplexy. Going forward, targeted educational residents and medical workers should be directed at those who was highly educated, living alone, and retired. It is also provide a theoretical basis for establishing a prehospital identification and transfer treatment system based on community in developing countries.
Objective To learn and evaluate the effect of health education of schistosomiasis in the Three Gorges Reservoir Area, so as to provide scientific basis for establishing the comprehensive prevention and control model in potential epidemic area of schistosomiasis. Methods Through adopting the multi-stage stratified cluster sampling method, the questionnaires were distributed to residents selected randomly from the demonstration, inside and outside control areas. The survey data were input with EpiData 3.0 software by two reviewers, and the differences among groups were analyzed by using SPSS 18.0 software to further evaluate the effect of health education of schistosomiasis. Results All the questionnaires distributed to 1 420 residents were retrieved (100%). The results of analysis showed that for the residents in the demonstration area, the average score of being aware of schistosomiasis prevention was (7.51±2.89), which was markedly higher than those in the inside (2.52±2.97) and the outside (3.13±3.51) control areas, with significant differences (Plt;0.05); and the passing rates were 78.15%, which was also obviously higher than those in the inside (17.26%) and the outside control areas (32.16%), with significant differences (Plt;0.05). Conclusion The health education of schistosomiasis can improve residents’ knowledge about schistosomiasis prevention, and it is very important to prevent and control the potential prevalence of schistosomiasis in the Three Gorges Area.
ObjectiveTo propose a lung artery segmentation method that integrates shape and position prior knowledge, aiming to solve the issues of inaccurate segmentation caused by the high similarity and small size differences between the lung arteries and surrounding tissues in CT images. MethodsBased on the three-dimensional U-Net network architecture and relying on the PARSE 2022 database image data, shape and position prior knowledge was introduced to design feature extraction and fusion strategies to enhance the ability of lung artery segmentation. The data of the patients were divided into three groups: a training set, a validation set, and a test set. The performance metrics for evaluating the model included Dice Similarity Coefficient (DSC), sensitivity, accuracy, and Hausdorff distance (HD95). ResultsThe study included lung artery imaging data from 203 patients, including 100 patients in the training set, 30 patients in the validation set, and 73 patients in the test set. Through the backbone network, a rough segmentation of the lung arteries was performed to obtain a complete vascular structure; the branch network integrating shape and position information was used to extract features of small pulmonary arteries, reducing interference from the pulmonary artery trunk and left and right pulmonary arteries. Experimental results showed that the segmentation model based on shape and position prior knowledge had a higher DSC (82.81%±3.20% vs. 80.47%±3.17% vs. 80.36%±3.43%), sensitivity (85.30%±8.04% vs. 80.95%±6.89% vs. 82.82%±7.29%), and accuracy (81.63%±7.53% vs. 81.19%±8.35% vs. 79.36%±8.98%) compared to traditional three-dimensional U-Net and V-Net methods. HD95 could reach (9.52±4.29) mm, which was 6.05 mm shorter than traditional methods, showing excellent performance in segmentation boundaries. ConclusionThe lung artery segmentation method based on shape and position prior knowledge can achieve precise segmentation of lung artery vessels and has potential application value in tasks such as bronchoscopy or percutaneous puncture surgery navigation.