We address the assignment of ICD-10 codes for causes of death by analyzing free-text descriptions in death certificates, together with the associated autopsy reports and clinical bulletins, from the Portuguese Ministry of Health. We leverage a deep neural network that combines word embeddings, recurrent units, and neural attention, for the generation of intermediate representations of the textual contents. The neural network also explores the hierarchical nature of the input data, by building representations from the sequences of words within individual fields, which are then combined according to the sequences of fields that compose the inputs. Moreover, we explore innovative mechanisms for initializing the weights of the final nodes of the network, leveraging co-occurrences between classes together with the hierarchical structure of ICD-10. Experimental results attest to the contribution of the different neural network components. Our best model achieves accuracy scores over 89%, 81%, and 76%, respectively for ICD-10 chapters, blocks, and full-codes. Through examples, we also show that our method can produce interpretable results, useful for public health surveillance.
Este trabalho analisou a eficiência e a adequação dos Equipamentos de Proteção Individuais (EPIs) utilizados na manipulação e na aplicação de agrotóxicos nas agriculturas brasileira e francesa. Essa análise foi feita a partir de dois casos: o primeiro, na cultura de tomate numa pequena comunidade rural do Brasil e o segundo, numa vinicultura na França. As evidências encontradas mostraram que os EPIs utilizados em ambos os casos, além de não protegerem integralmente o trabalhador contra o agrotóxico, ainda agravaram os riscos e perigos, pois se tornaram fontes de contaminação. Conclui-se que, nos casos analisados, os EPIs não eliminaram nem neutralizaram a insalubridade, conforme estatui a legislação, e ainda aumentaram a probabilidade de contaminação dos trabalhadores rurais em algumas atividades. Discutiu-se, ainda, a possibilidade dos EPIs apresentarem lacunas funcionais no projeto, na concepção, no uso, na manutenção, no armazenamento e no descarte, que devem ser melhor avaliadas através de futuros estudos de desenvolvimento tecnológico.
OBJECTIVE To systematize and analyze the evidence from qualitative studies that address the perception of Brazilian Community Health Agents about their work.METHODS This is a systematic review of the meta-synthesis type on the work of community health agents, carried out from the Virtual Health Library using the descriptors “Agente Comunitário de Saúde” and “Trabalho”, in Portuguese. The strategy was constructed by crossing descriptors, using the Boolean operator “AND”, and filtering Brazilian articles, published from 2004 to 2014, which resulted in 129 identified articles. We removed quantitative or quanti-qualitative research articles, essays, debates, literature reviews, reports of experiences, and research that did not include Brazilian Community Health Agents as subjects. Using these criteria, we selected and analyzed 33 studies that allowed us to identify common subjects and differences between them, to group the main conclusions, to classify subjects, and to interpret the content.RESULTS The analysis resulted in three thematic units: characteristics of the work of community health agents, problems related to the work of community health agents, and positive aspects of the work of community health agents. On the characteristics, we could see that the work of the community health agents is permeated by the political and social dimensions of the health work with predominant use of light technologies. The main input is the knowledge that this professional obtains with the contact with families, which is developed with home visits. On the problems in the work of community health agents, we could identify the lack of limits in their attributions, poor conditions, obstacles in the relationship with the community and teams, weak professional training, and bureaucracy. The positive aspects we identified were the recognition of the work by families, resolution, bonding, work with peers, and work close to home.CONCLUSIONS This review provided an overview of the difficulties and positive aspects that are present in the daily work of community health agents. Given this, we have raised two challenges. The first one refers to how public policy makers need to appropriation the research results and the second one refers to the need to invest in studies that are designed to generate solutions for the difficulties faced by community health agents in their work.
Shigella spp. are a leading cause of human diarrheal disease worldwide, with Shigella flexneri being the most frequently isolated species in developing countries. This serogroup is presently classified into 19 serotypes worldwide. We report here a multicenter validation of a multiplex-PCR-based strategy previously developed by Q. Sun, R. Lan, Y. Wang, A. Zhao, et al. (J Clin Microbiol 49:3766 -3770, 2011) for molecular serotyping of S. flexneri. This study was performed by seven international laboratories, with a panel of 71 strains (researchers were blind to their identity) as well as 279 strains collected from each laboratory's own local culture collections. This collaborative work found a high extent of agreement among laboratories, calculated through interrater reliability (IRR) measures for the PCR test that proved its robustness. Agreement with the traditional method (serology) was also observed in all laboratories for 14 serotypes studied, while specific genetic events could be responsible for the discrepancies among methodologies in the other 5 serotypes, as determined by PCR product sequencing in most of the cases. This work provided an empirical framework that allowed the use of this molecular method to serotype S. flexneri and showed several advantages over the traditional method of serological typing. These advantages included overcoming the problem of availability of suitable antisera in testing laboratories as well as facilitating the analysis of multiple samples at the same time. The method is also less time-consuming for completion and easier to implement in routine laboratories. We recommend that this PCR be adopted, as it is a reliable diagnostic and characterization methodology that can be used globally for laboratory-based shigella surveillance.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.