84 research outputs found
Attention-Based Recurrent Neural Networks (RNNs) for Short Text Classification: An Application in Public Health Monitoring
No full textIn this paper, we propose an attention-based approach to short text classification, which we have created for the practical application of Twitter mining for public health monitoring. Our goal is to automatically filter Tweets which are relevant to the syndrome of asthma/difficulty breathing. We describe a bi-directional Recurrent Neural Network architecture with an attention layer (termed ABRNN) which allows the network to weigh words in a Tweet differently based on their perceived importance. We further distinguish between two variants of the ABRNN based on the Long Short Term Memory and Gated Recurrent Unit architectures respectively, termed the ABLSTM and ABGRU. We apply the ABLSTM and ABGRU, along with popular deep learning text classification models, to a Tweet relevance classification problem and compare their performances. We find that the ABLSTM outperforms the other models, achieving an accuracy of 0.906 and an F1-score of 0.710. The attention vectors computed as a by-product of our models were also found to be meaningful representations of the input Tweets. As such, the described models have the added utility of computing document embeddings which could be used for other tasks besides classification. To further validate the approach, we demonstrate the ABLSTM’s performance in the real world application of public health surveillance and compare the results with real-world syndromic surveillance data provided by Public Health England (PHE). A strong positive correlation was observed between the ABLSTM surveillance signal and the real-world asthma/difficulty breathing syndromic surveillance data. The ABLSTM is a useful tool for the task of public health surveillance
Limiting worker exposure to highly pathogenic avian influenza a (H5N1): a repeat survey at a rendering plant processing infected poultry carcasses in the UK
No full textAbstract Background Current occupational and public health guidance does not distinguish between rendering plant workers and cullers/poultry workers in terms of infection risk in their respective roles during highly pathogenic avian influenza poultry outbreaks. We describe an operational approach to human health risk assessment decision making at a large rendering plant processing poultry carcasses stemming from two separate highly pathogenic avian influenza A (H5N1) outbreaks in England during 2007. Methods During the first incident a uniform approach assigned equal exposure risk to all rendering workers in or near the production line. A task based exposure assessment approach was adopted during the second incident based on a hierarchy of occupational activities and potential for infection exposure. Workers assessed as being at risk of infection were offered personal protective equipment; pre-exposure antiviral prophylaxis; seasonal influenza immunisation; hygiene advice; and health monitoring. A repeat survey design was employed to compare the two risk assessment approaches, with allocation of antiviral prophylaxis as the main outcome variable. Results Task based exposure assessment during the second incident reduced the number of workers assessed at risk of infection from 72 to 55 (24% reduction) when compared to the first incident. No cases of influenza like illness were reported in workers during both incidents. Conclusions Task based exposure assessment informs a proportionate public health response in rendering plant workers during highly pathogenic avian influenza H5N1 outbreaks, and reduces reliance on extensive antiviral prophylaxis.</p
Evaluation of the national Notifiable Diseases Surveillance System for dengue fever in Taiwan, 2010-2012.
No full textBACKGROUND:In Taiwan, around 1,500 cases of dengue fever are reported annually and incidence has been increasing over time. A national web-based Notifiable Diseases Surveillance System (NDSS) has been in operation since 1997 to monitor incidence and trends and support case and outbreak management. We present the findings of an evaluation of the NDSS to ascertain the extent to which dengue fever surveillance objectives are being achieved. METHODOLOGY:We extracted the NDSS data on all laboratory-confirmed dengue fever cases reported during 1 January 2010 to 31 December 2012 to assess and describe key system attributes based on the Centers for Disease Control and Prevention surveillance evaluation guidelines. The system's structure and processes were delineated and operational staff interviewed using a semi-structured questionnaire. Crude and age-adjusted incidence rates were calculated and key demographic variables were summarised to describe reporting activity. Data completeness and validity were described across several variables. PRINCIPAL FINDINGS:Of 5,072 laboratory-confirmed dengue fever cases reported during 2010-2012, 4,740 (93%) were reported during July to December. The system was judged to be simple due to its minimal reporting steps. Data collected on key variables were correctly formatted and usable in > 90% of cases, demonstrating good data completeness and validity. The information collected was considered relevant by users with high acceptability. Adherence to guidelines for 24-hour reporting was 99%. Of 720 cases (14%) recorded as travel-related, 111 (15%) had an onset >14 days after return, highlighting the potential for misclassification. Information on hospitalization was missing for 22% of cases. The calculated PVP was 43%. CONCLUSIONS/SIGNIFICANCE:The NDSS for dengue fever surveillance is a robust, well maintained and acceptable system that supports the collection of complete and valid data needed to achieve the surveillance objectives. The simplicity of the system engenders compliance leading to timely and accurate reporting. Completeness of hospitalization information could be further improved to allow assessment of severity of illness. To minimize misclassification, an algorithm to accurately classify travel cases should be established
An Evolutionary Approach to Automatic Keyword Selection for Twitter Data Analysis
No full textIn this paper, we propose an approach to intelligent and automatic keyword selection for the purpose of Twitter data collection and analysis. The proposed approach makes use of a combination of deep learning and evolutionary computing. As some context for application, we present the proposed algorithm using the case study of public health surveillance over Twitter, which is a field with a lot of interest. We also describe an optimization objective function particular to the keyword selection problem, as well as metrics for evaluating Twitter keywords, namely: reach and tweet retreival power, on top of traditional metrics such as precision. In our experiments, our evolutionary computing approach achieved a tweet retreival power of 0.55, compared to 0.35 achieved by the baseline human approach
Use of antimicrobial resistance information and prescribing guidance for management of urinary tract infections: survey of general practitioners in the West Midlands
No full textSurvey questionnaire. (PDF 403 kb
Deep Learning for Relevance Filtering in Syndromic Surveillance: A Case Study in Asthma/Difficulty Breathing
No full textIn this paper, we investigate deep learning methods that may extract some word context for Twitter mining for syndromic surveillance. Most of the work on syndromic surveillance has been done on the flu or Influenza- Like Illnesses (ILIs). For this reason, we decided to look at a different but equally important syndrome, asthma/difficulty breathing, as this is quite topical given global concerns about the impact of air pollution. We also compare deep learning algorithms for the purpose of filtering Tweets relevant to our syndrome of interest, asthma/difficulty breathing. We make our comparisons using different variants of the F-measure as our evaluation metric because they allow us to emphasise recall over precision, which is important in the context of syndromic surveillance so that we do not lose relevant Tweets in the classification. We then apply our relevance filtering systems based on deep learning algorithms, to the task of syndromic surveillance and compare the results with real-world syndromic surveillance data provided by Public Health England (PHE).We find that the RNN performs best at relevance filtering but can also be slower than other architectures which is important for consideration in real-time application. We also found that the correlation between Twitter and the real-world asthma syndromic surveillance data was positive and improved with the use of the deep- learning-powered relevance filtering. Finally, the deep learning methods enabled us to gather context and word similarity information which we can use to fine tune the vocabulary we employ to extract relevant Tweets in the first place
A scoping review of the use of Twitter for public health research
No full textPublic health practitioners and researchers have used traditional medical databases to study and understand public health for a long time. Recently, social media data, particularly Twitter, has seen some use for public health purposes. Every large technological development in history has had an impact on the behaviour of society. The advent of the internet and social media is no different. Social media creates public streams of communication, and scientists are starting to understand that such data can provide some level of access into the people's opinions and situations. As such, this paper aims to review and synthesize the literature on Twitter applications for public health, highlighting current research and products in practice. A scoping review methodology was employed and four leading health, computer science and cross-disciplinary databases were searched. A total of 755 articles were retreived, 92 of which met the criteria for review. From the reviewed literature, six domains for the application of Twitter to public health were identified: (i) Surveillance; (ii) Event Detection; (iii) Pharmacovigilance; (iv) Forecasting; (v) Disease Tracking; and (vi) Geographic Identification. From our review, we were able to obtain a clear picture of the use of Twitter for public health. We gained insights into interesting observations such as how the popularity of different domains changed with time, the diseases and conditions studied and the different approaches to understanding each disease, which algorithms and techniques were popular with each domain, and more
Communicable disease control and health protection handbook
No full text"Sales handles: Clear and concise content that combines science with practical guidance. Covers basic principles of communicable disease control and health protection, major syndromes, control of individual infections, main services and activities, organizational arrangements for all EU countries and sources of further information. All chapters updated in line with recent changes in epidemiology, new guidelines for control and administrative changes. New disease chapters include Zika virus, Schistosomiasis, Coronavirus including MERS + SARS, and Ebola. Market description: Public-health physicians, epidemiologists, infection control nurses, microbiologists and those training to work in these related fields"...Provided by publishe
Number of confirmed cases of dengue fever reported to the National Disease Surveillance System by month of report, Taiwan, 2010–2012.
No full text<p>Number of confirmed cases of dengue fever reported to the National Disease Surveillance System by month of report, Taiwan, 2010–2012.</p
- …
