PADI-web 3.0: A new framework for extracting and disseminating fine-grained information from the news for animal disease surveillance

Valentin, Sarah; Arsevska, Elena; Rabatel, Julien; Falala, Sylvain; Mercier, Alizé; Lancelot, Renaud; Roche, Mathieu

doi:10.1016/j.onehlt.2021.100357

Cited by 17 publications

(23 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…with Natural Language Processing (NLP) approaches. In 13 we summarized how the corpus described in this study is used to learn a fine-grained classification model based on a machine learning approach that is integrated into PADI-web 3.0. The proposed annotation scheme is intended to enhance EBS systems by enabling the automated classification of sentences from disease-related news.…”

Section: Methodsmentioning

confidence: 99%

Elaboration of a new framework for fine-grained epidemiological annotation

et al. 2022

View full text Add to dashboard Cite

Event-based surveillance (EBS) gathers information from a variety of data sources, including online news articles. Unlike the data from formal reporting, the EBS data are not structured, and their interpretation can overwhelm epidemic intelligence (EI) capacities in terms of available human resources. Therefore, diverse EBS systems that automatically process (all or part of) the acquired nonstructured data from online news articles have been developed. These EBS systems (e.g., GPHIN, HealthMap, MedISys, ProMED, PADI-web) can use annotated data to improve the surveillance systems. This paper describes a framework for the annotation of epidemiological information in animal disease-related news articles. We provide annotation guidelines that are generic and applicable to both animal and zoonotic infectious diseases, regardless of the pathogen involved or its mode of transmission (e.g., vector-borne, airborne, by contact). The framework relies on the successive annotation of all the sentences from a news article. The annotator evaluates the sentences in a specific epidemiological context, corresponding to the publication date of the news article.

show abstract

Section: Methodsmentioning

confidence: 99%

Elaboration of a new framework for fine-grained epidemiological annotation

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The uptake of EI systems, such as EIOS, was far from being generalised across all infectious diseases models. The generalisation of EI systems might be hampered by their limited performance for ltering noise and irrelevant items and validation of signals and analysis (24)., This requires time-consuming involvement of trained personnel, and leading agencies to favour the use of more trustworthy sources, such as scienti c literature, despite poor timeliness, and mainstream media, despite limited coverage of a range of online sources. For several disease models, the capture, ltering and validation of EBS signals were not part of the EI activities, raising concerns about countries capacities to early detect potential health threats, and the need to train PH and AH professionals in collection, analysis and interpretation of EBS signals, either through in-country initiatives, European capacity building, or within the development of the WHO's Pandemic Hub in the coming years (7,25,26).…”

Section: Event-based Activities: Data Sources and Processesmentioning

confidence: 99%

“…Our ndings showed that manual assessment of IBS and EBS signals i.e., expert review, was the main method used, emphasising the demand for continuous availability of trained personnel. The implementation of automatic or semi-automatic methods and tools (19,24,27), including forecasting or the use of data science and arti cial intelligence techniques for rapid processing of EI data could be of interest for better allocation of resources, to decrease the current workload of EI o cers and effectively support government agencies, healthcare service providers, and medical professionals in the future (28, 29).…”

Section: Signal Assessment and Communication Of Alertsmentioning

confidence: 99%

Epidemic intelligence activities among national public and animal health agencies: a European cross-sectional study

Dub¹,

Mäkelä²,

Kleef

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Epidemic Intelligence (EI) encompasses all activities related to early identification, verification, analysis, assessment, and investigation of health threats. It integrates an indicator-based (IBS) component using systematically collected surveillance data, and an event-based component (EBS), using non-official, non-verified, non-structured data from multiple sources. We described current EI practices in Europe by conducting a survey of national Public Health (PH) and Animal Health (AH) agencies. We included generic questions on the structure, mandate and scope of the institute, on the existence and coordination of EI activities, followed by a section where respondents provided a description of EI activities for three diseases out of seven disease models. Out of 81 gatekeeper agencies contacted, 34 agencies (42%) from 26 different countries responded, out of which, 32 conducted EI activities. Less than half (15/32; 47%) had teams dedicated to EI activities and 56% (18/34) had Standard Operating Procedures (SOPs) in place. On a national level, a combination of IBS and EBS was the most common data source. Most respondents monitored the epidemiological situation in bordering countries, the rest of Europe and the world. EI systems were heterogeneous across countries and diseases. National IBS activities strongly relied on mandatory laboratory-based surveillance systems. The collection, analysis and interpretation of IBS information was performed manually for most disease models. Depending on the disease, some respondents did not have any EBS activity. Most respondents conducted signal assessment manually through expert review. Cross-sectoral collaboration was heterogeneous. More than half of the responding institutes collaborated on various levels (data sharing, communication, etc.) with neighbouring countries and/or international structures, across most disease models. Our findings emphasise a notable engagement in EI activities across PH and AH institutes of Europe, but opportunities exist for better integration, standardisation, and automatization of these efforts. A strong reliance on traditional IBS and laboratory-based surveillance systems, emphasises the key role of in-country laboratories networks. EI activities may benefit particularly from investments in cross-border collaboration, the development of methods that can automatise signal assessment in both IBS and EBS data, as well as further investments in the collection of EBS data beyond scientific literature and mainstream media.

show abstract

“…The emergence and spread of any animal infectious disease, such as Avian Influenza, has serious consequences for animal health and a substantial socio-economic impact for agriculture. For instance, the 2021-2022 season have experienced the largest observed highly pathogenic avian influenza (HPAI) cases in Europe so far, with a total of 2,467 outbreaks in poultry, 3,573 HPAI events in wild birds, and 48 million birds culled in the affected establishments 2 . Due to this highly contagious nature, it is critical to monitor new and emergent infectious animal diseases.…”

Section: Introductionmentioning

confidence: 99%