Held in conjunction with the XXXVII International Conference of the Spanish Society for Natural Language Processing (SEPLN2021)
[Sep 13]: schedule updated
[Sep 9]: the proceedings are now available
[Aug 31]: preliminary version of the schedule and talks was published
[Mai 7]: website of the IberHealth2021 workshop launched
Workshop will be held on Tuesday, September 21, 10.00-14:30 (CEST timezone)
Overview
One of the most active application domains of language technologies and NLP is health and biomedicine. This scenario has been accentuated even more due to current COVID-19 pandemic, where more efficient access, processing and exploitation of multiple types of unstructured data sources, including the biomedical literature, electronic healthcare records, clinical trials, patents, news articles or even social media is of critical importance. Not to mention other data sources like clinical practice guidelines, clinical trials documentation, medical questionnaires, medical informed consent documents, etc.). Nevertheless, despite the considerable progress made in the last two years, most of the current resources and research in clinical and biomedical NLP and text mining is still done mostly on data in English. Adapting and applying such tools to healthcare data in Spanish, as well as other languages like Portuguese or Catalan is not straightforward even under the increasing promise of multilingual NLP applications. Among the main driving forces of clinical NLP research and development one can highlight scientific venues in the form of workshops devoted specifically to this topic, such as BioCreative, BioNLP, AMIA NLP workshops, i2b2, (and its successor n2c2). Other scientific venues like LOUHI, HealTAC or eHealth CLEF, although theoretically being multi-lingual, did only marginally cover content in Spanish. Considering the number of native Spanish speakers and the large community of healthcare-related professionals it is key to provide a more stable and specific scientific environment to present and exchange NLP research results, resources and tools not only to NLP professionals but also to the industrial sector, pharma industry, healthcare professionals including medical informatics and healthcare technology experts. The objective of this workshop is to foster research both at the theoretical as well as at the level of practical real-world applications of NLP technologies applied to a diversity of textual data types. This workshop will also specially cover contributions from industry and companies working on healthcare textual data and results obtained by pharmaceutical industry.Finally, in addition to the healthcare NLP session we will also include a second session covering language technologies applied to the domain of food, nutrition and agriculture. Despite the economic importance of this sector for Spain and Latin-American countries, the potential of NLP applications was only marginally explored.Schedule (CEST timezone)
9/21/2021 10:00
Welcoming to the Iberian Health and Food Language Technologies and BSC/Plan TL resources
Martin Krallinger - Barcelona Supercomputing Center (BSC)
9/21/2021 10:30
Mining the Sociome for Insights on Chronic Conditions and Healthy Lifestyles
Anália Lourenço - Universidade de Vigo, Spain
9/21/2021 11:15
Food Nutrition and Security Cloud: A case usage of NLP technologies to extract food – drug interactions from scientific and clinical texts.
Enrique Carrillo de Santa Pau - IMDEA Food Institute, Madrid. Spain
9/21/2021 11:45
Use of NLP and Text Mining for health, nutrition, and food: Plan TL/BSC resources, components, corpora and use cases
Antonio Miranda/Eulalia Farre - Barcelona Supercomputing Center (BSC)
9/21/2021 12:15 coffee break
9/21/2021 12:30
Food Information Extraction and Normalization: the Past, the Present, and the Future
Tome Eftimov, Gjorgjina Cenikj - Jožef Stefan Institute, Ljubljana, Slovenia
9/21/2021 13:30
Text mining as a way to select microbial strains to ferment new food products
Claire Nédellec - University Paris-Saclay, INRAE
9/21/2021 14:00
EFSA's initiative for exchanging content and news on Data Science for Food Safety Risk Assessment
Carsten Behring - European Food Safety Authority
9/21/2021 14:15
Wrap-up final session
Martin Krallinger - Barcelona Supercomputing Center (BSC)
Invited Talks (Preliminary)
Natural language processing of healthcare text including EHRs, literature, patents, clinical trials, clinical practice guidelines, social media, health forums, or informed consents
Information extraction of clinical variables, named entities, values and clinical modifiers like negations
Speech analytics and voice processing for healthcare applications
Medical controlled vocabularies and ontologies
Machine-learning and transfer learning for healthcare text analytics
Information fusion, integration and semantic interoperability of NLP results with structured data or image data
Explainable models and interpretability of clinical NLP results
Real-world application and use cases of healthcare text analytics
Evaluation and benchmarking of clinical NLP methods
Knowledge discovery and hypothesis generation for health, agriculture, food or nutrition applications
Information fusion, i.e. integrating data from various sources, e.g. structured and narrative documentation
Anonymization of sensitive health data
NLP applied to agriculture, food (food poisoning, allergies, food-medication interactions food contamination, etc.…), nutrition
Multilingual health, agriculture or food NLP approaches (methods, applications, annotated data and terminologies).
Shared tasks and evaluation campaigns
Machine translation of specialized content and document types applied to health, agriculture and food/nutrition domains.
Martin Krallinger
Text Mining unit at the Barcelona Supercomputing Center (BSC), Spain
Jocelyn Dunstan
Center for Mathematical Modeling, Faculty of Physical and Mathematical Sciences, University of Chile
Francisco M. Couto
LASIGE, Departamento de Informática, Faculdade de Ciências, Universidade de Lisboa, Portugal
The planned workshop functions as a venue for the different types of contributors, mainly task providers and solution providers, to meet together and exchange their experiences.
We expect that investigation on the topics of the task will continue after the workshop, based on new insights obtained through discussions during the workshop.
As a venue to compile the results of the follow-up investigation, a journal special issue will be organized to be published a few months after the workshop. The specific journal will be announced after negotiation with publishers