COVID-19 RESOURCES
by QUINTELLIGENCE

   
Since late 2016, Quintelligence initiated efforts to refocus and develop text mining and data analytics technologies aiming to provide meaningful ICT tools to decision-making over Big Data in the context of Public Health and Healthcare. The core data sources utilized by these tools are (i) worldwide news articles, through the Event Registry engine; and (ii) the biomedical research over scientific articles in MEDLINE/PubMed.

During the pandemic in early 2020, we have refocused some tools developed by our partners at the Jozef Stefan Institute, and build others to address problems as a contribution to the global fight against COVID-19 (read more about it). This work is in line with the global effort of WHO and the recent release of the news monitoring dashboard originating from the initiative of the UNESCO Research Centre for Artificial Intelligence (IRCAI). 

These tools allow the users to explore the worldwide media, configuring the topic to be monitored over a real-time news stream over 100 thousand news articles daily. Side-by-side it enables health professionals to explore the biomedical research open dataset MEDLINE, that feeds the well-established and worldwide adopted medical science search engine PubMed. 








qmidas medline explorer


 MEDLINE EXPLORER 

This tool is based on the SearchPoint technology and exhibits the clustered keywords of a query on the MEDLINE/PubMed open dataset, after searching for a keyword. This int
eractive visual tool helps to surface information we are looking for, avoiding the standard answer that is biased by definition. SearchPoint can be used in any language with any document set (public or private) that can be indexed, analyzed and visualized with this approach.
  • an interactive exploratory tool to change the often biased priority of MEDLINE search results
  • exhibits a word-cloud representing the k-means clusters of topics in the articles that include the searched keywords of a query
  • An interactive pointer that can be moved through the word-cloud and that will change the priority of the listed articles 
  • designed to improve the search engine experience; the user provides further information to the search by interacting with the system by dragging a pointer over word clouds
  • these word clouds are produced by cosine similarity to an "average" centred on the topics in each abstract of the set of selected papers, clustered using the k-means algorithm
  • available through a web app or an API to integrate into any system
  • evaluated based on annotations by health professionals with experience 
The technology is published Open Source under a BSD license.  


 Repository   https://github.com/quintelligence-health/medline-widget              Demonstrator   https://qmidas.quintelligence.com/searchpoint




 MESH CLASSIFIER  

qmidas mesh classifier
This t
ool developed aims to classify free text with the latest MeSH Headings provided by NHS. It is based on the DMOZ classifier, learning over 80+ years of MEDLINE data, and over the MeSH tree with 16 major categories and a max of 13 levels of deepness. It provides all the classifying categories with position number and (cosine) similarity weight, with a slider and a number of max categories visible. It available through a web app and an API. 

  • designed to classify free text of any nature with the classes of MeSH Headings where MEDLINE is based on, to which health professionals are familiar with  
  • it can classify articles that haven't yet been annotated by the NHS, or official WHO documents of interest
  • it can also classify news articles and be used to monitor worldwide news based on the classification provided at the MeSH Headings  
  • based on the DMOZ classifier, learning over 80+ years of MEDLINE data, and over the MeSH tree with 16 major categories and a max of 13 levels of deepness. 
  • provides all the classifying categories with position number and (cosine) similarity weight, with a slider and a number of max categories visible
  • available through a web app or an API to integrate into any system
  • evaluated based on annotations by health professionals with experience 

The technology is published Open Source under an Apache 2.0 license.  


 Repository   https://github.com/quintelligence-health/medline_classifier              Demonstrator   https://qmidas.quintelligence.com/classify-mesh-major/
.




 MEDLINE DASHBOARD 
qmidas medline dashboard


The Kibana dashboard permits the user to profit of data visualization modules that feed on his/her datasets built in ElasticSearch. This tool enables one to query the dataset and produce different types of data visualization modules that can later integrate a customized dashboard. Kibana can be used in any language with any document set (public or private) that can be indexed, analyzed and v
isualized with this approach.

  • designed to improve the user experience in exploring the MEDLINE dataset through visualization modules composing topic dedicated live monitoring dashboards 
  • the visualization modules based on Kibana do not require technical skills and enable the data exploration by a diversity of professionals
  • the system includes a powerful querying engine based on the open-source information retrieval software library Lucene 
  • enables to query large datasets – MEDLINE – and produce different types of visualisation modules that can be later integrated into customized dashboards 
  • feeds on that dataset through the elasticSearch index, permitting powerful queries, based on Lucene.
  • with 3 levels of access/expertise: (i) the management dashboard; (ii) the visual modules creator; and (iii) the data visualiser enabling the user to see the structure of the dataset
The technology is partially published Open Source under an Apache 2.0 license.  


 Repository   https://github.com/quintelligence-health/medline-dashboard              Demonstrator   ONLY through the MIDAS platform
.




qmidas news widget
 NEWS EXPLORER 


This tool allows the user to monitor news topics configured to the specificities that most relate to their public health priorities. In that, they can focus specific locations, subtopics, languages, news categories, etc. 

  • Based on a real-time stream of news including PRs and blogs
  • Allows for the control of the specificities of the data stream, the subtopics, the locations, the categories, etc.
  • Includes a tag cloud that allows for a fast perspective on the main news topics related to the defined choices
  • Monitor specific relations in the media as, e.g., COVID-19 and diabetes
  •   
The technology is published Open Source under an Apache 2.0 license.  


 Repository   https://github.com/quintelligence-health/news-widget              Demonstrator   ONLY through the MIDAS platform







 NEWS DASHBOARD 
qmidas news explorer


This dashboard is based on the Event Registry technology, allowing the user to explore different aspects of the worldwide and local news. It offers several visualisation modules that can help detecting insights or analyse the newstream to better avoid bias and fake news.

  • Provides more than 100k news daily and related  categories of news 
  • Search over 60+ languages using a powerful cross-lingual technology
  • Allows for the exploration of the main entities and trends of a subset based on any query 
  • Includes the exploration of news over a detailed timeline of events and retrieve related results
  • Permits the user to visualise the sentiment over a selection of news
  •   
The technology is only available under a commercial license.  


 Repository   Not available              Demonstrator   http://eventregistry.org/






 INFLUENZANET HUBS 

k

This tool serves the Influenzanet partners to explore their own data, based on the elasticsearch technology and on Kibana dashboards. The influenzanet system has been active throughout Europe for more than 15 years, allowing to monitor the behaviour of Influenza-Like-Ilness with the help of online volunteers that fill-in a profile and weekly questionnaires. It will be extended to help monitor COVID-19 throughout Europe. This tool enables one to query the dataset based on the Lucene syntax, and produce different types of data visualization modules that can later integrate a customized dashboard. Kibana can be used in any language with any document set (public or private) that can be indexed, analyzed and visualized with this approach.

  • designed to improve the user experience in exploring the Influenzanet dataset through visualization modules composing topic dedicated live monitoring dashboards 
  • the visualization modules based on Kibana do not require technical skills and enable the data exploration by a diversity of professionals
  • the system includes a powerful querying engine based on the open-source information retrieval software library Lucene   

The technology is published Open Source under an Apache 2.0 license.  


 Repository   https://github.com/quintelligence-health/news-widget              Demonstrator   ONLY through the Influenzanet platform






 TOPOLOGICAL DATA ANALYSIS 
qmidas TDA


This research is based on the data from Influenzanet and is being applied to understand the behaviour of COVID-19 throughout Europe. It is based on the Topological Data Analysis methodology, which examines the structure of data through the topological structure of data. It allows comparing the evolution of the epidemics through the encoded topology of their incidence time series.

  • designed to encode the topology of time-series to allow easy comparison on topology 
  • brings the incidence time-series to a higher dimension point cloud using a time window that runs through the series
  • computes the persistence diagrams out of the simplicial complex build over the point cloud  

The related technology is published Open Source and the results of the research are state-of-the-art, published throughout several scientific publications.  







 STREAMSTORY 
qmidas streamstory


This tool allows for the analysis of dynamical Markov processes, analysing simultaneous time-series through transitions between states. It offers several customisation options and data visualisation modules. The approach using this methodology and technology was developed to analyse the Influenzanet data and is being applied to analyse COVID-19. 

  • can compare the evolution of the disease between countries by comparing their time-series of incidence
  • contrasts the incidence of the disease with weather conditions and other impact factors
  • can analyse the dynamics of the evolution of the disease based on incidence, morbidity and recovery     

The related technology is published Open Source and the results of the research are state-of-the-art, published throughout several scientific publications. 



Disclaimer: The core system was developed by the AI Lab at the IJS and refocused by Quintelligence within the MIDAS project to analyze the MEDLINE dataset. It can be implemented in premises to work with proprietary data. It is currently available as Open Source under the BSD license.


Contributors

F. Fuart - J. Costa - L. Stopar - M. Grobelnik - D. Mladenić - M. Karlovac - M. Jermol - G. Leban - E. Novak - I. Novalija - P. Škraba - B. Fortuna - L. Rei - A. Košmerlj - J. Belyaeva


logos



contacts






3