[Paper Review] Global Health Monitor: A Web-based System for Detecting and Mapping Infectious Diseases

Son Doan, Hung Q. Ngo+2 more|arXiv (Cornell University)|Nov 21, 2019

Data-Driven Disease SurveillanceMedicine4 references35 citations

TL;DR

A web-based system that analyzes English news stories to detect infectious disease outbreaks, classifies relevance, and maps geo-temporal spread using the BioCaster ontology. It runs on a cluster monitoring 1500 news feeds 24/7 with hourly map updates.

ABSTRACT

We present the Global Health Monitor, an online Web-based system for detecting and mapping infectious disease outbreaks that appear in news stories. The system analyzes English news stories from news feed providers, classifies them for topical relevance and plots them onto a Google map using geo-coding information, helping public health workers to monitor the spread of diseases in a geo-temporal context. The background knowledge for the system is contained in the BioCaster ontology (BCO) (Collier et al., 2007a) which includes both information on infectious diseases as well as geographical locations with their latitudes/longitudes. The system consists of four main stages: topic classification, named entity recognition (NER), disease/location detection and visualization. Evaluation of the system shows that it achieved high accuracy on a gold standard corpus. The system is now in practical use. Running on a clustercomputer, it monitors more than 1500 news feeds 24/7, updating the map every hour.

Motivation & Objective

• Detect infectious disease outbreaks from English news stories.
• Classify news items for topical relevance to diseases and health events.
• Identify disease and location entities and geocode them.
• Visualize outbreaks on a geo-temporal map for public health monitoring.

Proposed method

• Utilizes topic classification, named entity recognition (NER), and disease/location detection as core pipeline stages.
• Uses the BioCaster ontology (BCO) as background knowledge for diseases and locations including coordinates.
• Plots detected outbreaks onto Google Maps using latitude/longitude geocoding.
• Operates on a 24/7 cluster to monitor over 1500 news feeds and refreshes the map hourly.
• Evaluates performance against a gold standard corpus to assess accuracy.

Experimental results

Research questions

• RQ1 Can news-derived signals be accurately classified as relevant infectious disease reports?
• RQ2 Can the system reliably extract disease and location entities and geocode them for mapping?
• RQ3 Does an ontology-driven approach (BCO) support effective detection and visualization of outbreaks?
• RQ4 Is the end-to-end system capable of real-time monitoring with frequent updates suitable for public health use?

Key findings

• System achieves high accuracy on a gold standard corpus.
• The platform monitors 1500 news feeds continuously and updates the map hourly.
• The approach combines topic classification, NER, and disease/location detection with geo-visualization.
• Background knowledge via BioCaster ontology supports disease and geographical entity recognition.
• System demonstrated practical use and deployment at IJCNLP 2008.
• Runs on a cluster computer to support real-time monitoring.