Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DrugWatch: A Comprehensive Multi-Source Data Visualisation Platform for Drug Safety Information

Artem Bobrov, Domantas Saltenis, Zhaoyue Sun, Gabriele Pergola, Yulan He

TL;DR

DrugWatch tackles the challenge of fragmented drug-safety information by integrating FAERS-derived statistics with PubMed case reports and private data annotation through NLP-driven event extraction. It comprises two sub-platforms: DrugWatch Search for interactive visualization and literature retrieval, and DrugWatch Annotate for live and bulk annotation of user data. The architecture combines a ReactJS frontend, a SvelteKit frontend for annotation, and a Flask backend, with a PubMed integration pipeline and multiple annotation models (Flan-T5, UIE, Mistral-7B) and a PHEE-based evaluation dataset. Early evaluations of ADE classification and extraction demonstrate competitive performance, and DrugWatch aims to streamline pharmacovigilance workflows while preserving user privacy and planning future data-source expansion and literature-summarization capabilities.

Abstract

Drug safety research is crucial for maintaining public health, often requiring comprehensive data support. However, the resources currently available to the public are limited and fail to provide a comprehensive understanding of the relationship between drugs and their side effects. This paper introduces DrugWatch, an easy-to-use and interactive multi-source information visualisation platform for drug safety study. It allows users to understand common side effects of drugs and their statistical information, flexibly retrieve relevant medical reports, or annotate their own medical texts with our automated annotation tool. Supported by NLP technology and enriched with interactive visual components, we are committed to providing researchers and practitioners with a one-stop information analysis, retrieval, and annotation service. The demonstration video is available at https://www.youtube.com/watch?v=RTqDgxzETjw. We also deployed an online demonstration system at https://drugwatch.net/.

DrugWatch: A Comprehensive Multi-Source Data Visualisation Platform for Drug Safety Information

TL;DR

DrugWatch tackles the challenge of fragmented drug-safety information by integrating FAERS-derived statistics with PubMed case reports and private data annotation through NLP-driven event extraction. It comprises two sub-platforms: DrugWatch Search for interactive visualization and literature retrieval, and DrugWatch Annotate for live and bulk annotation of user data. The architecture combines a ReactJS frontend, a SvelteKit frontend for annotation, and a Flask backend, with a PubMed integration pipeline and multiple annotation models (Flan-T5, UIE, Mistral-7B) and a PHEE-based evaluation dataset. Early evaluations of ADE classification and extraction demonstrate competitive performance, and DrugWatch aims to streamline pharmacovigilance workflows while preserving user privacy and planning future data-source expansion and literature-summarization capabilities.

Abstract

Drug safety research is crucial for maintaining public health, often requiring comprehensive data support. However, the resources currently available to the public are limited and fail to provide a comprehensive understanding of the relationship between drugs and their side effects. This paper introduces DrugWatch, an easy-to-use and interactive multi-source information visualisation platform for drug safety study. It allows users to understand common side effects of drugs and their statistical information, flexibly retrieve relevant medical reports, or annotate their own medical texts with our automated annotation tool. Supported by NLP technology and enriched with interactive visual components, we are committed to providing researchers and practitioners with a one-stop information analysis, retrieval, and annotation service. The demonstration video is available at https://www.youtube.com/watch?v=RTqDgxzETjw. We also deployed an online demonstration system at https://drugwatch.net/.
Paper Structure (25 sections, 14 figures, 2 tables)

This paper contains 25 sections, 14 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Architecture of DrugWatch
  3. User Interaction Design
  4. DrugWatch Search Sub-platform
  5. Flexible Search
  6. Drug Information Display
  7. Statistical Information Display
  8. Pubmed Case Report Retrieval
  9. DrugWatch Annotate Sub-platform
  10. Live Annotation
  11. Bulk Annotation
  12. Backend Implementation
  13. Data Storage and Retrieval
  14. Relevant Medical Case Reports Integration
  15. Case Reports Collection
  16. ...and 10 more sections

Figures (14)

  • Figure 1: The overall architecture of DrugWatch.
  • Figure 2: Top frequent side effects related to Acetaminophen, presented by bar chart and word cloud.
  • Figure 3: A breakdown of top side effects for each age group when searching for reports of Acetaminophen in males.
  • Figure 4: Retrieved PubMed case reports related to "Acetaminophen" and "Liver failure".
  • Figure 5: Illustration of DrugWatch Annotate model annotation result comparison interface.
  • ...and 9 more figures