Using AI to Monitor Drug Adverse Events and Capture Causal Relations

Ontotext's highly configurable and flexible solution for scientific literature monitoring of adverse events efficiently identifies potential ADRs and captures causal relations between key structured ICSR data elements

  • Reduced time and effort on identifying and validating adverse events
  • Increased efficiency by providing access to vast amounts of textual data to identify potential adverse drug reaction signals
  • Support for compliance monitoring, adhering to both regulatory agencies and healthcare needs
  • Improved user experience throughout the entire drug development lifecycle

The Goal

Regulatory agencies such as the FDA and  European Medicinal Agency’s pharmacovigilance guidelines require Pharma companies to provide a strict ‘signal detection’ process for identifying new adverse events. Depending on the severity of the ADR, these companies also need to quickly assess such signals and promptly notify the regulatory bodies. 

For this purpose, Pharma companies require efficient solutions that can provide systematic and comprehensive approaches to monitoring adverse events in scientific literature where the information is far less structured than in ICSRs. Semantic technology can help automate this complex process, thus significantly improving the overall healthcare of the patients. 

The Challenge

The existing process of monitoring and screening for reported ADR in scientific literature has always been time and effort-consuming but generally fails to meet its goal. The main challenges are:

  • handling the exponentially growing safety information in diverse data sources and different formats that needs to be scanned 
  • full coverage of all important ADRs data elements
  • ambiguity of using generic and brand drug names interchangeably
  • capturing the complexity of the causal relations between different entities involved in a case report in a consistent way

The Solution: KG-based Scientific Literature Monitoring for Adverse Events

The solution for scientific literature monitoring of adverse events that Ontotext can provide is based on semantic text analysis and knowledge graphs. It is highly configurable and flexible, going beyond the standard deep indexing of content. 

The solution allows Pharma companies to automatically process any scientific literature data feed delivered in a standard format and loaded with an up-to-date subset of case reports abstracts from NCBI PubMed or other sources. The extracted data is complemented with structured ADR knowledge from a public Adverse Event Reporting system (such as FAERS from US FDA,  EudraVigilance, VigiBase from WHO). The information in it is normalized to established medical and drug terminologies like MedDRA, SNOMED CT, MeSH, and more.

On top of being able to efficiently identify potential adverse events mentioned in scientific literature, Ontotext’s solution can also capture causal relations between key structured ICSR data elements such as patient profile (gender, age), drug information (drug name, active ingredients, structured dosage information, co-treatment drug, route of administration), etc.

Building AI powered Drug Adverse Events Monitoring Service

Business Benefits

  • Utilization of publicly available safety information to build a comprehensive and up-to-date drug safety knowledge graph, assisting in causality assessment
  • Connecting knowledge locked in their proprietary safety data with large public repositories of semi-structured drug adverse reaction data
  • Tracing safety information throughout the entire drug development lifecycle in a single intelligent system
  • Automatically detecting new safety signals as new data is ingested in the knowledge graph
  • Decreased time and effort to identify and validate adverse events

Why Choose Ontotext?

With Ontotext’s scientific literature monitoring solution, using semantic text analysis, Pharma companies can scour vast amounts of textual data, such as medical records, and clinical notes, to identify potential ADR signals. This also supports compliance monitoring, adhering to both regulatory agencies and healthcare needs.  Additionally, users gain an integrated drug safety signals detection process that helps them identify and validate ADR signals timely and efficiently.

Do you think this case resembles your particular needs?

New call-to-action

 

Contact Us Now