We have asked 7 of our customers to list which activities are the most cumbersome and repetitive in order to improve our service in automation.
Currently automation in Pharmacovigilance (PV) and drug safety can be seen from two different standpoints:
- Artificial Intelligence (AI): where a system interprets an input pattern and gives an output by performing tasks upon decision-making processes.
- Robotic Process Automation (RPA). Substituting repetitive manual steps with no need of human supervisioning or intervention.
PV and Drug Safety relies on a series of repetitive tasks that can be considered minor or major. Some are easy to monitor and automate, other ones are trickier.
1. Search Strategies
You need to be comprehensive and find all the relevant literature minimizing the risk of missing evidence.
An AI system can improve your recall and precision.
- Use services that include synonym expansion and related concepts: “Neurodegenerative disease” → “Neurodegeneration”.
- Connect to APIs related to PubMed for MeSH terms inclusion: “Alzheimer’s disease”→ MeSH added “Cognitive Dysfunction”.
- Correct syntax and expand it: “Alzheimer disease”→ “Alzheimer's disease”.
You’ll end up with more relevant data in less time.
2. Downloading articles PDF/meta
You now need to download to your laptop or shared space all the articles you deemed relevant from your previous step.
- Connect a script towards PubMed and PubMedCentral API.
- Retrieve links automatically by reconstructing them with PMID or DOI.
- Filter out duplicates and download each remaining link.
You won’t miss any reference while skipping obvious duplicates.
3. Monitoring Literature
You want to be notified when a new data point related to your research appears.
- Synch all your feeds and recommendations into a single email, as a single centralized data point.
- Connect programmatically to your email and download the reference from each feed directly.
You will automatically collect references from your email without needing to open it and check in.
4. Searching in multiple documents
In order to find information, you need to open several documents all at once and browse among them. This is not only cumbersome but quite annoying. As you won’t remember where you read which part and why.
- Use a document based search engine.
- Upload all your documents and start searching.
You can now search with “inclusion criteria”, exclude documents based on “exclusion criteria”, and let the search engine do the most of the work.
Time saved here varies upon quantity of data such as the amount of search strategies and how many articles you are supposed to download and screen. But on average by introducing AI and RPA in your processes, you’ll shrink a week of full-time work in a few hours while improving quality and quantity of data.
Do you think you could benefit from saving time in these tasks but don’t know where to start? Contact Us, we’ll be happy to set you up as fast as possible.
Everything starts with search.
With a smart suite of search tools to help you find the information you need, when you need it. Enhance your Search Experience with PapersHive Today!Contact Us