Our NLP research has been published in Gates Open Research!

Clinical trials are the backbone of medical progress, but a worrying trend is emerging: a large portion end without delivering useful results. This “uninformativeness” wastes valuable resources and delays advancements.

Fast Data Science is excited to announce the publication of a technical research paper the Clinical Trial Risk Tool, a game-changer in identifying potential uninformativeness at the protocol stage!

• Wood TA and McNair D. Clinical Trial Risk Tool: software application using natural language processing to identify the risk of trial uninformativeness [version 1; peer review: 1 approved with reservations]. Gates Open Res 2023, 7:56 doi.org/10.12688/gatesopenres.14416.1

The Clinical Trial Risk Tool is a browser-based tool which uses Natural Language Processing (NLP) to analyse clinical trial protocols. Here’s how it works:

• Drag and drop your PDF protocol: The tool reads and parses the text, automatically identifying key features of the trial design.
• Risk assessment: These features are fed into a risk model, pinpointing areas that might lead to uninformative results.
• User-friendly interface: Visualize the risk indicators and their locations directly in the text. You can even correct any parsing inaccuracies.
• Detailed report: Get a PDF report summarizing the extracted key features and potential risks.

Ready to fight uninformativeness? Head over to app.clinicaltrialrisk.org and access this free, open-source software.

A BibTex citation is as follows:

@article{wood2023clinical,
  title={Clinical Trial Risk Tool: software application using natural language processing to identify the risk of trial uninformativeness},
  author={Wood, Thomas A and McNair, Douglas},
  journal={Gates Open Research},
  volume={7},
  number={56},
  pages={56},
  year={2023},
  publisher={F1000 Research Limited}
}

Read more about research in AI and Fast Data Science’s publications here