Semi-Automated Data Extraction: Dextr and Laser AI

Title:  Evaluation of a semi-automated data extraction tool for public health literature-based reviews: Dextr

Date Published: January 2022

Environment International: Peer-Reviewed Publication

Citation: Walker VR, Schmitt CP, Wolfe MS, Nowak AJ, Kulesza K, Williams AR, Shin R, Cohen J, Burch D, Stout MD, Shipkowski KA, Rooney AA. Evaluation of a semi-automated data extraction tool for public health literature-based reviews: Dextr. Environ Int. 2022 Jan 15;159:107025. doi: 10.1016/j.envint.2021.107025. Epub 2021 Dec 14. PMID: 34920276.

‍What is this report about?

The aim was to evaluate a web-based tool that semi-automates data extraction based on three specific goals.   

‍How did Laser AI help?

Laser AI, previously known as Dextr, was chosen as the web-based tool to help automate data extraction in animal studies. Artificial intelligence within Laser AI offered the authors a unique tool to extract and connect complex study elements while testing its efficacy.

‍What were the results?

Dextr reduces the average extraction time (from 933 seconds to 436 seconds per study), while the precision and recall rate remain similar to a manual extraction workflow (96.0% vs. 95.4% and 91.8% vs. 97.0%, respectively).