RobotReviewer provides data on the trial PICO characteristics (Population, Interventions/Comparators, and Outcomes), and also automatically assesses trials for likely biases using the Cochrane Risk of Bias tool.
RobotReviewer, a third-party application property of Byron Wallace, Iain Marshall, Joël Kuiper, and Frank Soboczenski.
RobotReviewer, an AI tool, uses natural language processing to analyze text in PDF files of clinical trials. It combines linear models and convolutional neural networks to classify the documents as having a low or unclear/high risk of bias for the first four questions of the Cochrane RoB1 tool. RobotReviewer has been trained on 12,808 trial PDFs from the Cochrane Database of Systematic Reviews. During development, it achieved higher accuracy than new human assessors by 1-2% when compared to the original dataset (Zhang, Y., 2016).
If you use RobotReviewer in your research, please cite:
Marshall, I. J., Kuiper, J., Banner, E., & Wallace, B. C. (2017, July). Automating biomedical evidence synthesis: RobotReviewer. In Proceedings of the conference. Association for Computational Linguistics. Meeting (Vol. 2017, p. 7).