- Home
- AI Literature Review Software
- RobotReviewer
RobotReviewer
What it is
RobotReviewer is a clinical-trial analysis tool that annotates PDF reports with PICO characteristics and likely bias assessments using the Cochrane Risk of Bias tool. Upload PDFs through its website to extract key data and view a summary table. Its REST API can process groups of reports to synthesize results across trials, and the project can run as local Docker services with a web app and API server. The software is free to use under the GNU GPL version 3.0, with a free plan listed as free. Self-hosting uses Grobid for document processing, RabbitMQ and Celery for workers, and SciBERT through bert-as-a-service. Manual setup calls for Python 3.6 and TensorFlow 1.12.0, along with those services. Docker instructions cover GPU use, and the worker can also run CPU-only. Installation is documented as tested on Ubuntu and Mac OS; Windows is possible but described as substantially difficult. The maintainers ask users to cite the 2017 ACL paper and invite questions by email, including requests for integration help.
For and against
For
- Annotates PICO characteristics and likely bias assessments
- PDF uploads produce extracted data and a summary table
- API can synthesize groups of trial reports
- Free to use under GNU GPL version 3.0
Against
- Manual setup requires Python 3.6 and TensorFlow 1.12.0
- Windows installation involves substantial difficulty
- Self-hosting requires several processing services
The verdict
RobotReviewer offers PDF analysis through a website and an API for groups of trial reports. Its free, open-source licensing is useful, but manual deployment requires a collection of services and Windows setup is difficult.
Who it suits
RobotReviewer suits researchers and review teams working with clinical-trial PDFs who need structured PICO and bias annotations. Its API and self-hosted setup may also suit teams integrating trial analysis into other software.
Plans and prices
Free
Free of charge · citation requested when used · attribution link required for integrations
Read 15 Oct 2026 · pypi.org
On the record
- API
- A REST API exposes the underlying models for integration into systematic-review software.pmc.ncbi.nlm.nih.gov · read 15 Oct 2026
- API extensibility
- The API endpoint implementation is available in the repository's robotreviewer/app.py file.github.com · read 15 Oct 2026
- Batch synthesis
- The newer API works with groups of trial reports so results can be synthesized across multiple studies.github.com · read 15 Oct 2026
- Citation
- The maintainers ask users to cite the 2017 ACL paper when using RobotReviewer.github.com · read 7 Oct 2026
- Citation year
- The cited RobotReviewer evaluation was published in the Journal of the American Medical Informatics Association in 2015.github.com · read 15 Oct 2026
- Dependencies
- Installation requires Grobid, Java, RabbitMQ, Celery, spaCy, NLTK, TensorFlow, and SciBERT model files.github.com · read 15 Oct 2026
- Deployment
- The project can run locally as Docker services, with a web app and API server exposed on local ports.github.com · read 7 Oct 2026
- Docker deployment
- Docker Compose files are provided for building and running the web, API, and supporting services.github.com · read 15 Oct 2026
- Documentation depth
- The README includes Docker, manual installation, configuration, startup, API, testing, and troubleshooting instructions.github.com · read 15 Oct 2026
- Funding
- The project acknowledges NIH National Library of Medicine and UK Medical Research Council grant support.github.com · read 15 Oct 2026
- GPU option
- The project provides Docker instructions for GPU use and also documents running the worker CPU-only.github.com · read 7 Oct 2026
- GPU support
- A separate Docker Compose configuration supports NVIDIA GPU execution.github.com · read 15 Oct 2026
- Integration
- The research paper identifies Covidence and similar evidence-synthesis platforms as potential integration targets.pmc.ncbi.nlm.nih.gov · read 15 Oct 2026
- Integrations
- The self-hosted setup uses Grobid for document processing, RabbitMQ and Celery for worker processing, and SciBERT through bert-as-a-service.github.com · read 7 Oct 2026
- License
- RobotReviewer is open source under the GNU GPL version 3.0 license.github.com · read 15 Oct 2026
- Limit: human oversight
- Researchers in the real-time study still considered human input and discussion superior to semi-automated assessment.pubmed.ncbi.nlm.nih.gov · read 15 Oct 2026
- Limit: incomplete automation
- A 2026 scoping review describes RobotReviewer as time-saving but still requiring human supervision and reports uncertain reliability in some studies.link.springer.com · read 15 Oct 2026
- Limit: lower accuracy than humans
- The original evaluation reported overall risk-of-bias classification accuracy five to ten percentage points below published human systematic reviews.pmc.ncbi.nlm.nih.gov · read 15 Oct 2026
