Puraite GmbH

Puraite is a German deep-tech SME based in Bielefeld, founded by AI researchers from Paderborn University. We build an explainable AI platform that automates systematic reviews and other forms of evidence synthesis for health research. What the platform does The system runs the full review workflow: search strategy, deduplication, title and abstract screening, full-text screening, data extraction, risk-of-bias assessment, meta-analysis and reporting. Every decision the AI makes is traceable to the source text and can be checked by a human reviewer. Methods follow the Cochrane Handbook, PRISMA 2020, GRADE, RoB 2 and ROBINS-I. All statistics run on validated R packages. Why this matters for Horizon Europe projects Health and cancer projects depend on evidence synthesis at several points: the state-of-the-art analysis at proposal stage, living reviews during the project, input for health economic models and HTA dossiers, and guideline-ready evidence at the end. Done by hand this takes many months per review. Puraite reduces it to days while keeping the method fully transparent and auditable. What we bring to a consortium - An AI and methods partner for work packages on evidence synthesis, living reviews and HTA input - Explainable AI expertise: traceable model decisions, human-in-the-loop validation, audit trails - A working platform available for pilot use from month one of the project - SME participation for exploitation and dissemination tasks Calls of interest HORIZON-MISS-2027-02-CANCER-04 (health economics, health systems and outcomes research), HORIZON-HLTH-2027-03-TOOL-08 (towards AGI for healthcare) and HORIZON-HLTH-2027-01-CARE-02 (adverse drug reactions in polypharmacy). We are looking for coordinators and consortia that need a partner for evidence synthesis, explainable AI or trustworthy clinical AI.