# Lumina Reviewer

> AI-assisted systematic literature review and citation screening platform. Prioritize abstracts with active learning and dense vector embeddings from paper #1, search across 7 scholarly and grey literature databases, use advisory Pixel-Bot AI recommendations, and automate PRISMA 2020 flow reporting.

Lumina Reviewer is designed for biomedical, clinical, social science, and interdisciplinary evidence synthesis. It automates repetitive literature screening tasks while keeping human researchers in rigorous methodological control.

## Core Capabilities
- **Active Learning Screening**: Ranks screening queues from paper #1 using dense vector embeddings and semantic representations (`text-embedding-3-small`) combined with active learning classifiers, eliminating cold-start screening lag.
- **Multi-Database Search**: Unified live search across 7 databases: PubMed, OpenAlex, Europe PMC, arXiv, bioRxiv/medRxiv, OpenAIRE, and official Government publications.
- **Pixel-Bot AI Assistant**: Provides advisory Include, Exclude or Maybe recommendations with reasoning. It does not count as a second independent human reviewer; researchers make the final decisions.
- **Reviewer-Blind Dual Screening**: Multi-reviewer workspace with blinded screening, automated disagreement detection, and conflict resolution workflows.
- **PRISMA 2020 Generator**: Real-time interactive PRISMA 2020 flow tracking with exportable high-resolution SVG and PNG diagrams.
- **Open-Access Full-Text Discovery**: Automated PDF retrieval via Unpaywall and DOI resolvers with built-in split-screen web PDF viewer.
- **Citation Ingestion & Deduplication**: Imports RIS, NBIB, CSV, BibTeX and supported tagged text. Citation-import duplicate checks use normalized DOI, normalized title similarity and an author/year/title fallback, with guards for conflicting identifiers and publication types.
- **Early Stopping Rules**: Work Saved over Sampling (WSS) and heuristic stopping strategies based on screening history. These heuristics do not establish recall against all relevant studies.

## Full Documentation
- [Complete Platform Documentation](https://luminareviewer.com/llms-full.txt): In-depth platform architecture, vector embeddings pipeline, multi-database search integrations, stopping heuristics, and competitor comparison matrix.

## Pricing
- **Free Tier**: Free interactive screening demo, free online PRISMA 2020 flow diagram generator, and preview workspace.
- **Starter (€9/month)**: Designed for solo researchers. 1 active project, up to 750 papers per project, active learning AI screening, export to CSV/RIS.
- **Duo (€19/month)**: For larger reviews and ongoing research, solo or together. Up to 100 active projects, 25,000 papers per project and priority support. A second reviewer, one shared workspace, independent dual screening and conflict resolution are included. Capacity is shared when collaborating.
- **Team (€20/month base, 1 seat included)**: Designed for collaborative research teams. Up to 100 active team projects, up to 100,000 papers per project, reviewer-blind dual screening, conflict resolution dashboard, PRISMA audit trails, seat add-ons (€7/month per extra seat).

## Tools & Interactive Features
- [PRISMA 2020 Flow Diagram Generator](https://luminareviewer.com/tools/prisma-flow-diagram-generator/): Free online tool to generate publication-ready PRISMA 2020 flow diagrams in SVG and PNG.
- [Multi-Database Literature Search](https://luminareviewer.com/tools/multi-database-search/): Search 7 scholarly and grey literature databases simultaneously without API keys.
- [Interactive Screening Demo](https://luminareviewer.com/demo/): Test active learning screening on real benchmark datasets without creating an account.
- [Pricing Plans & Calculator](https://luminareviewer.com/pricing/): Transparent pricing starting at €9/month with no vendor lock-in.

## Systematic Review Guides & Methodology
- [Systematic Review Protocol Guide](https://luminareviewer.com/guide/systematic-review-protocol/): Developing rigorous, PROSPERO-ready systematic review protocols.
- [How to Do a Systematic Review](https://luminareviewer.com/guide/how-to-do-a-systematic-review/): Complete handbook for evidence synthesis from search to synthesis.
- [PICO Framework Guide](https://luminareviewer.com/guide/pico-framework/): Structuring research questions using Population, Intervention, Comparison, and Outcome.
- [Inclusion and Exclusion Criteria](https://luminareviewer.com/guide/inclusion-exclusion-criteria/): Establishing objective eligibility boundaries for screening.
- [Title and Abstract Screening](https://luminareviewer.com/guide/title-abstract-screening/): Practical strategies for rapid, accurate title and abstract screening.
- [Active Learning in Systematic Reviews](https://luminareviewer.com/guide/active-learning-systematic-reviews/): How machine learning and text embeddings prioritize candidate studies.
- [Dual Screening & Conflict Resolution](https://luminareviewer.com/guide/dual-screening/): Coordinating independent dual screening and adjudicating reviewer disagreements.
- [PRISMA 2020 Checklist Guide](https://luminareviewer.com/guide/prisma-2020-checklist/): Navigating the 27-item reporting guideline for systematic reviews.
- [PRISMA Flow Diagram Explained](https://luminareviewer.com/guide/prisma-flow-diagram-explained/): How to calculate and document identification, screening, and inclusion phases.
- [Stopping Strategies Guide](https://luminareviewer.com/guide/stopping-strategies/): Target recall thresholds and empirical stopping heuristics in AI screening.
- [Full-Text PDF Retrieval](https://luminareviewer.com/guide/full-text-pdf-retrieval/): Open-access discovery pipelines, Unpaywall integration, and document management.
- [Grey Literature Search Guide](https://luminareviewer.com/guide/grey-literature-search/): Locating preprints, government reports, and institutional grey literature.
- [Deduplicating Search Results](https://luminareviewer.com/guide/deduplicate-systematic-review-search-results/): Multi-pass deduplication strategies across overlapping scientific databases.
- [How to Import RIS Files](https://luminareviewer.com/guide/how-to-import-ris-files/): Technical specifications of RIS tags, character encodings, and common syntax fixes.
- [Importing Scientific Databases](https://luminareviewer.com/guide/import-from-scientific-databases/): Multi-database citation import and unified screening preparation.

## Database Export Guides
- [Export from PubMed](https://luminareviewer.com/guide/export-from-pubmed/): Exporting search results to .nbib and CSV formats with complete abstracts.
- [Export from Scopus](https://luminareviewer.com/guide/export-from-scopus/): Configuring RIS exports up to 20,000 records with abstracts and keywords.
- [Export from Web of Science](https://luminareviewer.com/guide/export-from-web-of-science/): Managing 1,000-record batch limits and Full Record RIS settings.
- [Export from Embase](https://luminareviewer.com/guide/export-from-embase/): Exporting biomedical literature and EMTREE indexing in RIS format.
- [Export from CINAHL](https://luminareviewer.com/guide/export-from-cinahl/): Nursing and allied health reference exports via EBSCOhost folder workflows.
- [Export from Cochrane Library](https://luminareviewer.com/guide/export-from-cochrane-library/): Exporting CDSR systematic reviews and CENTRAL clinical trials in RIS format.
- [Export from Google Scholar](https://luminareviewer.com/guide/export-from-google-scholar/): Exporting citations via My Library and desktop citation harvesters.
- [Export from PsycINFO](https://luminareviewer.com/guide/export-from-psycinfo/): Exporting behavioral science citations across APA PsycNet, EBSCOhost, and Ovid.

## Software Comparisons
- [Rayyan Alternative](https://luminareviewer.com/compare/rayyan-alternative/): Lumina Reviewer vs Rayyan for AI-assisted systematic review screening.
- [Covidence Alternative](https://luminareviewer.com/compare/covidence-alternative/): Lumina Reviewer vs Covidence feature comparison, workflow flexibility, and pricing.
- [ASReview Alternative](https://luminareviewer.com/compare/asreview-alternative/): Lumina Reviewer vs ASReview active learning, team collaboration, and cloud access.
- [DistillerSR Alternative](https://luminareviewer.com/compare/distillersr-alternative/): Modern, affordable alternative to DistillerSR for review teams.
- [EndNote Alternative](https://luminareviewer.com/compare/endnote-alternative/): Why systematic review teams use dedicated screening platforms over reference managers.
- [Excel Alternative](https://luminareviewer.com/compare/excel-alternative/): Replacing error-prone spreadsheets with purpose-built systematic review screening software.
- [Elicit Alternative](https://luminareviewer.com/compare/elicit-alternative/): Comparing AI literature synthesis with systematic review screening methodology.

## Practical Citation Workflows & Reproducible Examples
- [PubMed NBIB to RIS: conversion and field checks](https://luminareviewer.com/guide/pubmed-nbib-to-ris/): Convert PubMed NBIB references to RIS, compare the fields, and download matched synthetic examples. Lumina also imports NBIB directly.
- [RIS import missing abstracts: diagnose and fix the export](https://luminareviewer.com/guide/ris-import-missing-abstracts/): Find out why abstracts are missing from RIS imports. Inspect AB tags, compare export settings, and test downloadable full and citation-only examples.
- [Import a Rayyan export into Lumina: citations and migration limits](https://luminareviewer.com/guide/import-rayyan-export/): Move Rayyan citation exports into Lumina using RIS. Check what transfers and preserve decisions, notes and the original audit log separately.
- [Deduplicate PubMed and Embase results: a worked example](https://luminareviewer.com/guide/deduplicate-pubmed-embase/): Follow a downloadable PubMed and Embase overlap example: four records, one duplicate, three unique citations, with PRISMA counting and matching limits.
- [Citation import benchmark: reproducible metadata checks](https://luminareviewer.com/guide/citation-import-benchmark/): Download Lumina’s synthetic NBIB and RIS fixtures and reproduce record counts, missing-abstract checks, migration limits and cross-database deduplication.
