Systematic Literature Review
AI-Assisted Search to Synthesis
From search to evidence synthesis: SaiLR supports every step, the decisions stay yours.
Search academic databases in one go, remove duplicates, screen titles, abstracts and full texts against your criteria, extract the data and summarise the evidence.
For staff and doctoral researchers of the University of Augsburg and University Hospital Augsburg, provided by the University Library.
Already have access? Sign InHow it works
From research question
to evidence synthesis
SaiLR handles the mechanical stages so you can focus on interpretation.
Set up your protocol
Write your research question. SaiLR drafts the search strings and the inclusion and exclusion criteria, and you check and change them before anything runs.
- Choose the databases to search
- Boolean search strings, adapted to each database
- Re-run a review with the same protocol, or adjust it first
Search and remove duplicates
SaiLR searches the chosen databases at the same time and merges the records that several of them report, matching by DOI, title and other metadata.
- All chosen databases in one search
- Duplicates matched by DOI, title and metadata
- Each merged record shows which databases reported it
Deep learning in oncology: A review
PubMedDeep learning in oncology: A review
arXivAI for cancer detection: Systematic analysis
Semantic ScholarMachine learning clinical trials meta-analysis
SpringerScreen and export
Each paper is checked against your criteria and marked Include, Maybe or Exclude, with a written reason. You can change any decision; the includes go on to full-text screening.
- A written reason for every decision
- Fixed rules decide the clear cases, the same way every time
- Export as RIS or CSV, with a PRISMA 2020 flow diagram
Deep Learning for Cancer Prognosis Using EHR Data
Matches all inclusion criteria. RCT, 2022, human.
Attention Mechanisms in Medical Image Transformers
Unclear clinical endpoint. Recommend manual check.
Sentiment Analysis with Traditional ML Approaches
Outside domain scope. Non-clinical.
From full text to synthesis
SaiLR retrieves the open-access full texts and assesses them strictly against your criteria. From the included studies it extracts the data, assesses risk of bias and drafts the synthesis. The AI proposes and you decide, on your own or as two reviewers following Cochrane.
- Full-text screening with a reason for every exclusion; upload missing PDFs yourself
- Data extraction and risk of bias (RoB 2), every value with a supporting quote
- Summary of findings and a cited narrative per research question, as Word, Markdown or CSV
Full-text screening
31 included · 21 excluded with reasons
Data extraction
31 studies · all quotes verified
Risk of bias
RoB 2 and a generic tool
Evidence synthesis
5 outcomes · 3 research questions
What's Included
What SaiLR does,
step by step
The parts of a systematic review that SaiLR takes on: drafting the protocol, searching, removing duplicates, screening, assessing full texts, extracting data, judging risk of bias, synthesising the evidence and exporting.
Multi-Database Search
One search goes to every database you choose, with the search string adapted to each.
AI Screening
Each paper is checked against your inclusion and exclusion criteria. Every decision comes with its reason, and you can change it.
Deduplication
Records that several databases report are merged into one, and you can see which were merged.
Protocol Generation
Describe your research question. SaiLR drafts Boolean search strings, inclusion and exclusion criteria, and a structured version of your question for you to check.
PRISMA Export
A PRISMA 2020 flow diagram with the counts of your run, and the paper list as RIS or CSV.
From full text to synthesis
Included papers go on to full-text screening, with a reason for every exclusion. SaiLR then extracts the data, assesses risk of bias and drafts a summary-of-findings table with a narrative per research question. Every value is backed by a quote, and you confirm or change every judgment.
Frequently Asked Questions
Common Questions
Answers to what researchers typically ask before using SaiLR.
Does AI screening replace manual review?
No. SaiLR classifies each paper as Include, Maybe, or Exclude with a written rationale against your criteria. Papers marked Maybe always require your judgment. The tool reduces the volume of papers you need to read manually - it does not make final decisions on your behalf.
AI Screening
Sorts the papers and gives a reason
Your review
You check and change decisions
Final Decision
Always yours
No. SaiLR classifies each paper as Include, Maybe, or Exclude with a written rationale against your criteria. Papers marked Maybe always require your judgment. The tool reduces the volume of papers you need to read manually - it does not make final decisions on your behalf.
AI Screening
Sorts the papers and gives a reason
Your review
You check and change decisions
Final Decision
Always yours
Request Access
SaiLR is currently available to a small group of researchers by invitation. Leave your details and we'll be in touch.
No account required to request access. We'll reach out via email.