A Hybrid Semi-Automated Workflow for Systematic and Literature Review Processes with Large Language Model Analysis
Problem
Systematic reviews are rigorous but labor-intensive due to collaborative protocols, strict procedures, and large numbers of documents to screen.
Why it matters
The manual workload of systematic reviews slows evidence synthesis used across research and practice.
What the AI did
A semi-automated workflow uses Gemini-Pro to autonomously summarize text and extract key information, replacing manual title/abstract screening, full-text screening, and data extraction while keeping a human in the loop for inclusion/exclusion decisions.
Results
not stated in source
Impact
Peer-reviewed workflow proposal; claimed accuracy maintenance is not quantified in the abstract.
Record
| Venue | Future Internet |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Anjia Ye, Ananda Maiti, Matthew W. Schmidt, SJ Pedersen |
| Citations | 28 (OpenAlex, at import time) |
| References | 28 |
| Model or method | Gemini-Pro (GPT-4/ChatGPT also referenced) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/fi16050167 |
| Open access | gold |
| Keywords | systematic review, LLM, Gemini-Pro, screening automation, human-in-the-loop |
Importance rating
significant — thousands to millions affected
Systematic reviews underpin evidence synthesis across research and practice; reducing their manual workload benefits the broader research community, though the abstract states no scale of affected users.