---
title: Dmitry Scherbakov
type: entity
tags: [researcher, biomedical-informatics, systematic-review, llm, automation]
affiliation: Biomedical Informatics Center, Medical University of South Carolina (MUSC)
sources: ["[[sources/2025-scherbakov-llms-as-tools-literature-reviews]]"]
created: 2026-04-20
updated: 2026-04-20
---

# Dmitry Scherbakov

Biomedical-informatics researcher at the **Medical University of South Carolina (MUSC)**. Co-first author of the JAMIA 2025 LLM-assisted systematic review of LLM-assisted review automation.

## Contributions in this wiki
- [[sources/2025-scherbakov-llms-as-tools-literature-reviews]] — JAMIA 2025: a self-demonstrating SR (172 eligible studies) that deploys a Covidence + GPT-4o plugin for screening, full-text review and extraction. Reports 83.0% precision / 86.0% recall on extraction.

## Role
Landscape mapper and toolbuilder for LLM-assisted SR; contributes the Covidence plugin and the "LLM as third reviewer" pattern as an operational recipe.