Toward Medical Deepfake Detection: A Comprehensive Dataset and Novel Method

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Shuaibo, Xing, Zhaohu, Wang, Hongqiu, Hao, Pengfei, Li, Xingyu, Liu, Zekai, Zhu, Lei
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914047109103616
author Li, Shuaibo
Xing, Zhaohu
Wang, Hongqiu
Hao, Pengfei
Li, Xingyu
Liu, Zekai
Zhu, Lei
author_facet Li, Shuaibo
Xing, Zhaohu
Wang, Hongqiu
Hao, Pengfei
Li, Xingyu
Liu, Zekai
Zhu, Lei
contents The rapid advancement of generative AI in medical imaging has introduced both significant opportunities and serious challenges, especially the risk that fake medical images could undermine healthcare systems. These synthetic images pose serious risks, such as diagnostic deception, financial fraud, and misinformation. However, research on medical forensics to counter these threats remains limited, and there is a critical lack of comprehensive datasets specifically tailored for this field. Additionally, existing media forensic methods, which are primarily designed for natural or facial images, are inadequate for capturing the distinct characteristics and subtle artifacts of AI-generated medical images. To tackle these challenges, we introduce \textbf{MedForensics}, a large-scale medical forensics dataset encompassing six medical modalities and twelve state-of-the-art medical generative models. We also propose \textbf{DSKI}, a novel \textbf{D}ual-\textbf{S}tage \textbf{K}nowledge \textbf{I}nfusing detector that constructs a vision-language feature space tailored for the detection of AI-generated medical images. DSKI comprises two core components: 1) a cross-domain fine-trace adapter (CDFA) for extracting subtle forgery clues from both spatial and noise domains during training, and 2) a medical forensic retrieval module (MFRM) that boosts detection accuracy through few-shot retrieval during testing. Experimental results demonstrate that DSKI significantly outperforms both existing methods and human experts, achieving superior accuracy across multiple medical modalities.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15711
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Toward Medical Deepfake Detection: A Comprehensive Dataset and Novel Method
Li, Shuaibo
Xing, Zhaohu
Wang, Hongqiu
Hao, Pengfei
Li, Xingyu
Liu, Zekai
Zhu, Lei
Computer Vision and Pattern Recognition
The rapid advancement of generative AI in medical imaging has introduced both significant opportunities and serious challenges, especially the risk that fake medical images could undermine healthcare systems. These synthetic images pose serious risks, such as diagnostic deception, financial fraud, and misinformation. However, research on medical forensics to counter these threats remains limited, and there is a critical lack of comprehensive datasets specifically tailored for this field. Additionally, existing media forensic methods, which are primarily designed for natural or facial images, are inadequate for capturing the distinct characteristics and subtle artifacts of AI-generated medical images. To tackle these challenges, we introduce \textbf{MedForensics}, a large-scale medical forensics dataset encompassing six medical modalities and twelve state-of-the-art medical generative models. We also propose \textbf{DSKI}, a novel \textbf{D}ual-\textbf{S}tage \textbf{K}nowledge \textbf{I}nfusing detector that constructs a vision-language feature space tailored for the detection of AI-generated medical images. DSKI comprises two core components: 1) a cross-domain fine-trace adapter (CDFA) for extracting subtle forgery clues from both spatial and noise domains during training, and 2) a medical forensic retrieval module (MFRM) that boosts detection accuracy through few-shot retrieval during testing. Experimental results demonstrate that DSKI significantly outperforms both existing methods and human experts, achieving superior accuracy across multiple medical modalities.
title Toward Medical Deepfake Detection: A Comprehensive Dataset and Novel Method
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.15711