New K-Space Signature framework detects medical deepfakes in MRI scans

Researchers have developed a new framework called the K-Space Signature (KSS) to detect medical deepfakes in MRI scans. This method analyzes images in the frequency domain, specifically the Logarithmic Power Spectral Density (Log-PSD) space, to identify synthetic data. The KSS framework utilizes a 3D MLP-Mixer architecture with an ArcFace metric-learning head to process spectral artifacts. Experiments show this approach achieves over 0.99 Accuracy and ROC-AUC on synthetic datasets and demonstrates robust zero-shot generalization to unseen scanners. AI

IMPACT This research introduces a novel method for detecting synthetic medical images, which could enhance the integrity of medical data and protect against malicious use of generative models in healthcare.

RANK_REASON Academic paper detailing a new method for detecting medical deepfakes. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• 3D MLP-Mixer
• ArcFace
• arXiv
• Kearns-Sayre syndrome
• K-Space Signature
• Logarithmic Power Spectral Density
• Log-PSD
• magnetic resonance imaging

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →