← 研踪 PaperTrailarXiv:2611.21955 · v1
SCAN: Site-Aware Collaborative Privacy-Preserving Adaptation of Vision-Language Models for Neurological Disease Classification
首次提交 2026-11-17 06:36
版本更新 2026-11-17 06:36
雷达观测 2026-11-18 19:43
摘要
Worldwide neuroimaging data could strengthen diagnostic foundation models, but privacy rules often prevent centralizing brain MRI scans. We present SCAN, a site-aware collaborative adaptation framework for Neurological Disease Classification that lets new sites adapt a 3D medical vision-language model (VLM) without retraining or sharing patient scans. Each site keeps exact class prototypes for local prediction and can share privacy-protected class summaries: norm-bounded sums and counts perturbed once with calibrated noise. Using 7,641 scans from 67 sites across 12 countries, covering Alzheimer's disease (AD), stroke, epilepsy and controls, we assess local adaptation, cross-site benefits, and effects on existing sites. At seven unseen sites, the frozen VLM's zero-shot AUCs were 0.833-0.945 for AD and stroke. Training-free local prototypes improved epilepsy AUC at three sites from 0.598/0.625/0.508 to 0.684/0.729/0.762 and AD AUC at one site from 0.833 to 0.863, nearly matching a classifier trained at each site (mean AUC 0.757 vs. 0.759). With differential privacy ( = 4), new sites contributed without materially reducing accuracy at existing sites (AUC change 0.2%), although gains from sharing were limited, and membership attacks performed near chance. SCAN offers hospitals a worldwide framework to adapt and share brain MRI VLMs without exchanging scans or retraining.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-17 06:36 · 最新观测版本
SCAN: Site-Aware Collaborative Privacy-Preserving Adaptation of Vision-Language Models for Neurological Disease Classification
Worldwide neuroimaging data could strengthen diagnostic foundation models, but privacy rules often prevent centralizing brain MRI scans. We present SCAN, a site-aware collaborative adaptation framework for Neurological Disease Classification that lets new sites adapt a 3D medical vision-language model (VLM) without retraining or sharing patient scans. Each site keeps exact class prototypes for local prediction and can share privacy-protected class summaries: norm-bounded sums and counts perturbed once with calibrated noise. Using 7,641 scans from 67 sites across 12 countries, covering Alzheimer's disease (AD), stroke, epilepsy and controls, we assess local adaptation, cross-site benefits, and effects on existing sites. At seven unseen sites, the frozen VLM's zero-shot AUCs were 0.833-0.945 for AD and stroke. Training-free local prototypes improved epilepsy AUC at three sites from 0.598/0.625/0.508 to 0.684/0.729/0.762 and AD AUC at one site from 0.833 to 0.863, nearly matching a classifier trained at each site (mean AUC 0.757 vs. 0.759). With differential privacy ( = 4), new sites contributed without materially reducing accuracy at existing sites (AUC change 0.2%), although gains from sharing were limited, and membership attacks performed near chance. SCAN offers hospitals a worldwide framework to adapt and share brain MRI VLMs without exchanging scans or retraining.
查看此版本解读 →查看此版本原文 ↗