pith:CAPME2KJ
Towards Modality-Agnostic Medical Image Anomaly Detection: A Training-Free Manifold Refinement Approach
Mean Shift Density Enhancement refines latent features from pretrained backbones to sharpen one-class anomaly scoring in medical images.
arxiv:2604.19191 v2 · 2026-04-21 · cs.CV · cs.AI
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{CAPME2KJDFSMKE7VRWNIHUC6LX}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
MSDE achieves the highest AUC on four datasets and the highest Average Precision on five datasets, including near-perfect performance on brain tumor detection (0.981 AUC/AP).
That the iterative manifold-shifting performed by Mean Shift Density Enhancement reliably moves normal samples toward higher-likelihood regions in the latent space of arbitrary pretrained backbones, thereby sharpening the subsequent Gaussian density estimate for anomaly scoring.
A new Mean Shift Density Enhancement procedure applied to self-supervised embeddings yields state-of-the-art anomaly detection AUC and average precision on seven medical imaging datasets.
Receipt and verification
| First computed | 2026-06-30T01:16:30.054884Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
101ec269491964c513f58d9a83d05e5ddbae08b03079464f5a86567981c4b606
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/CAPME2KJDFSMKE7VRWNIHUC6LX \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 101ec269491964c513f58d9a83d05e5ddbae08b03079464f5a86567981c4b606
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "9d0eae0cb8656a187a83f681977a39ffdfff081f23d0f37b632aa768aa5381da",
"cross_cats_sorted": [
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-21T07:58:44Z",
"title_canon_sha256": "2f350d25ff3cb5168fd6facd99724f111f661f8c30f422cd5f2cdacf7039baf8"
},
"schema_version": "1.0",
"source": {
"id": "2604.19191",
"kind": "arxiv",
"version": 2
}
}