Metraipolis › archive › dc_d626dc0a088f

M3-Score: Fidelity, Memorization and Coverage as Separate Axes for Evaluating Generative Radiology Image Models — arXiv:2609.33769

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33769
License (as archived)
CC0-1.0
Archived
2026-10-07T23:51:45.397Z
SHA-256 of the archived text
23c3a908111457e6d0f4c0f652adab8c61a3af98cf2e2bfe2630e12bec6e6d64
Stable URL of this page
https://metraipolis.com/documento/dc_d626dc0a088f

How to cite

M3-Score: Fidelity, Memorization and Coverage as Separate Axes for Evaluating Generative Radiology Image Models — arXiv:2609.33769 — arXiv (en), https://arxiv.org/abs/2609.33769 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d626dc0a088f (Metraipolis archive dc_d626dc0a088f, sha256 23c3a908111457e6d0f4c0f652adab8c61a3af98cf2e2bfe2630e12bec6e6d64)
BibTeX
@misc{dcd626dc0a088f,
  title = {M3-Score: Fidelity, Memorization and Coverage as Separate Axes for Evaluating Generative Radiology Image Models — arXiv:2609.33769},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33769},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d626dc0a088f; sha256 23c3a908111457e6d0f4c0f652adab8c61a3af98cf2e2bfe2630e12bec6e6d64},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33769. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Quantitative evaluation of generative models for radiology remains challenging. Clinically relevant structures are often small and infrequent, feature spaces learned from natural images may represent them poorly, and a single summary score cannot distinguish limited fidelity from limited diversity. This study proposes the Medical Multi-axis Maximum Mean Discrepancy score (M3-Score), an evaluation framework based on RadioDINO-s16, a frozen vision transformer pretrained on radiology images. M3-Score reports three complementary axes computed at pre-specified encoder depths: \emph{fidelity}, measured by an unbiased multi-bandwidth radial basis function (RBF) MMD$^2$ at the final block; \emph{memorization}, measured by nearest-neighbor distances at 75\% depth; and \emph{coverage}, defined as the fraction of real images with a generated neighbor within their $k$-nearest-neighbor radius at 33\% depth. Reference sets are sampled across subjects to limit the influence of correlated slices. On BraTS brain MRI, the fidelity axis ordered five comparison sets of increasing severity (Spearman $\rho = 1.00$), and a subject-disjoint real set yielded $\mathrm{MMD}^2 = 0$ (permutation $p = 1$). An unconditional denoising diffusion probabilistic model achieved $\mathrm{MMD}^2 = 0.073$ (95\% confidence interval $[0.071, 0.080]$) but covered only 38\% of the real distribution. Under progressive mode dropping, $k$-NN manifold recall increased at all twelve encoder blocks, whereas the proposed coverage estimator decreased monotonically ($\rho = -1.00$). RadioDINO-s16 features separated real brain MRI from generated samples with a ROC-AUC of 0.819, compared with 0.555 for InceptionV3 and 0.582 for CLIP. Across a twentyfold range of sample sizes, the mean M3 value varied by a factor of 1.05, compared with 2.52 for the Fréchet Inception Distance.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.33769 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T23:29:27.254Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).