Salvato in:
Dettagli Bibliografici
Autori principali: Pan, Tan, Mei, Shuhao, Sun, Yixuan, Guo, Kaiyu, Jiang, Chen, Tan, Zhaorui, Li, Mengzhu, Han, Limei, Zou, Xiang, Cheng, Yuan, Baktashmotlagh, Mahsa
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2605.14654
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917495567286272
author Pan, Tan
Mei, Shuhao
Sun, Yixuan
Guo, Kaiyu
Jiang, Chen
Tan, Zhaorui
Li, Mengzhu
Han, Limei
Zou, Xiang
Cheng, Yuan
Baktashmotlagh, Mahsa
author_facet Pan, Tan
Mei, Shuhao
Sun, Yixuan
Guo, Kaiyu
Jiang, Chen
Tan, Zhaorui
Li, Mengzhu
Han, Limei
Zou, Xiang
Cheng, Yuan
Baktashmotlagh, Mahsa
contents Self-supervised pre-training methods in medical imaging typically treat each individual as an isolated instance, learning representations through augmentation-based objectives or masked reconstruction. They often do not adequately capitalize on a key characteristic of physiological features: anatomical structures maintain consistent spatial relationships across individuals (instances), such as the thalamus being medial to the basal ganglia, regardless of variations in brain size, shape, or pathology. We propose leveraging this cross-instance topological consistency as a supervisory signal. The challenge arises from the inherent variability in medical imaging, which can differ significantly across instances and modalities. To tackle this, we focus on two alignment regimes. (i) Intra-instance: with pixel-level correspondences available, a cross-modal triplet objective explicitly preserves local neighborhood topology. (ii) Inter-instance: without such supervision, we derive pseudo-correspondences to control partial neighborhood alignment and prevent topology collapse across modalities. We validate our approach across 7 downstream multi-modal tasks, achieving average improvements of 1.1% and 5.94% in segmentation and classification tasks, respectively, and demonstrating significantly better robustness when modalities are missing at test time.
format Preprint
id arxiv_https___arxiv_org_abs_2605_14654
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imaging
Pan, Tan
Mei, Shuhao
Sun, Yixuan
Guo, Kaiyu
Jiang, Chen
Tan, Zhaorui
Li, Mengzhu
Han, Limei
Zou, Xiang
Cheng, Yuan
Baktashmotlagh, Mahsa
Computer Vision and Pattern Recognition
Self-supervised pre-training methods in medical imaging typically treat each individual as an isolated instance, learning representations through augmentation-based objectives or masked reconstruction. They often do not adequately capitalize on a key characteristic of physiological features: anatomical structures maintain consistent spatial relationships across individuals (instances), such as the thalamus being medial to the basal ganglia, regardless of variations in brain size, shape, or pathology. We propose leveraging this cross-instance topological consistency as a supervisory signal. The challenge arises from the inherent variability in medical imaging, which can differ significantly across instances and modalities. To tackle this, we focus on two alignment regimes. (i) Intra-instance: with pixel-level correspondences available, a cross-modal triplet objective explicitly preserves local neighborhood topology. (ii) Inter-instance: without such supervision, we derive pseudo-correspondences to control partial neighborhood alignment and prevent topology collapse across modalities. We validate our approach across 7 downstream multi-modal tasks, achieving average improvements of 1.1% and 5.94% in segmentation and classification tasks, respectively, and demonstrating significantly better robustness when modalities are missing at test time.
title Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imaging
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.14654