Learning How Much, Not Just What: Cross-Patient Burden Order for CT Vision-Language Pretraining

• URL: http://arxiv.org/abs/2608.00231v1
• Date: Fri, 31 Jul 2026 19:23:02 GMT
• Title: Learning How Much, Not Just What: Cross-Patient Burden Order for CT Vision-Language Pretraining
• Abstract summary: Volumetric CT vision-language pretraining learns 3D representations from scan-report pairs.<n>Nothing separates a mild from an extensive case of the same finding along a consistent direction.<n>Cross-sectional cohorts already encode weak burden cues across different patients.
• Score: 4.757530530297623
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Volumetric CT vision-language pretraining learns 3D representations from scan-report pairs, but global and anatomy-aware objectives supervise only correspondence: they establish what is present and leave how much unconstrained. Nothing separates a mild from an extensive case of the same finding along a consistent direction, so the graded burden language in reports collapses into a present/absent signal. Longitudinal supervision would supply this order, but patient-matched CT pairs are scarce at scale; cross-sectional cohorts already encode weak burden cues across different patients. We introduce Spectrum, an anatomy-conditioned framework that represents each study at whole-study and organ scopes. For each organ-mapped pathology, a rule-based scorer mines confidence-filtered lower-to-higher pairs of different patients, and Burden-Direction Alignment (BDA) aligns the pathology-conditioned image delta with the report delta at each scope, separating that direction from its reverse. Because the endpoints are different people, a target-conditioned aligner first makes them comparable, so the delta reflects burden rather than between-patient variation. BDA further separates the selected direction from its reverse, anchors it to the observed higher-burden endpoint, and enforces consistency across ordered triplets. Since every pair is drawn within a single pathology, BDA is designed to constrain intra-class structure that image-report contrast alone never touches. Spectrum attains 85.6 zero-shot AUROC on CT-RATE and 72.7 on external RAD-ChestCT, with consistent gains in linear probing and retrieval. Weak cross-patient order is thus a scalable complement to anatomy-aware correspondence, yielding burden-aware CT representations without longitudinal data.

Related papers

• Big, Bright, or Invisible: A Frozen-Feature Benchmark of 3D CT Foundation Models [30.102620002840904]
  3D CT foundation models could assist routine interpretation by providing generalizable representations of anatomy and pathology.<n>We benchmark ten frozen CT encoders across three cohorts of thoracic CT scans.<n>We find no universal state-of-the-art, with rankings fluctuating significantly depending on the evaluation context.
  arXiv  Detail & Related papers  (2026-08-06T12:34:17Z)
• Rad-JEPA 3D: Radiology Joint-Embedding Predictive Model for 3D Computed Tomography [10.665332135327205]
  We introduce Rad-JEPA 3D, a joint-embedding predictive framework that learns volumetric CT representations.<n>To improve the quality of intermediate representations, we propose Hidden States Orthogonal Regularization (HSOR)<n>HSOR aligns student-teacher hidden states and reduces feature redundancy throughout the encoder.
  arXiv  Detail & Related papers  (2026-07-28T19:00:17Z)
• Fine-Grained Vision-Language Pretraining with Organ-Conditioned Pattern Tokens for CT Understanding [2.9220076568786335]
  OCP-CT is an organ-conditioned pattern-token alignment framework for CT vision-language pretraining.<n>It preserves a stable global CT-report contrastive branch and introduces an organ pattern interface.<n>On the publicly available CT-RATE and RAD-ChestCT benchmarks, OCP-CT achieves average AUROCs of 84.5% and 69.9% for zero-shot abnormality diagnosis.
  arXiv  Detail & Related papers  (2026-07-15T14:36:43Z)
• MAC-XA: Multi-view Anatomy-Correspondence Fusion for Coronary Stenosis Reporting from X-ray Angiography [46.98348758460767]
  Multi-view reasoning in coronary X-ray angiography is inherently a cross-projection geometric problem.<n>We introduce a controllable synthetic angiography generation strategy to expose geometry-derived patch-level correspondence supervision.<n>Experiments on synthetic data and zero-shot transfer to real angiograms show that this alignment-constrained design improves correspondence consistency and structured stenosis reporting.
  arXiv  Detail & Related papers  (2026-07-07T13:33:25Z)
• Patient-Aware Contrastive Learning Preserves Per-Patient Structure in RR-Interval Representations [0.0]
  We study the problem in the setting of Paroxysmal Atrial Fibrillation(PAF) detection from RR-interval(RRI) sequences.<n>We propose a patient-aware contrastive objective that forms positive pairs only from same-patient, same-class segments.
  arXiv  Detail & Related papers  (2026-06-22T16:35:18Z)
• Disentangled Anatomy-Disease Diffusion (DADD) for Controllable Ulcerative Colitis Progression Synthesis [4.554894288663752]
  We address the challenge of Synthesizing longitudinal medical images at controllable disease stages.<n>Our framework, Disentangled Anatomy Diffusion, conditions a latent diffusion model on two complementary embeddings.<n>We introduce a Feature, a cross-attention-based erasure mechanism that identifies and suppresses disease-related channels.<n>Our approach produces high-fidelity images across all severity levels and effectively rebalances skewed class distributions.
  arXiv  Detail & Related papers  (2026-05-03T12:35:12Z)
• PHASOR: Anatomy- and Phase-Consistent Volumetric Diffusion for CT Virtual Contrast Enhancement [60.81732730684265]
  This paper introduces PHASOR, a volumetric diffusion framework for high-fidelity virtual contrast enhancement (VCE)<n>By treating CT volumes as coherent sequences, we leverage a video diffusion model to enhance structural coherence and volumetric accuracy.<n>To ensure anatomy-phase consistent synthesis, we introduce two complementary modules. First, anatomy-routed mixture-of-experts (AR-MoE) anchors distinct enhancement patterns to anatomical semantics, with organ-specific memory to capture salient details.<n>Second, intensity-phase aware representation alignment (IP-REPA) highlights intricate contrast signals while mitigating the impact of imperfect spatial alignment.
  arXiv  Detail & Related papers  (2026-04-01T15:57:18Z)
• DMD-augmented Unpaired Neural Schrödinger Bridge for Ultra-Low Field MRI Enhancement [45.02203416553436]
  As paired 64 mT - 3 T scans are scarce, we propose an unpaired 64 mT $rightarrow$ 3 T translation framework.<n>Our method builds upon the Unpaired Neural Schrdinge Bridge (UNSB) with multi-step refinement.
  arXiv  Detail & Related papers  (2026-03-04T06:26:52Z)
• Hide-and-Seek Attribution: Weakly Supervised Segmentation of Vertebral Metastases in CT [68.09387763135236]
  We introduce a weakly supervised method trained solely on vertebra-level healthy/malignant labels, without any lesion masks.<n>We achieve strong blastic/lytic performance despite no mask supervision.
  arXiv  Detail & Related papers  (2025-12-07T14:03:28Z)
• Thoracic Cartilage Ultrasound-CT Registration using Dense Skeleton Graph [49.11220791279602]
  It is challenging to accurately map planned paths from a generic atlas to individual patients, particularly for thoracic applications. A graph-based non-rigid registration is proposed to enable transferring planned paths from the atlas to the current setup.
  arXiv  Detail & Related papers  (2023-07-07T18:57:21Z)
• Pseudo-Label Guided Multi-Contrast Generalization for Non-Contrast Organ-Aware Segmentation [21.792652078631868]
  We propose an unsupervised approach to compute non-contrast segmentation without ground-truth label. Unlike generative adversarial approaches, we compute the pairwise morphological context with CECT to provide teacher guidance. We validate our approach on multi-organ segmentation with paired non-contrast & contrast-enhanced CT scans.
  arXiv  Detail & Related papers  (2022-05-12T06:36:33Z)
• Building Brains: Subvolume Recombination for Data Augmentation in Large Vessel Occlusion Detection [56.67577446132946]
  A large training data set is required for a standard deep learning-based model to learn this strategy from data. We propose an augmentation method that generates artificial training samples by recombining vessel tree segmentations of the hemispheres from different patients. In line with the augmentation scheme, we use a 3D-DenseNet fed with task-specific input, fostering a side-by-side comparison between the hemispheres.
  arXiv  Detail & Related papers  (2022-05-05T10:31:57Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.