CIV-DG: Conditional Instrumental Variables for Domain Generalization in Medical Imaging

• URL: http://arxiv.org/abs/2603.25202v1
• Date: Thu, 26 Mar 2026 09:07:08 GMT
• Title: CIV-DG: Conditional Instrumental Variables for Domain Generalization in Medical Imaging
• Abstract summary: Cross-site generalizability in medical AI is compromised by selection bias.<n>We propose CIV-DG, a causal framework that disentangles pathological semantics from scanner-induced artifacts.<n>Experiments on the Camelyon17 benchmark and large-scale Chest X-Ray datasets demonstrate that CIV-DG significantly outperforms leading baselines.
• Score: 28.44504764194173
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Cross-site generalizability in medical AI is fundamentally compromised by selection bias, a structural mechanism where patient demographics (e.g., age, severity) non-randomly dictate hospital assignment. Conventional Domain Generalization (DG) paradigms, which predominantly target image-level distribution shifts, fail to address the resulting spurious correlations between site-specific variations and diagnostic labels. To surmount this identifiability barrier, we propose CIV-DG, a causal framework that leverages Conditional Instrumental Variables to disentangle pathological semantics from scanner-induced artifacts. By relaxing the strict random assignment assumption of standard IV methods, CIV-DG accommodates complex clinical scenarios where hospital selection is endogenously driven by patient demographics. We instantiate this theory via a Deep Generalized Method of Moments (DeepGMM) architecture, employing a conditional critic to minimize moment violations and enforce instrument-error orthogonality within demographic strata. Extensive experiments on the Camelyon17 benchmark and large-scale Chest X-Ray datasets demonstrate that CIV-DG significantly outperforms leading baselines, validating the efficacy of conditional causal mechanisms in resolving structural confounding for robust medical AI.

Related papers

• Dementia Etiology Diagnosis via Collaborative Meta Knowledge Enhancement [81.27196033392559]
  We propose a Collaborative Meta Knowledge Enhancement (COME) framework for dementia etiology diagnosis.<n>COME injects multi-center acquisition semantics, source identifiers, and modality indicators as heterogeneity-aware embeddings into a unified Transformer architecture for scale-up training.<n>Our method achieves state-of-the-art in-domain performance with a mean macro-averaged AUC of 85.62% and a 4.29-point gain over the strongest baseline.
  arXiv  Detail & Related papers  (2026-07-24T04:26:47Z)
• CIPHER: Causal Intervention Pathways for Healthcare Equity and Robustness [12.8473199678431]
  We formalize the medical image formation process via a structural causal model.<n>We introduce CIPHER, a framework designed to intervene on all four causal paths.<n>By employing a multi-pathway intervention strategy, our model reduced worst-group disparities by an average of 35.8%.
  arXiv  Detail & Related papers  (2026-07-01T09:36:53Z)
• Towards a Virtual Neuroscientist: Autonomous Neuroimaging Analysis via Multi-Agent Collaboration [53.772014300855375]
  We introduce NIAgent, a multi-agent system for autonomous end-to-end neuroimaging analysis.<n>Unlike conventional flat tool-calling agents, NIAgent adopts a code-centric execution paradigm.<n>We propose a hierarchical verification framework for autonomous quality control, integrating cohort-level metric screening with agentic visual inspection.
  arXiv  Detail & Related papers  (2026-05-10T06:30:19Z)
• Graph-Augmented Topological Internalization with Dual-Stream Classifiers for Medical Report Generation [3.3887144970801555]
  We propose a Graph-Augmented Dual-Stream Medical Report Generation with Topological Internalization, GDMRG.<n>Our framework introduces a Topological Knowledge Internalization module, TKI, which generates an explicit parameterized weight matrix based on global disease co-occurrence priors.<n>Experiments on the MIMIC-CXR dataset demonstrate that GDMRG achieves competitive clinical efficacy, CE, while maintaining natural language fluency.
  arXiv  Detail & Related papers  (2026-05-04T09:17:27Z)
• Clinical Cognition Alignment for Gastrointestinal Diagnosis with Multimodal LLMs [63.535652574541764]
  Multimodal Large Language Models (MLLMs) have demonstrated remarkable potential in medical image analysis.<n>Their application in gastrointestinal endoscopy is currently hindered by two critical limitations.<n>We propose a novel Clinical-Cognitive-Aligned (CogAlign) framework to address these challenges.
  arXiv  Detail & Related papers  (2026-03-21T07:47:37Z)
• AngioDG: Interpretable Channel-informed Feature-modulated Single-source Domain Generalization for Coronary Vessel Segmentation in X-ray Angiography [3.515262079463771]
  Cardiovascular diseases are the leading cause of death globally.<n>X-ray Coronary Angiography (XCA) is the gold standard during real-time cardiac interventions.<n>Generalizable vessel segmentation models for XCA is challenging due to variations in imaging protocols and patient demographics.
  arXiv  Detail & Related papers  (2025-11-21T19:21:21Z)
• Single Domain Generalization in Diabetic Retinopathy: A Neuro-Symbolic Learning Approach [0.5581472054346949]
  KG-DG is a neuro-symbolic framework for diabetic retinopathy (DR) classification.<n>It integrates vision transformers with expert-guided symbolic reasoning to enable generalization across unseen domains.
  arXiv  Detail & Related papers  (2025-09-03T01:07:49Z)
• UdonCare: Hierarchy Pruning for Unseen Domain Discovery in Predictive Healthcare [8.077539743672732]
  We propose a hierarchy-guided method to divide patients into latent domains that decomposes domain-inlabel information from patient data.<n>Our method identifies patient domains by pruning medical hierarchy (e.g. ICD-9-IV-CM)<n>On two public datasets, MIMIC-III and MIMIC-III, UdonCare shows superiority across four clinical prediction tasks with substantial domain gaps.
  arXiv  Detail & Related papers  (2025-06-08T03:20:34Z)
• Adaptable Cardiovascular Disease Risk Prediction from Heterogeneous Data using Large Language Models [70.64969663547703]
  AdaCVD is an adaptable CVD risk prediction framework built on large language models extensively fine-tuned on over half a million participants from the UK Biobank.<n>It addresses key clinical challenges across three dimensions: it flexibly incorporates comprehensive yet variable patient information; it seamlessly integrates both structured data and unstructured text; and it rapidly adapts to new patient populations using minimal additional data.
  arXiv  Detail & Related papers  (2025-05-30T14:42:02Z)
• How Well Does GPT-4V(ision) Adapt to Distribution Shifts? A Preliminary Investigation [90.93999543169296]
  GPT-4V acts as the most advanced publicly accessible multimodal foundation model. This study rigorously evaluates GPT-4V's adaptability and generalization capabilities in dynamic environments.
  arXiv  Detail & Related papers  (2023-12-12T16:48:07Z)
• Instrumental Variable Learning for Chest X-ray Classification [52.68170685918908]
  We propose an interpretable instrumental variable (IV) learning framework to eliminate the spurious association and obtain accurate causal representation. Our approach's performance is demonstrated using the MIMIC-CXR, NIH ChestX-ray 14, and CheXpert datasets.
  arXiv  Detail & Related papers  (2023-05-20T03:12:23Z)
• An Adversarial Domain Separation Framework for Septic Shock Early Prediction Across EHR Systems [7.058760708627898]
  We propose a general domain adaptation (DA) framework that tackles two categories of discrepancies in EHRs collected from different medical systems. We evaluate our framework for early diagnosis of an extremely challenging condition, septic shock, using two real-world EHRs from distinct medical systems in the U.S.
  arXiv  Detail & Related papers  (2020-10-26T23:41:33Z)

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.