HarMoE: Multi-Source Chest Radiograph Pretraining with Dataset-Disentangled Experts

• URL: http://arxiv.org/abs/2608.02252v1
• Date: Mon, 03 Aug 2026 14:00:16 GMT
• Title: HarMoE: Multi-Source Chest Radiograph Pretraining with Dataset-Disentangled Experts
• Abstract summary: HarMoE is a dataset-aware mixture-of-experts framework that learns shared cross-dataset medical semantics.<n> Experiments on large-scale chest X-ray benchmarks show that HarMoE consistently improves zero-shot classification transfer.
• Score: 6.917612472314974
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent vision-language models for chest X-ray understanding are largely built on image-report alignment and therefore rely heavily on MIMIC-CXR as the dominant pretraining source. While effective at scale, this paradigm underexplores an important alternative source of supervision: a range of existing multi-label classification datasets, which provide cleaner and more explicit disease signals than free-text reports, and can offer broader pathology coverage when combined across sources. However, learning from such heterogeneous datasets is nontrivial, as differences in label ontologies, annotation protocols, acquisition pipelines, and report styles can cause models to entangle clinical semantics with dataset identity, leading to poor transfer despite increased scale. In this work, we revisit radiology VLM construction from the perspective of harmonized multi-source learning. We propose HarMoE, a dataset-aware mixture-of-experts framework that learns shared cross-dataset medical semantics while confining source-specific variation to lightweight residual experts in deeper decoder layers. To further exploit clean supervision from labeled datasets, we train in a unified disease vocabulary with masked multi-dataset supervision, enabling the model to leverage complementary annotations without introducing false negatives. Experiments on large-scale chest X-ray benchmarks show that HarMoE consistently improves zero-shot classification, out-of-distribution transfer, and grounding over strong baselines. Our results suggest that building robust radiology VLMs requires moving beyond single-source image-report alignment toward structured knowledge construction from heterogeneous datasets with cleaner supervision and broader coverage. Code and the 873k harmonized dataset will be released at https://github.com/Roypic/harmoe.

Related papers

• AnchorDiff: Topology-Aware Masked Diffusion with Confidence-based Rewriting for Radiology Report Generation [2.2748974006378933]
  Radiology report generation aims to automatically produce clinically accurate textual reports from medical images.<n>Existing methods rely on autoregressive (AR) language models, whose causal dependency structure restricts generation to a unidirectional left-to-right process.<n>We propose AnchorDiff, the first masked-diffusion framework for RRG that integrates knowledge-graph-derived clinical anchors into diffusion language modeling.
  arXiv  Detail & Related papers  (2026-05-16T16:42:43Z)
• Adapting HFMCA to Graph Data: Self-Supervised Learning for Generalizable fMRI Representations [57.054499278843856]
  Functional magnetic resonance imaging (fMRI) analysis faces significant challenges due to limited dataset sizes and domain variability between studies.<n>Traditional self-supervised learning methods inspired by computer vision often rely on positive and negative sample pairs.<n>We propose adapting a recently developed Hierarchical Functional Maximal Correlation Algorithm (HFMCA) to graph-structured fMRI data.
  arXiv  Detail & Related papers  (2025-10-05T12:35:01Z)
• Exploring the Capabilities of LLM Encoders for Image-Text Retrieval in Chest X-rays [8.019362739504087]
  Vision-language pretraining has advanced image-text alignment, yet progress in radiology remains constrained by the heterogeneity of clinical reports.<n>We ask whether large language model (LLM) encoders can provide robust clinical representations that transfer across diverse styles.<n>We introduce LLM2VEC4CXR, a domain-adapted encoder for chest X-ray reports, and LLM2CLIP4CXR, a dual-tower framework that couples this encoder with a vision backbone.
  arXiv  Detail & Related papers  (2025-09-17T09:44:59Z)
• SimCroP: Radiograph Representation Learning with Similarity-driven Cross-granularity Pre-training [25.763109982379703]
  We propose a Similarity-Driven Cross-Granularity Pre-training framework on chest CTs.<n>It combines similarity-driven alignment and cross-granularity fusion to improve radiograph interpretation.<n>SimCroP is pre-trained on a large-scale paired CT-reports dataset and validated on image classification and segmentation tasks.
  arXiv  Detail & Related papers  (2025-09-10T06:20:53Z)
• Privacy-Preserving Chest X-ray Report Generation via Multimodal Federated Learning with ViT and GPT-2 [0.1874930567916036]
  The study proposes a Multimodal Federated Learning framework for chest X-ray report generation using the IU-Xray dataset.<n>The system utilizes a Vision Transformer (ViT) as the encoder and GPT-2 as the report generator, enabling decentralized training without sharing raw data.<n>The results show that FL can match or surpass centralized models in generating clinically relevant and semantically rich radiology reports.
  arXiv  Detail & Related papers  (2025-05-27T20:01:12Z)
• Source-Free Collaborative Domain Adaptation via Multi-Perspective Feature Enrichment for Functional MRI Analysis [55.03872260158717]
  Resting-state MRI functional (rs-fMRI) is increasingly employed in multi-site research to aid neurological disorder analysis. Many methods have been proposed to reduce fMRI heterogeneity between source and target domains. But acquiring source data is challenging due to concerns and/or data storage burdens in multi-site studies. We design a source-free collaborative domain adaptation framework for fMRI analysis, where only a pretrained source model and unlabeled target data are accessible.
  arXiv  Detail & Related papers  (2023-08-24T01:30:18Z)
• Cross-Modal Causal Intervention for Medical Report Generation [107.76649943399168]
  Radiology Report Generation (RRG) is essential for computer-aided diagnosis and medication guidance.<n> generating accurate lesion descriptions remains challenging due to spurious correlations from visual-linguistic biases.<n>We propose a two-stage framework named CrossModal Causal Representation Learning (CMCRL)<n> Experiments on IU-Xray and MIMIC-CXR show that our CMCRL pipeline significantly outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-03-16T07:23:55Z)
• Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training [23.506879497561712]
  We employ a contrastive global-local dual-encoder architecture to learn concepts directly from unstructured medical reports. We evaluate our approach on the large-scale chest X-Ray datasets MIMIC-CXR, CheXpert, and ChestX-Ray14 for disease classification.
  arXiv  Detail & Related papers  (2022-05-14T21:44:05Z)
• Many-to-One Distribution Learning and K-Nearest Neighbor Smoothing for Thoracic Disease Identification [83.6017225363714]
  deep learning has become the most powerful computer-aided diagnosis technology for improving disease identification performance. For chest X-ray imaging, annotating large-scale data requires professional domain knowledge and is time-consuming. In this paper, we propose many-to-one distribution learning (MODL) and K-nearest neighbor smoothing (KNNS) methods to improve a single model's disease identification performance.
  arXiv  Detail & Related papers  (2021-02-26T02:29:30Z)
• G-MIND: An End-to-End Multimodal Imaging-Genetics Framework for Biomarker Identification and Disease Classification [49.53651166356737]
  We propose a novel deep neural network architecture to integrate imaging and genetics data, as guided by diagnosis, that provides interpretable biomarkers. We have evaluated our model on a population study of schizophrenia that includes two functional MRI (fMRI) paradigms and Single Nucleotide Polymorphism (SNP) data.
  arXiv  Detail & Related papers  (2021-01-27T19:28:04Z)
• Learning Invariant Feature Representation to Improve Generalization across Chest X-ray Datasets [55.06983249986729]
  We show that a deep learning model performing well when tested on the same dataset as training data starts to perform poorly when it is tested on a dataset from a different source. By employing an adversarial training strategy, we show that a network can be forced to learn a source-invariant representation.
  arXiv  Detail & Related papers  (2020-08-04T07:41:15Z)
• Deep Mining External Imperfect Data for Chest X-ray Disease Screening [57.40329813850719]
  We argue that incorporating an external CXR dataset leads to imperfect training data, which raises the challenges. We formulate the multi-label disease classification problem as weighted independent binary tasks according to the categories. Our framework simultaneously models and tackles the domain and label discrepancies, enabling superior knowledge mining ability.
  arXiv  Detail & Related papers  (2020-06-06T06:48:40Z)

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.