Scaling Generative Foundation Models for Chest Radiography with Rectified Flow Transformers

• URL: http://arxiv.org/abs/2606.19460v1
• Date: Wed, 17 Jun 2026 18:01:18 GMT
• Title: Scaling Generative Foundation Models for Chest Radiography with Rectified Flow Transformers
• Abstract summary: Controlled, high-fidelity synthesis of chest radiographs is a promising path toward diversifying clinical datasets.<n>We present the largest specialist generative foundation model for chest radiographs to date, with over 1.3B parameters, trained for 1.6T tokens on a curated, heterogeneous dataset.<n>Our model supports controllable radiograph generation and editing across multiple demographic subgroups, acquisition views, and a dozen pathologies.
• Score: 17.58664012517527
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We introduce the first generative foundation model for chest radiograph synthesis trained from scratch at the billion-parameter scale. Existing radiographic AI models often suffer from poor generalisation across patient subpopulations, institutions, and acquisition settings, resulting in limited real-world clinical utility. Controlled, high-fidelity synthesis of chest radiographs is a promising path toward diversifying clinical datasets and evaluating the robustness of diagnostic models. Therefore, we present the largest specialist generative foundation model for chest radiographs to date, with over 1.3B parameters, trained for 1.6T tokens on a curated, heterogeneous dataset comprising 1.2M radiographs and clinical expert-guided metadata. Our model supports controllable radiograph generation and editing across multiple demographic subgroups, acquisition views, and a dozen pathologies. Moreover, we significantly advance the state of the art in radiograph synthesis fidelity, producing images that are indistinguishable from real radiographs to clinical experts.

Related papers

• Vision Foundation Models in Radiology: A Scoping Review of Data, Methodology, Evaluation and Clinical Translation [35.18016233072556]
  Vision foundation models (VFMs) are increasingly being developed for radiological imaging, yet their definition, development and evaluation remain heterogeneous.<n>We conducted a PRISMAScR scoping review of peer-reviewed studies describing foundation models trained exclusively on radiological imaging data.<n> Transformer-based architectures and self-supervised pretraining predominated, particularly masked image modeling, contrastive learning and multi-stage approaches.<n>Overall, radiology-specific VFMs show promising transferability, but clinical translation remains constrained by limited data representativeness, heterogeneous benchmarks, incomplete reporting and insufficient deployment-oriented evaluation.
  arXiv  Detail & Related papers  (2026-07-08T10:01:33Z)
• DiffusionXRay: A Diffusion and GAN-Based Approach for Enhancing Digitally Reconstructed Chest Radiographs [0.13048920509133807]
  We introduce DiffusionXRay, a novel image restoration pipeline for Chest X-ray images.<n>We train a DDPM-based model on paired low-quality and high-quality images, enabling it to learn the nuances of X-ray image restoration.<n>Our method demonstrates promising results in enhancing image clarity, contrast, and overall diagnostic value of chest X-rays.
  arXiv  Detail & Related papers  (2026-03-02T10:14:50Z)
• Curia: A Multi-Modal Foundation Model for Radiology [3.5025024631649857]
  We introduce Curia, a foundation model trained on the entire cross-sectional imaging output of a major hospital.<n>Curia accurately identifies organs, detects conditions like brain hemorrhages and myocardial infarctions, and predicts outcomes in tumor staging.
  arXiv  Detail & Related papers  (2025-09-08T16:04:12Z)
• A Generative Foundation Model for Chest Radiography [43.32039823252467]
  ChexGen is a generative vision-language foundation model that introduces a unified framework for text-, mask-, and bounding box-guided synthesis of chest radiographs.<n>ChexGen was pretrained on the largest curated chest X-ray dataset to date, consisting of 960,000 radiograph-report pairs.
  arXiv  Detail & Related papers  (2025-09-04T05:53:58Z)
• ChatRadio-Valuer: A Chat Large Language Model for Generalizable Radiology Report Generation Based on Multi-institution and Multi-system Data [115.0747462486285]
  ChatRadio-Valuer is a tailored model for automatic radiology report generation that learns generalizable representations. The clinical dataset utilized in this study encompasses a remarkable total of textbf332,673 observations. ChatRadio-Valuer consistently outperforms state-of-the-art models, especially ChatGPT (GPT-3.5-Turbo) and GPT-4 et al.
  arXiv  Detail & Related papers  (2023-10-08T17:23:17Z)
• CT Reconstruction from Few Planar X-rays with Application towards Low-resource Radiotherapy [20.353246282326943]
  We propose a method to generate CT volumes from few (5) planar X-ray observations using a prior data distribution. To focus the generation task on clinically-relevant features, our model can also leverage anatomical guidance during training. Our method is better than recent sparse CT reconstruction baselines in terms of standard pixel and structure-level metrics.
  arXiv  Detail & Related papers  (2023-08-04T01:17:57Z)
• Revisiting Computer-Aided Tuberculosis Diagnosis [56.80999479735375]
  Tuberculosis (TB) is a major global health threat, causing millions of deaths annually. Computer-aided tuberculosis diagnosis (CTD) using deep learning has shown promise, but progress is hindered by limited training data. We establish a large-scale dataset, namely the Tuberculosis X-ray (TBX11K) dataset, which contains 11,200 chest X-ray (CXR) images with corresponding bounding box annotations for TB areas. This dataset enables the training of sophisticated detectors for high-quality CTD.
  arXiv  Detail & Related papers  (2023-07-06T08:27:48Z)
• Medical Image Captioning via Generative Pretrained Transformers [57.308920993032274]
  We combine two language models, the Show-Attend-Tell and the GPT-3, to generate comprehensive and descriptive radiology records. The proposed model is tested on two medical datasets, the Open-I, MIMIC-CXR, and the general-purpose MS-COCO.
  arXiv  Detail & Related papers  (2022-09-28T10:27:10Z)
• Image Synthesis with Disentangled Attributes for Chest X-Ray Nodule Augmentation and Detection [52.93342510469636]
  Lung nodule detection in chest X-ray (CXR) images is common to early screening of lung cancers. Deep-learning-based Computer-Assisted Diagnosis (CAD) systems can support radiologists for nodule screening in CXR. To alleviate the limited availability of such datasets, lung nodule synthesis methods are proposed for the sake of data augmentation.
  arXiv  Detail & Related papers  (2022-07-19T16:38:48Z)
• Generative Residual Attention Network for Disease Detection [51.60842580044539]
  We present a novel approach for disease generation in X-rays using a conditional generative adversarial learning. We generate a corresponding radiology image in a target domain while preserving the identity of the patient. We then use the generated X-ray image in the target domain to augment our training to improve the detection performance.
  arXiv  Detail & Related papers  (2021-10-25T14:15:57Z)
• 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)
• Computer-aided abnormality detection in chest radiographs in a clinical setting via domain-adaptation [0.23624125155742057]
  Deep learning (DL) models are being deployed at medical centers to aid radiologists for diagnosis of lung conditions from chest radiographs. These pre-trained DL models' ability to generalize in clinical settings is poor because of the changes in data distributions between publicly available and privately held radiographs. In this work, we introduce a domain-shift detection and removal method to overcome this problem.
  arXiv  Detail & Related papers  (2020-12-19T01:01:48Z)

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.