The paper presents the first systematic evaluation of out‑of‑distribution generalization for congenital heart disease (CHD) segmentation, using the ImageCHD cohort as a held‑out target. It compares several segmentation architectures under different training regimes, showing that in‑distribution performance is a poor predictor of cross‑cohort robustness: nnU‑Net drops from 0.77 to 0.51 Dice, while SwinUNETR maintains higher performance at 0.67 Dice. Limited target‑domain adaptation with only 11 labeled ImageCHD cases boosts all SwinUNETR variants above 0.76 Dice, highlighting the importance of explicit cross‑dataset testing.
By Aniketh Vijesh, Shrisharanyan Vasu, Abhijit Ramesh, Clare Pomeroy-Ward, Harikrishnan Anil Maya, Sarin Xavier, Mahesh Kappanayil, Gilad Gressel
The study benchmarks foundation models for chest X‑ray classification across a domain shift from adult to pediatric patients. Six foundation models and two CNN baselines were evaluated on a harmonized five‑label task using adult data for adaptation and tested on both adult and pediatric sets. Results show all models experience a substantial performance drop—over 10 percentage points—when applied to pediatric data, and deeper adaptation does not markedly improve cross‑population generalization.
By Sara In\'acio, Hugo Proen\c{c}a, Jo\~ao C. Neves
AnyCXR is a unified framework that performs multi-organ segmentation on chest X-rays taken at any projection angle using only synthetic data. It employs a Multi-stage Domain Randomization engine to generate over 100,000 anatomically accurate synthetic radiographs from 3D CT volumes, and a Conditional Joint Annotation Regularization strategy to learn from partial, imperfect labels by enforcing anatomical consistency in a latent space. Trained solely on synthetic data, AnyCXR achieves strong zero-shot generalization on real-world datasets, accurately delineating 54 anatomical structures across PA, lateral, and oblique views, and supports downstream clinical tasks such as cardiothoracic ratio estimation, spine curvature assessment, and disease classification.
By Zifei Dong, Wenjie Wu, Jinkui Hao, Tianqi Chen, Ziqiao Weng, Bo Zhou
arXiv:2606. 16991v1 Announce Type: cross Abstract: Multiphasic contrast-enhanced CT (CECT) is widely used for abdominal lesion characterization, yet it carries inherent risks of contrast-induced nephropathy, escalates acquisition burden, and heavily contributes to radiologist workload.
By Mariam Elbakry, Aliaa Sayed Sheha, Salma Hassan Tantawy, Aya Yassin, Concetto Spampinato, Karim Lekadir, Xiaomeng Li, Marawan Elbatel
Congenital heart disease (CHD) diagnosis and surgical planning often require patient-specific 3D anatomical models, but manual segmentation is labor-intensive, particularly in complex anatomies. Altho...
arXiv:2606. 19460v1 Announce Type: cross Abstract: We introduce the first generative foundation model for chest radiograph synthesis trained from scratch at the billion-parameter scale.
By Fabio De Sousa Ribeiro, Emma A. M. Stanley, Charles Jones, Tian Xia, Dominic C. Marshall, Laurent Renard Trich\'e, Christopher V. Cosgriff, Panagiotis Dimitrakopoulos, Sotirios A. Tsaftaris, Ben Glocker
arXiv:2606. 02035v1 Announce Type: new Abstract: Medical imaging interpretation is a foundational pillar of modern clinical diagnostics, yet the manual generation of radiology reports remains a time-consuming process prone to interpretation inconsistencies.
By Yogesh Kumar Meena, Saurabh Agarwal, K. V. Arya
The paper introduces an uncertainty‑aware clinical knowledge graph for chest X‑ray device reasoning, capturing device instances, tip estimates, placement assessments, provenance, report events, and temporal links as interconnected evidence. The graph builder processes 30,083 studies from 3,255 patients, producing 914,632 evidence nodes and 884,549 typed relationships, while preserving detailed uncertainty and provenance information for each predicted device. The authors also outline typed data contracts, uncertainty representations, abstention rules, report‑image grounding, and longitudinal query mechanisms, though the current analysis is post‑hoc descriptive and does not yet demonstrate clinical utility.
By Harshil Lodhiya
arXiv:2608.28455v1 Announce Type: new
Abstract: Contrastive vision-language learning uses paired chest CT volumes and radiology reports to learn abnormality classifiers without manually annotated lab...
By Huseyin Umut Isik, Mehmet Alp Ozaydin, Sila Kurugol, \c{S}eyda Ertekin
Meta-CLIP is a metadata‑conditioned contrastive learning framework that incorporates DICOM acquisition information to reweight cross‑modal negatives during medical vision‑language pretraining. By defining a bounded pairwise margin over numerical and categorical DICOM fields, the method emphasizes distinctions among studies with similar imaging conditions while keeping the kernel fixed during optimization. Evaluations on chest radiography and volumetric CT show that Meta‑CLIP improves zero‑shot classification performance, achieving higher macro‑AUROC scores than baseline CLIP models on multiple benchmarks.
By Haozhe Luo, Nirosh Sivanesan, Shelley Zixin Shu, Francesco Nitti, Mauricio Reyes
arXiv:2409.16183v2 Announce Type: replace
Abstract: Radiology is a vital and complex component of modern clinical workflow and covers many tasks. Recently, vision-language (VL) foundation models in m...
By Xiaohong Liu, Guoxing Yang, Yulin Luo, Jiaji Mao, Xiang Zhang, Haibo Wang, Zhiyang He, Ming Gao, Shanghang Zhang, Jun Shen, Guangyu Wang
ICGF-CXRMate is a framework that compresses heterogeneous patient data into a fixed set of clinical context tokens conditioned on chest X‑ray images. These tokens are selectively fused with visual features via gated cross‑attention and also retained in the decoder prompt, yielding shorter prompts and higher report quality. Experiments on MIMIC‑CXR and MIMIC‑IV‑ED show significant gains in RadGraph‑F1, CXR‑BERT similarity, and BLEU‑4, while reducing prompt length by 22.9%.
By Vu Minh Tran, Thuc Huynh, Son Thanh Luu, Doanh C. Bui