5 papers
Solving the Needle-in-a-Haystack Problem in Mammography Vision-Language Model with Differentiable Subset Sampling
Young Seok Jeon, Beatrice Brown-Mulry, Rohan Satya Isaac +5
There is growing interest in adopting CLIP-style vision--language model (VLM) pretraining for mammography. However, models that directly employ the standard CLIP architecture and t…
Positive-Unlabeled Preference Optimization For Chest X-ray Report Generation
Yuta Kobayashi, Pradyun Ramesh, Muhammad Ahmed Chaudhry +5
Vision-Language Models (VLMs) for radiology report generation are typically trained on retrospective clinical reports, which suffer from omission noise: clinically present findings…
RadHarmony: Radiological Data Handling in the Era of Agentic AI
Frank Li, Bardia Khosravi, Mohammadreza Chavoshi +5
Training deep learning models on radiological images requires integrating heterogeneous datasets across different sources, file formats, directory layouts, label schemas, and annot…
Frozen Foundation-Model Embeddings Discard Small-Lesion Signal in Chest Radiography: Implications for Pre-Deployment Evaluation
Raajitha Muthyala, Zhenan Yin, Alekhya Jilla +6
Frozen vision-transformer (ViT) foundation-model embeddings increasingly serve as the substrate for downstream chest-radiography (CXR) pipelines, yet where small-scale, low-contras…
MultiMedVision: Multi-Modal Medical Vision Framework
Frank Li, Bardia Khosravi, Mohammadreza Chavoshi +5
Multi-modal medical imaging enables comprehensive diagnostics, yet current foundation models process 2D (e.g. X-ray) and 3D (e.g. CT) data with separate, dimensionality-specific ar…