# CheXanatomy: Anatomy-Aware Vision-Language Modeling for Chest Radiographs

arXiv: https://arxiv.org/abs/2606.08420 · HTML: https://mlsanity.com/paper/2606.08420/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 7 Jun 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 19 of 19)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Gemma 2: Improving Open Language Models at a Practical Size](https://mlsanity.com/paper/2408.00118/) (2024)
- [PaliGemma: A versatile 3B VLM for transfer](https://mlsanity.com/paper/2407.07726/) (2024)
- [Contrastive Learning of Medical Visual Representations from Paired Images and Text](https://mlsanity.com/paper/2010.00747/) (2020)
- [Image Segmentation Using Text and Image Prompts](https://mlsanity.com/paper/2112.10003/) (2021)
- [Pix2seq: A Language Modeling Framework for Object Detection](https://mlsanity.com/paper/2109.10852/) (2021)
- [CLIP-Driven Universal Model for Organ Segmentation and Tumor Detection](https://mlsanity.com/paper/2301.00785/) (2023)
- [Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography](https://mlsanity.com/paper/2403.17834/) (2024)
- [MedCLIP-SAMv2: Towards Universal Text-Driven Medical Image Segmentation](https://mlsanity.com/paper/2409.19483/) (2024)
- [Fast Auto-Differentiable Digitally Reconstructed Radiographs for Solving Inverse Problems in Intraoperative Imaging](https://mlsanity.com/paper/2208.12737/) (2022)
- [CheXmask: a large-scale dataset of anatomical segmentation masks for multi-center chest x-ray images](https://mlsanity.com/paper/2307.03293/) (2023)
- [Detailed Annotations of Chest X-Rays via CT Projection for Report Understanding](https://mlsanity.com/paper/2210.03416/) (2022)
- [Accurate Fine-Grained Segmentation of Human Anatomy in Radiographs via Volumetric Pseudo-Labeling](https://mlsanity.com/paper/2306.03934/) (2023)
- [VinDr-RibCXR: A Benchmark Dataset for Automatic Segmentation and Labeling of Individual Ribs on Chest X-rays](https://mlsanity.com/paper/2107.01327/) (2021)
- [Shadow and Light: Digitally Reconstructed Radiographs for Disease Classification](https://mlsanity.com/paper/2406.03688/) (2024)
- [AnyCXR: Human Anatomy Segmentation of Chest X-ray at Any Acquisition Position using Multi-stage Domain Randomized Synthetic Data with Imperfect Annotations and Conditional Joint Annotation Regularization Learning](https://mlsanity.com/paper/2512.17263/) (2025)

## Abstract (arXiv)

Vision-language models (VLMs) pretrained on large-scale image-text pairs demonstrate strong image-level understanding, but are primarily optimized for global alignment and do not explicitly encode fine-grained anatomical structure, limiting their suitability for spatially precise tasks such as segmentation. We introduce CheXanatomy, a framework that integrates explicit anatomical knowledge into a pretrained VLM through autoregressive token-space supervision. Instead of adding task-specific decoder heads, the model is trained to generate anatomical segmentation masks via next-token prediction. To enable scalable supervision, we synthesize realistic chest radiographs from CT volumes and forward-project CT segmentation labels to obtain anatomically consistent 2D masks. We evaluate the approach on synthetic and real chest radiographs against a U-Net baseline, including ablations on model scale, input resolution, and vision encoder fine-tuning. Autoregressive anatomical supervision achieves performance comparable to specialized convolutional models in-distribution and demonstrates improved geometric robustness under domain shift to real CXR data. In addition, anatomy-pretrained models exhibit improved sample efficiency when adapting to novel localization tasks under limited supervision. Larger models and higher input image resolution improve performance, while vision encoder fine-tuning has limited effect. These results show that embedding anatomical structure directly into the generative objective promotes spatially grounded representations and supports anatomy-aware medical vision-language modeling.