# Task Driven Generative Modeling for Unsupervised Domain Adaptation: Application to X-ray Image Segmentation

Source: [Task Driven Generative Modeling for Unsupervised Domain Adaptation: Application to X-ray Image Segmentation](<https://www.sota2.com/research/paper/68fa68551e209131cea1b918>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:35:00.393Z

Authors: Yue Zhang, Shun Miao, Tommaso Mansi, Rui Liao

Published: 2018-06-11T22:39:35.000Z

## Abstract

Automatic parsing of anatomical objects in X-ray images is critical to many clinical applications in particular towards image-guided invention and workflow automation. Existing deep network models require a large amount of labeled data. However, obtaining accurate pixel-wise labeling in X-ray images relies heavily on skilled clinicians due to the large overlaps of anatomy and the complex texture patterns. On the other hand, organs in 3D CT scans preserve clearer structures as well as sharper boundaries and thus can be easily delineated. In this paper, we propose a novel model framework for learning automatic X-ray image parsing from labeled CT scans. Specifically, a Dense Image-to-Image network (DI2I) for multi-organ segmentation is first trained on X-ray like Digitally Reconstructed Radiographs (DRRs) rendered from 3D CT volumes. Then we introduce a Task Driven Generative Adversarial Network (TD-GAN) architecture to achieve simultaneous style transfer and parsing for unseen real X-ray images. TD-GAN consists of a modified cycle-GAN substructure for pixel-to-pixel translation between DRRs and X-ray images and an added module leveraging the pre-trained DI2I to enforce segmentation consistency. The TD-GAN framework is general and can be easily adapted to other learning tasks. In the numerical experiments, we validate the proposed model on 815 DRRs and 153 topograms. While the vanilla DI2I without any adaptation fails completely on segmenting the topograms, the proposed model does not require any topogram labels and is able to provide a promising average dice of 85% which achieves the same level accuracy of supervised training (88%).

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Optic Disc and Cup Segmentation on RIM-ONE r3 (target domain)](<https://www.sota2.com/research/sota/optic-disc-and-cup-segmentation-on-rim-one-r3-target-domain.md>): Optic disc and cup segmentation; RIM-ONE r3 (target domain); Dice (Disc): 85.3.

- [Optic Disc and Cup Segmentation on Drishti-GS (target domain)](<https://www.sota2.com/research/sota/optic-disc-and-cup-segmentation-on-drishti-gs-target-domain.md>): Optic disc and cup segmentation; Drishti-GS (target domain); Dice (Disc): 0.924.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1806.07201>)

[PDF](<https://arxiv.org/pdf/1806.07201.pdf>)
