Published February 20, 2025
| Version v1
Model
Open
Generative Modeling for Interpretable Anomaly Detection in Medical Imaging: Applications in Failure Detection and Data Curation
Authors/Creators
-
1.
The University of Texas MD Anderson Cancer Center
Contributors
Data collector (3):
Researcher (8):
Supervisor (2):
-
1.
The University of Texas MD Anderson Cancer Center
-
2.
Texas Southern University
-
3.
University of Washington
-
4.
Rice University
- 5. Baylor College of Medicine
Description
This archive contains the ChestX-ray14 StyleGAN2-ADA model weights that were used for anomaly detection within radiographs from the Medical Imaging and Data Resource Center (MIDRC) in the manuscript entitled "Generative Modeling for Interpretable Failure Detection in Liver CT Segmentation and Scalable Data Curation of Chest Radiographs". The model weights were saved at 25,000 kimgs, having achieved the lowest Fréchet Inception Distance (5.12) across those kimgs.
Files
Files
(364.0 MB)
| Name | Size | |
|---|---|---|
|
md5:2799e55e13123ee76e46f1fa1e721e7b
|
364.0 MB | Download |
Additional details
Funding
- National Institutes of Health
- P30CA016672
- National Institutes of Health
- R01CA235564
- National Institutes of Health
- R01CA221971
Dates
- Available
-
2025-02-20
Software
- Repository URL
- https://github.com/mckellwoodland/gan_anom_detect/
- Programming language
- Python
- Development Status
- Active
References
- Karras, T., et al.: Analyzing and improving the image quality of stylegan. In: Zabih, R., et al. (eds.) CVPR. IEEE (2020)
- Karras, T., et al.: Training generative adversarial networks with limited data. In: Larochelle, H., et al. (eds.) NeurIPS. vol. 33, pp. 12104–12114. Curran Associates, Inc. (2020)
- Wang, X., et al.: Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In: Chellappa, R., et al. (eds.) CVPR. IEEE (2017)