We present a method based on a generative model for detection of disturbances such as prosthesis, screws, zippers, and metals in 2D radiographs. The generative model is trained in an unsupervised fashion using clinical radiographs as well as simulated data, none of which contain disturbances. Our approach employs a latent space consistency loss which has the benefit of identifying similarities, and is enforced to reconstruct X-rays without disturbances. In order to detect images with disturbances, an anomaly score is computed also employing the Frechet distance between the input X-ray and the reconstructed one using our generative model. Validation was performed using clinical pelvis radiographs. We achieved an AUC of 0.77 and 0.83 with clinical and synthetic data, respectively. The results demonstrated a good accuracy of our method for detecting outliers as well as the advantage of utilizing synthetic data.
|Title of host publication||2021 IEEE 18th International Symposium on Biomedical Imaging, ISBI 2021|
|Publisher||IEEE Computer Society|
|Number of pages||4|
|State||Published - 13 Apr 2021|
|Event||18th IEEE International Symposium on Biomedical Imaging, ISBI 2021 - Nice, France|
Duration: 13 Apr 2021 → 16 Apr 2021
|Name||Proceedings - International Symposium on Biomedical Imaging|
|Conference||18th IEEE International Symposium on Biomedical Imaging, ISBI 2021|
|Period||13/04/21 → 16/04/21|
Bibliographical noteFunding Information:
This work was supported by grant 234-2015-FONDECYT (Master Program) from Cienciactiva of CONCYTEC-PERU and German BMBF research campus MODAL (grant no. 3FO18501). The authors thank CiTeSoft-UNSA for the database access. The authors report no conflicts of interest.
© 2021 IEEE.
- Anomaly detection
- Generative models
- Pelvic radiographs
- Unsupervised learning