Skip to main content

Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index

Publication ,  Conference
Peng, Y; Hou, R; Ren, Y; Grimm, LJ; Marks, JR; Hwang, ES; Lo, JY
Published in: Progress in Biomedical Optics and Imaging - Proceedings of SPIE
January 1, 2020

Detecting microcalcification clusters in mammograms is important to the diagnosis of breast diseases. Previous studies which mainly focused on supervised methods require abundant annotated training data but these data are usually hard to acquire. In this work, we leverage unsupervised convolutional autoencoders and structural similarity (SSIM) based post-processing to detect and localize microcalcification clusters in full-field digital mammograms (FFDMs). Our models were trained by patches extracted from 3,632 normal cases, in total with 16,702 mammograms. Evaluations were conducted in three aspects, including patch-based anomaly detection, pixel-wise microcalcification localization, and microcalcification cluster detection. Specifically, the receiver operating characteristic (ROC) analysis was used for patch-based anomaly detection. Then, a pixel-wise ROC analysis and a cluster-based free-response ROC (FROC) analysis were performed to assess our detection algorithms of individual microcalcifications and microcalcification clusters, respectively. We achieved a pixel-wise AUC of 0.97 as well as a cluster-based sensitivity of 0.62 at 1 false positive per image and 0.75 at 2.5 false positives per image. Both qualitative and quantitative results demonstrated the effectiveness of our method.

Duke Scholars

Published In

Progress in Biomedical Optics and Imaging - Proceedings of SPIE

DOI

ISSN

1605-7422

ISBN

9781510633957

Publication Date

January 1, 2020

Volume

11314
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Peng, Y., Hou, R., Ren, Y., Grimm, L. J., Marks, J. R., Hwang, E. S., & Lo, J. Y. (2020). Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index. In Progress in Biomedical Optics and Imaging - Proceedings of SPIE (Vol. 11314). https://doi.org/10.1117/12.2551263
Peng, Y., R. Hou, Y. Ren, L. J. Grimm, J. R. Marks, E. S. Hwang, and J. Y. Lo. “Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index.” In Progress in Biomedical Optics and Imaging - Proceedings of SPIE, Vol. 11314, 2020. https://doi.org/10.1117/12.2551263.
Peng Y, Hou R, Ren Y, Grimm LJ, Marks JR, Hwang ES, et al. Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index. In: Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2020.
Peng, Y., et al. “Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index.” Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 11314, 2020. Scopus, doi:10.1117/12.2551263.
Peng Y, Hou R, Ren Y, Grimm LJ, Marks JR, Hwang ES, Lo JY. Microcalcification localization and cluster detection using unsupervised convolutional autoencoders and structural similarity index. Progress in Biomedical Optics and Imaging - Proceedings of SPIE. 2020.

Published In

Progress in Biomedical Optics and Imaging - Proceedings of SPIE

DOI

ISSN

1605-7422

ISBN

9781510633957

Publication Date

January 1, 2020

Volume

11314