Spectral Data Augmentation Techniques to Quantify Lung Pathology from CT-Images
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Spectral Data Augmentation Techniques to Quantify Lung Pathology from CT-Images. / Kayal, Subhradeep; Dubost, Florian; Tiddens, Harm A.W.M.; De Bruijne, Marleen.
ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging. IEEE, 2020. p. 586-590 9098581.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Spectral Data Augmentation Techniques to Quantify Lung Pathology from CT-Images
AU - Kayal, Subhradeep
AU - Dubost, Florian
AU - Tiddens, Harm A.W.M.
AU - De Bruijne, Marleen
PY - 2020/4
Y1 - 2020/4
N2 - Data augmentation is of paramount importance in biomedical image processing tasks, characterized by inadequate amounts of labelled data, to best use all of the data that is present. In-use techniques range from intensity transformations and elastic deformations, to linearly combining existing data points to make new ones. In this work, we propose the use of spectral techniques for data augmentation, using the discrete cosine and wavelet transforms. We empirically evaluate our approaches on a CT texture analysis task to detect abnormal lung-tissue in patients with cystic fibrosis. Empirical experiments show that the proposed spectral methods perform favourably as compared to the existing methods. When used in combination with existing methods, our proposed approach can increase the relative minor class segmentation performance by 44.1% over a simple replication baseline.
AB - Data augmentation is of paramount importance in biomedical image processing tasks, characterized by inadequate amounts of labelled data, to best use all of the data that is present. In-use techniques range from intensity transformations and elastic deformations, to linearly combining existing data points to make new ones. In this work, we propose the use of spectral techniques for data augmentation, using the discrete cosine and wavelet transforms. We empirically evaluate our approaches on a CT texture analysis task to detect abnormal lung-tissue in patients with cystic fibrosis. Empirical experiments show that the proposed spectral methods perform favourably as compared to the existing methods. When used in combination with existing methods, our proposed approach can increase the relative minor class segmentation performance by 44.1% over a simple replication baseline.
KW - Cystic Fibrosis
KW - Data Augmentation
KW - Discrete Cosine Transform
KW - Discrete Wavelet Transform
KW - Lung CT
KW - Lung Texture Analysis
KW - Spectral Transforms
U2 - 10.1109/ISBI45749.2020.9098581
DO - 10.1109/ISBI45749.2020.9098581
M3 - Article in proceedings
AN - SCOPUS:85085857028
SP - 586
EP - 590
BT - ISBI 2020 - 2020 IEEE International Symposium on Biomedical Imaging
PB - IEEE
T2 - 17th IEEE International Symposium on Biomedical Imaging, ISBI 2020
Y2 - 3 April 2020 through 7 April 2020
ER -
ID: 248471614