Wavelet transforms for detecting microcalcifications in mammograms
Abstract
Clusters of fine, granular microcalcifications in mammograms may be an early sign of disease. Individual grains are difficult to detect and segment due to size and shape variability and because the background mammogram texture is typically inhomogeneous. The authors develop a two-stage method based on wavelet transforms for detecting and segmenting calcifications. The first stage is based on an undecimated wavelet transform, which is simply the conventional filter bank implementation without downsampling, so that the low-low (LL), low-high (LH), high-low (HL), and high-high (HH) sub-bands remain at full size. Detection takes place in HH and the combination LH + HL. Four octaves are compared with two inter-octave voices for finer scale resolution. By appropriate selection of the wavelet basis the detection of microcalcifications in the relevant size range can be nearly optimized. In fact, the filters which transform the input image into HH and LH + HL are closely related to prewhitening matched filters for detecting Gaussian objects (idealized microcalcifications) in two common forms of Markov (background) noise. The second stage is designed to overcome the limitations of the simplistic Gaussian assumption and provides an accurate segmentation of calcification boundaries. Detected pixel sites in HH and LH + HL aremore »
- Authors:
-
- Univ. of Arizona, Tucson, AZ (United States). Dept. of Electrical and Computer Engineering
- Publication Date:
- OSTI Identifier:
- 227874
- Resource Type:
- Journal Article
- Journal Name:
- IEEE Transactions on Medical Imaging
- Additional Journal Information:
- Journal Volume: 15; Journal Issue: 2; Other Information: PBD: Apr 1996
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 55 BIOLOGY AND MEDICINE, BASIC STUDIES; MAMMARY GLANDS; BIOMEDICAL RADIOGRAPHY; NEOPLASMS; DIAGNOSIS; X RADIATION; IMAGE PROCESSING
Citation Formats
Strickland, R N, and Hahn, H I. Wavelet transforms for detecting microcalcifications in mammograms. United States: N. p., 1996.
Web. doi:10.1109/42.491423.
Strickland, R N, & Hahn, H I. Wavelet transforms for detecting microcalcifications in mammograms. United States. https://doi.org/10.1109/42.491423
Strickland, R N, and Hahn, H I. 1996.
"Wavelet transforms for detecting microcalcifications in mammograms". United States. https://doi.org/10.1109/42.491423.
@article{osti_227874,
title = {Wavelet transforms for detecting microcalcifications in mammograms},
author = {Strickland, R N and Hahn, H I},
abstractNote = {Clusters of fine, granular microcalcifications in mammograms may be an early sign of disease. Individual grains are difficult to detect and segment due to size and shape variability and because the background mammogram texture is typically inhomogeneous. The authors develop a two-stage method based on wavelet transforms for detecting and segmenting calcifications. The first stage is based on an undecimated wavelet transform, which is simply the conventional filter bank implementation without downsampling, so that the low-low (LL), low-high (LH), high-low (HL), and high-high (HH) sub-bands remain at full size. Detection takes place in HH and the combination LH + HL. Four octaves are compared with two inter-octave voices for finer scale resolution. By appropriate selection of the wavelet basis the detection of microcalcifications in the relevant size range can be nearly optimized. In fact, the filters which transform the input image into HH and LH + HL are closely related to prewhitening matched filters for detecting Gaussian objects (idealized microcalcifications) in two common forms of Markov (background) noise. The second stage is designed to overcome the limitations of the simplistic Gaussian assumption and provides an accurate segmentation of calcification boundaries. Detected pixel sites in HH and LH + HL are dilated then weighted before computing the inverse wavelet transform. Individual microcalcifications are greatly enhanced in the output image, to the point where straightforward thresholding can be applied to segment them. FROC curves are computed from tests using a freely distributed database of digitized mammograms.},
doi = {10.1109/42.491423},
url = {https://www.osti.gov/biblio/227874},
journal = {IEEE Transactions on Medical Imaging},
number = 2,
volume = 15,
place = {United States},
year = {Mon Apr 01 00:00:00 EST 1996},
month = {Mon Apr 01 00:00:00 EST 1996}
}