Wavelet octave energy for breast tumor classification on sonography: A new shape feature

Yueh Ching Liao, King Chu Hung, Cheng Tung Ku, Chin Feng Tsai, Shu Mei Guo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Infiltrative nature of lesions is a significant feature of malignant breast lesion in ultrasound images. Characterizing infiltrative nature is crucial for the realization of computer-aided diagnosis system. In this study, the infiltrative nature is regarded as an energy that produces irregularly & considerably local variances in a 1-D signal. The local variances can be enhanced by few high octave energies in 1-D discrete periodized wavelet transform (DPWT). A test dataset of breast sonograms with the lesion contour delineated by an experienced physician & two inexperienced students are built for feature efficacy evaluation. A high individual performance result implies that the proposed feature is well correlated with radiologist's perception & closer to match those in trained physician than morphometric parameters. Experimental results also reveal that with a great performance improvement, the proposed feature is suitable for the combination with some morphometric parameters.

Original languageEnglish
Title of host publicationProceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009
Pages388-392
Number of pages5
DOIs
Publication statusPublished - 2009 Sep 21
Event2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009 - Okayama, Japan
Duration: 2009 Mar 262009 Mar 29

Publication series

NameProceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009

Other

Other2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009
CountryJapan
CityOkayama
Period09-03-2609-03-29

Fingerprint

Ultrasonography
Computer aided diagnosis
Discrete wavelet transforms
Tumors
Ultrasonics
Students

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

Liao, Y. C., Hung, K. C., Ku, C. T., Tsai, C. F., & Guo, S. M. (2009). Wavelet octave energy for breast tumor classification on sonography: A new shape feature. In Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009 (pp. 388-392). [4919307] (Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009). https://doi.org/10.1109/ICNSC.2009.4919307
Liao, Yueh Ching ; Hung, King Chu ; Ku, Cheng Tung ; Tsai, Chin Feng ; Guo, Shu Mei. / Wavelet octave energy for breast tumor classification on sonography : A new shape feature. Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009. 2009. pp. 388-392 (Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009).
@inproceedings{076d01920696452c9f20aa60cb4d58a9,
title = "Wavelet octave energy for breast tumor classification on sonography: A new shape feature",
abstract = "Infiltrative nature of lesions is a significant feature of malignant breast lesion in ultrasound images. Characterizing infiltrative nature is crucial for the realization of computer-aided diagnosis system. In this study, the infiltrative nature is regarded as an energy that produces irregularly & considerably local variances in a 1-D signal. The local variances can be enhanced by few high octave energies in 1-D discrete periodized wavelet transform (DPWT). A test dataset of breast sonograms with the lesion contour delineated by an experienced physician & two inexperienced students are built for feature efficacy evaluation. A high individual performance result implies that the proposed feature is well correlated with radiologist's perception & closer to match those in trained physician than morphometric parameters. Experimental results also reveal that with a great performance improvement, the proposed feature is suitable for the combination with some morphometric parameters.",
author = "Liao, {Yueh Ching} and Hung, {King Chu} and Ku, {Cheng Tung} and Tsai, {Chin Feng} and Guo, {Shu Mei}",
year = "2009",
month = "9",
day = "21",
doi = "10.1109/ICNSC.2009.4919307",
language = "English",
isbn = "9781424434923",
series = "Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009",
pages = "388--392",
booktitle = "Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009",

}

Liao, YC, Hung, KC, Ku, CT, Tsai, CF & Guo, SM 2009, Wavelet octave energy for breast tumor classification on sonography: A new shape feature. in Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009., 4919307, Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009, pp. 388-392, 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009, Okayama, Japan, 09-03-26. https://doi.org/10.1109/ICNSC.2009.4919307

Wavelet octave energy for breast tumor classification on sonography : A new shape feature. / Liao, Yueh Ching; Hung, King Chu; Ku, Cheng Tung; Tsai, Chin Feng; Guo, Shu Mei.

Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009. 2009. p. 388-392 4919307 (Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Wavelet octave energy for breast tumor classification on sonography

T2 - A new shape feature

AU - Liao, Yueh Ching

AU - Hung, King Chu

AU - Ku, Cheng Tung

AU - Tsai, Chin Feng

AU - Guo, Shu Mei

PY - 2009/9/21

Y1 - 2009/9/21

N2 - Infiltrative nature of lesions is a significant feature of malignant breast lesion in ultrasound images. Characterizing infiltrative nature is crucial for the realization of computer-aided diagnosis system. In this study, the infiltrative nature is regarded as an energy that produces irregularly & considerably local variances in a 1-D signal. The local variances can be enhanced by few high octave energies in 1-D discrete periodized wavelet transform (DPWT). A test dataset of breast sonograms with the lesion contour delineated by an experienced physician & two inexperienced students are built for feature efficacy evaluation. A high individual performance result implies that the proposed feature is well correlated with radiologist's perception & closer to match those in trained physician than morphometric parameters. Experimental results also reveal that with a great performance improvement, the proposed feature is suitable for the combination with some morphometric parameters.

AB - Infiltrative nature of lesions is a significant feature of malignant breast lesion in ultrasound images. Characterizing infiltrative nature is crucial for the realization of computer-aided diagnosis system. In this study, the infiltrative nature is regarded as an energy that produces irregularly & considerably local variances in a 1-D signal. The local variances can be enhanced by few high octave energies in 1-D discrete periodized wavelet transform (DPWT). A test dataset of breast sonograms with the lesion contour delineated by an experienced physician & two inexperienced students are built for feature efficacy evaluation. A high individual performance result implies that the proposed feature is well correlated with radiologist's perception & closer to match those in trained physician than morphometric parameters. Experimental results also reveal that with a great performance improvement, the proposed feature is suitable for the combination with some morphometric parameters.

UR - http://www.scopus.com/inward/record.url?scp=70349145405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349145405&partnerID=8YFLogxK

U2 - 10.1109/ICNSC.2009.4919307

DO - 10.1109/ICNSC.2009.4919307

M3 - Conference contribution

AN - SCOPUS:70349145405

SN - 9781424434923

T3 - Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009

SP - 388

EP - 392

BT - Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009

ER -

Liao YC, Hung KC, Ku CT, Tsai CF, Guo SM. Wavelet octave energy for breast tumor classification on sonography: A new shape feature. In Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009. 2009. p. 388-392. 4919307. (Proceedings of the 2009 IEEE International Conference on Networking, Sensing and Control, ICNSC 2009). https://doi.org/10.1109/ICNSC.2009.4919307