Mammographic Density Classification using Multiresolution Histogram Information

Izzati Muhimmah; Reyer Zwiggelaar

Mammographic Density Classification using Multiresolution Histogram Information

Izzati Muhimmah, Reyer Zwiggelaar

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

4 Downloads (Pure)

Abstract

Mammographic density is known to be an important indicator of breast cancer risk. Quantitative estimation approaches based on histogram information have been investigated previously. However, claims have been made that greylevel information might be insufficient to discriminate between complex density classes. A multi-resolution histogram technique, which was developed as a texture analysis approach, has been investigated as an alternative classification space. Using a DAG-SVM classifier on the MIAS database the result shows an agreement of 77.57% between automatic and expert radiologist manual classification.

Original language	English
Title of host publication	Proceedings of the International Special Topic Conference on Information Technology in Biomedicine
Subtitle of host publication	ITAB
Publisher	IEEE Press
Number of pages	6
Publication status	Published - 26 Oct 2006

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

Mammographic density classification ...Final published version, 238 KB

Permanent link

Persistent link

Cite this

@inproceedings{acf630264dbc4dc4818f6023c908a799,

title = "Mammographic Density Classification using Multiresolution Histogram Information",

abstract = "Mammographic density is known to be an important indicator of breast cancer risk. Quantitative estimation approaches based on histogram information have been investigated previously. However, claims have been made that greylevel information might be insufficient to discriminate between complex density classes. A multi-resolution histogram technique, which was developed as a texture analysis approach, has been investigated as an alternative classification space. Using a DAG-SVM classifier on the MIAS database the result shows an agreement of 77.57% between automatic and expert radiologist manual classification.",

author = "Izzati Muhimmah and Reyer Zwiggelaar",

year = "2006",

month = oct,

day = "26",

language = "English",

booktitle = "Proceedings of the International Special Topic Conference on Information Technology in Biomedicine",

publisher = "IEEE Press",

address = "United States of America",

}

TY - GEN

T1 - Mammographic Density Classification using Multiresolution Histogram Information

AU - Muhimmah, Izzati

AU - Zwiggelaar, Reyer

PY - 2006/10/26

Y1 - 2006/10/26

N2 - Mammographic density is known to be an important indicator of breast cancer risk. Quantitative estimation approaches based on histogram information have been investigated previously. However, claims have been made that greylevel information might be insufficient to discriminate between complex density classes. A multi-resolution histogram technique, which was developed as a texture analysis approach, has been investigated as an alternative classification space. Using a DAG-SVM classifier on the MIAS database the result shows an agreement of 77.57% between automatic and expert radiologist manual classification.

AB - Mammographic density is known to be an important indicator of breast cancer risk. Quantitative estimation approaches based on histogram information have been investigated previously. However, claims have been made that greylevel information might be insufficient to discriminate between complex density classes. A multi-resolution histogram technique, which was developed as a texture analysis approach, has been investigated as an alternative classification space. Using a DAG-SVM classifier on the MIAS database the result shows an agreement of 77.57% between automatic and expert radiologist manual classification.

M3 - Conference Proceeding (Non-Journal item)

BT - Proceedings of the International Special Topic Conference on Information Technology in Biomedicine

PB - IEEE Press

ER -

Mammographic Density Classification using Multiresolution Histogram Information

Abstract

UN SDGs

Access to Document

Permanent link

Fingerprint

Cite this