Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering

Carlos S. Benitez; Said Pertuz; Otso Arponen; Anna Leena Lääperi; Irina Rinta-Kiikka

doi:10.1109/ColCACI56938.2022.9905248

Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering

Carlos S. Benitez, Said Pertuz, Otso Arponen, Anna Leena Lääperi, Irina Rinta-Kiikka

Tutkimustuotos: Konferenssiartikkeli › Scientific › vertaisarvioitu

Abstrakti

Mammographic breast percent density (PD) is one of the strongest risk factors associated with the development of breast cancer. As a result, the accurate estimation of PD from screening mammograms is an important problem for breast cancer risk assessment. Nevertheless, automatic segmentation of the dense fibroglandular tissue (FGT) is a difficult task due to the complexity of morphological characteristics and heterogeneity of the breast. In this work, we present a hybrid algorithm based on convolutional neural networks (CNN) and intensity-based clustering used for the fully-automated segmentation of dense tissue in mammograms. We utilize a dataset of 582 mammograms with expert reader's manually segmented dense tissue areas as a reference. The PD estimates obtained with the proposed method yield a median PD error of 7.7% with no statistically significant differences with respect to the expert.

Alkuperäiskieli	Englanti
Otsikko	2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings
Toimittajat	Alvaro David Orjuela-Canon
Kustantaja	IEEE
Sivumäärä	4
ISBN (elektroninen)	978-1-6654-7470-2
DOI - pysyväislinkit	https://doi.org/10.1109/ColCACI56938.2022.9905248
Tila	Julkaistu - 2022
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisussa
Tapahtuma	IEEE Colombian Conference on Applications of Computational Intelligence - Cali, Kolumbia Kesto: 27 heinäk. 2022 → 29 heinäk. 2022 Konferenssinumero: 2022

Julkaisusarja

Nimi	IEEE Colombian Conference on Applications of Computational Intelligence
Vuosikerta	2022

Conference

Conference	IEEE Colombian Conference on Applications of Computational Intelligence
Lyhennettä	ColCACI
Maa/Alue	Kolumbia
Kaupunki	Cali
Ajanjakso	27/07/22 → 29/07/22

Julkaisufoorumi-taso

Jufo-taso 1

!!ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Computer Vision and Pattern Recognition
Control and Optimization

YK:n kestävän kehityksen tavoitteet

Tämä tuotos edistää seuraavia kestävän kehityksen tavoitteita:

Pääsy asiakirjaan

10.1109/ColCACI56938.2022.9905248

Siteeraa tätä

Benitez, C. S., Pertuz, S., Arponen, O., Lääperi, A. L., & Rinta-Kiikka, I. (2022). Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering. teoksessa A. D. Orjuela-Canon (Toimittaja), 2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings (IEEE Colombian Conference on Applications of Computational Intelligence; Vuosikerta 2022). IEEE. https://doi.org/10.1109/ColCACI56938.2022.9905248

Benitez, Carlos S. ; Pertuz, Said ; Arponen, Otso et al. / Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering. 2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings. Toimittaja / Alvaro David Orjuela-Canon. IEEE, 2022. (IEEE Colombian Conference on Applications of Computational Intelligence).

@inproceedings{d66aa1429ef64900befb701ef370d973,

title = "Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering",

abstract = "Mammographic breast percent density (PD) is one of the strongest risk factors associated with the development of breast cancer. As a result, the accurate estimation of PD from screening mammograms is an important problem for breast cancer risk assessment. Nevertheless, automatic segmentation of the dense fibroglandular tissue (FGT) is a difficult task due to the complexity of morphological characteristics and heterogeneity of the breast. In this work, we present a hybrid algorithm based on convolutional neural networks (CNN) and intensity-based clustering used for the fully-automated segmentation of dense tissue in mammograms. We utilize a dataset of 582 mammograms with expert reader's manually segmented dense tissue areas as a reference. The PD estimates obtained with the proposed method yield a median PD error of 7.7% with no statistically significant differences with respect to the expert.",

keywords = "breast cancer, convolutional neural network, digital mammography, fibroglandular tissue, percent density, risk",

author = "Benitez, {Carlos S.} and Said Pertuz and Otso Arponen and L{\"a}{\"a}peri, {Anna Leena} and Irina Rinta-Kiikka",

note = "Funding Information: The authors have no conflict of interest to declare. This work was partially supported by MINCIENCIAS (project code 110284467139: Software de an{\'a}lisis parenquimatoso de im{\'a}genes mamogr{\'a}ficas para la estimaci{\'o}n de riesgo de c{\'a}ncer de seno) and by Universidad Industrial de Santander (project VIE2544: An{\'a}lisis Espacial de Densidad en Mamograf{\'i}a Digital para la Evaluaci{\'o}n de Riesgo de C{\'a}ncer de Seno). Publisher Copyright: {\textcopyright} 2022 IEEE.; IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI ; Conference date: 27-07-2022 Through 29-07-2022",

year = "2022",

doi = "10.1109/ColCACI56938.2022.9905248",

language = "English",

series = "IEEE Colombian Conference on Applications of Computational Intelligence",

publisher = "IEEE",

editor = "Orjuela-Canon, {Alvaro David}",

booktitle = "2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings",

}

Benitez, CS, Pertuz, S, Arponen, O, Lääperi, AL & Rinta-Kiikka, I 2022, Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering. julkaisussa AD Orjuela-Canon (Toimittaja), 2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings. IEEE Colombian Conference on Applications of Computational Intelligence, Vuosikerta 2022, IEEE, IEEE Colombian Conference on Applications of Computational Intelligence, Cali, Kolumbia, 27/07/22. https://doi.org/10.1109/ColCACI56938.2022.9905248

Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering. / Benitez, Carlos S.; Pertuz, Said; Arponen, Otso et al.
2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings. toim. / Alvaro David Orjuela-Canon. IEEE, 2022. (IEEE Colombian Conference on Applications of Computational Intelligence; Vuosikerta 2022).

Tutkimustuotos: Konferenssiartikkeli › Scientific › vertaisarvioitu

TY - GEN

T1 - Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering

AU - Benitez, Carlos S.

AU - Pertuz, Said

AU - Arponen, Otso

AU - Lääperi, Anna Leena

AU - Rinta-Kiikka, Irina

N1 - Conference code: 2022

PY - 2022

Y1 - 2022

N2 - Mammographic breast percent density (PD) is one of the strongest risk factors associated with the development of breast cancer. As a result, the accurate estimation of PD from screening mammograms is an important problem for breast cancer risk assessment. Nevertheless, automatic segmentation of the dense fibroglandular tissue (FGT) is a difficult task due to the complexity of morphological characteristics and heterogeneity of the breast. In this work, we present a hybrid algorithm based on convolutional neural networks (CNN) and intensity-based clustering used for the fully-automated segmentation of dense tissue in mammograms. We utilize a dataset of 582 mammograms with expert reader's manually segmented dense tissue areas as a reference. The PD estimates obtained with the proposed method yield a median PD error of 7.7% with no statistically significant differences with respect to the expert.

AB - Mammographic breast percent density (PD) is one of the strongest risk factors associated with the development of breast cancer. As a result, the accurate estimation of PD from screening mammograms is an important problem for breast cancer risk assessment. Nevertheless, automatic segmentation of the dense fibroglandular tissue (FGT) is a difficult task due to the complexity of morphological characteristics and heterogeneity of the breast. In this work, we present a hybrid algorithm based on convolutional neural networks (CNN) and intensity-based clustering used for the fully-automated segmentation of dense tissue in mammograms. We utilize a dataset of 582 mammograms with expert reader's manually segmented dense tissue areas as a reference. The PD estimates obtained with the proposed method yield a median PD error of 7.7% with no statistically significant differences with respect to the expert.

KW - breast cancer

KW - convolutional neural network

KW - digital mammography

KW - fibroglandular tissue

KW - percent density

KW - risk

U2 - 10.1109/ColCACI56938.2022.9905248

DO - 10.1109/ColCACI56938.2022.9905248

M3 - Conference contribution

AN - SCOPUS:85141381995

T3 - IEEE Colombian Conference on Applications of Computational Intelligence

BT - 2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings

A2 - Orjuela-Canon, Alvaro David

PB - IEEE

T2 - IEEE Colombian Conference on Applications of Computational Intelligence

Y2 - 27 July 2022 through 29 July 2022

ER -

Benitez CS, Pertuz S, Arponen O, Lääperi AL, Rinta-Kiikka I. Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering. julkaisussa Orjuela-Canon AD, toimittaja, 2022 IEEE Colombian Conference on Applications of Computational Intelligence, ColCACI 2022 - Proceedings. IEEE. 2022. (IEEE Colombian Conference on Applications of Computational Intelligence). doi: 10.1109/ColCACI56938.2022.9905248

Automatic Dense Tissue Segmentation in Digital Mammography Images Based on Fully Convolutional Network and Intensity-Based Clustering

Abstrakti

Julkaisusarja

Conference

Julkaisufoorumi-taso

!!ASJC Scopus subject areas

YK:n kestävän kehityksen tavoitteet

Pääsy asiakirjaan

Sormenjälki

Siteeraa tätä