Detection of acini in histopathology slides: towards automated prediction of breast cancer risk

Suzanne Wetstein, Allison Onken, Gabrielle Baker, Michael Pyle, Josien Pluim, Rulla Tamimi, Yujing Heng, Mitko Veta

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

3 Downloads (Pure)

Abstract

Terminal duct lobular units (TDLUs) are structures in the breast which involute with the completion of childbearing and physiological ageing. Women with less TDLU involution are more likely to develop breast cancer than those with more involution. Thus, TDLU involution may be utilized as a biomarker to predict invasive cancer risk. Manual assessment of TDLU involution is a cumbersome and subjective process. This makes it amenable for automated assessment by image analysis. In this study, we developed and evaluated an acini detection method as a first step towards automated assessment of TDLU involution using a dataset of histopathological whole-slide images (WSIs) from the Nurses’ Health Study (NHS) and NHSII. The NHS/NHSII is among the world's largest investigations of epidemiological risk factors for major chronic diseases in women. We compared three different approaches to detect acini in WSIs using the U-Net convolutional neural network architecture. The approaches differ in the target that is predicted by the network: circular mask labels, soft labels and distance maps. Our results showed that soft label targets lead to a better detection performance than the other methods. F1 scores of 0.65, 0.73 and 0.66 were obtained with circular mask labels, soft labels and distance maps, respectively. Our acini detection method was furthermore validated by applying it to measure acini count per mm2 of tissue area on an independent set of WSIs. This measure was found to be significantly negatively correlated with age.
Original languageEnglish
Title of host publicationMedical Imaging 2019
Subtitle of host publicationDigital Pathology
EditorsJohn E. Tomaszewski, Aaron D. Ward
Place of PublicationBellingham
Number of pages7
ISBN (Electronic)9781510625594
DOIs
Publication statusPublished - 18 Mar 2019
EventSPIE Medical Imaging 2019 - San Diego, United States
Duration: 16 Feb 201921 Feb 2019
http://spie.org/MI/entireprogram/2019-2-20?print=2&SSO=1

Publication series

NameProceedings of SPIE
Volume10956

Conference

ConferenceSPIE Medical Imaging 2019
CountryUnited States
CitySan Diego
Period16/02/1921/02/19
Internet address

Fingerprint

Ducts
Labels
Masks
Health
Biomarkers
Network architecture
Image analysis
Aging of materials
Tissue
Neural networks

Cite this

Wetstein, S., Onken, A., Baker, G., Pyle, M., Pluim, J., Tamimi, R., ... Veta, M. (2019). Detection of acini in histopathology slides: towards automated prediction of breast cancer risk. In J. E. Tomaszewski, & A. D. Ward (Eds.), Medical Imaging 2019: Digital Pathology [109560Q] (Proceedings of SPIE; Vol. 10956). Bellingham. https://doi.org/10.1117/12.2511408
Wetstein, Suzanne ; Onken, Allison ; Baker, Gabrielle ; Pyle, Michael ; Pluim, Josien ; Tamimi, Rulla ; Heng, Yujing ; Veta, Mitko. / Detection of acini in histopathology slides : towards automated prediction of breast cancer risk. Medical Imaging 2019: Digital Pathology. editor / John E. Tomaszewski ; Aaron D. Ward. Bellingham, 2019. (Proceedings of SPIE).
@inproceedings{9bc0103ead9f4d87bd2b623c761d05ff,
title = "Detection of acini in histopathology slides: towards automated prediction of breast cancer risk",
abstract = "Terminal duct lobular units (TDLUs) are structures in the breast which involute with the completion of childbearing and physiological ageing. Women with less TDLU involution are more likely to develop breast cancer than those with more involution. Thus, TDLU involution may be utilized as a biomarker to predict invasive cancer risk. Manual assessment of TDLU involution is a cumbersome and subjective process. This makes it amenable for automated assessment by image analysis. In this study, we developed and evaluated an acini detection method as a first step towards automated assessment of TDLU involution using a dataset of histopathological whole-slide images (WSIs) from the Nurses’ Health Study (NHS) and NHSII. The NHS/NHSII is among the world's largest investigations of epidemiological risk factors for major chronic diseases in women. We compared three different approaches to detect acini in WSIs using the U-Net convolutional neural network architecture. The approaches differ in the target that is predicted by the network: circular mask labels, soft labels and distance maps. Our results showed that soft label targets lead to a better detection performance than the other methods. F1 scores of 0.65, 0.73 and 0.66 were obtained with circular mask labels, soft labels and distance maps, respectively. Our acini detection method was furthermore validated by applying it to measure acini count per mm2 of tissue area on an independent set of WSIs. This measure was found to be significantly negatively correlated with age.",
author = "Suzanne Wetstein and Allison Onken and Gabrielle Baker and Michael Pyle and Josien Pluim and Rulla Tamimi and Yujing Heng and Mitko Veta",
year = "2019",
month = "3",
day = "18",
doi = "10.1117/12.2511408",
language = "English",
series = "Proceedings of SPIE",
editor = "Tomaszewski, {John E.} and Ward, {Aaron D.}",
booktitle = "Medical Imaging 2019",

}

Wetstein, S, Onken, A, Baker, G, Pyle, M, Pluim, J, Tamimi, R, Heng, Y & Veta, M 2019, Detection of acini in histopathology slides: towards automated prediction of breast cancer risk. in JE Tomaszewski & AD Ward (eds), Medical Imaging 2019: Digital Pathology., 109560Q, Proceedings of SPIE, vol. 10956, Bellingham, SPIE Medical Imaging 2019, San Diego, United States, 16/02/19. https://doi.org/10.1117/12.2511408

Detection of acini in histopathology slides : towards automated prediction of breast cancer risk. / Wetstein, Suzanne; Onken, Allison; Baker, Gabrielle; Pyle, Michael; Pluim, Josien; Tamimi, Rulla; Heng, Yujing; Veta, Mitko.

Medical Imaging 2019: Digital Pathology. ed. / John E. Tomaszewski; Aaron D. Ward. Bellingham, 2019. 109560Q (Proceedings of SPIE; Vol. 10956).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Detection of acini in histopathology slides

T2 - towards automated prediction of breast cancer risk

AU - Wetstein, Suzanne

AU - Onken, Allison

AU - Baker, Gabrielle

AU - Pyle, Michael

AU - Pluim, Josien

AU - Tamimi, Rulla

AU - Heng, Yujing

AU - Veta, Mitko

PY - 2019/3/18

Y1 - 2019/3/18

N2 - Terminal duct lobular units (TDLUs) are structures in the breast which involute with the completion of childbearing and physiological ageing. Women with less TDLU involution are more likely to develop breast cancer than those with more involution. Thus, TDLU involution may be utilized as a biomarker to predict invasive cancer risk. Manual assessment of TDLU involution is a cumbersome and subjective process. This makes it amenable for automated assessment by image analysis. In this study, we developed and evaluated an acini detection method as a first step towards automated assessment of TDLU involution using a dataset of histopathological whole-slide images (WSIs) from the Nurses’ Health Study (NHS) and NHSII. The NHS/NHSII is among the world's largest investigations of epidemiological risk factors for major chronic diseases in women. We compared three different approaches to detect acini in WSIs using the U-Net convolutional neural network architecture. The approaches differ in the target that is predicted by the network: circular mask labels, soft labels and distance maps. Our results showed that soft label targets lead to a better detection performance than the other methods. F1 scores of 0.65, 0.73 and 0.66 were obtained with circular mask labels, soft labels and distance maps, respectively. Our acini detection method was furthermore validated by applying it to measure acini count per mm2 of tissue area on an independent set of WSIs. This measure was found to be significantly negatively correlated with age.

AB - Terminal duct lobular units (TDLUs) are structures in the breast which involute with the completion of childbearing and physiological ageing. Women with less TDLU involution are more likely to develop breast cancer than those with more involution. Thus, TDLU involution may be utilized as a biomarker to predict invasive cancer risk. Manual assessment of TDLU involution is a cumbersome and subjective process. This makes it amenable for automated assessment by image analysis. In this study, we developed and evaluated an acini detection method as a first step towards automated assessment of TDLU involution using a dataset of histopathological whole-slide images (WSIs) from the Nurses’ Health Study (NHS) and NHSII. The NHS/NHSII is among the world's largest investigations of epidemiological risk factors for major chronic diseases in women. We compared three different approaches to detect acini in WSIs using the U-Net convolutional neural network architecture. The approaches differ in the target that is predicted by the network: circular mask labels, soft labels and distance maps. Our results showed that soft label targets lead to a better detection performance than the other methods. F1 scores of 0.65, 0.73 and 0.66 were obtained with circular mask labels, soft labels and distance maps, respectively. Our acini detection method was furthermore validated by applying it to measure acini count per mm2 of tissue area on an independent set of WSIs. This measure was found to be significantly negatively correlated with age.

UR - http://www.scopus.com/inward/record.url?scp=85068679186&partnerID=8YFLogxK

U2 - 10.1117/12.2511408

DO - 10.1117/12.2511408

M3 - Conference contribution

T3 - Proceedings of SPIE

BT - Medical Imaging 2019

A2 - Tomaszewski, John E.

A2 - Ward, Aaron D.

CY - Bellingham

ER -

Wetstein S, Onken A, Baker G, Pyle M, Pluim J, Tamimi R et al. Detection of acini in histopathology slides: towards automated prediction of breast cancer risk. In Tomaszewski JE, Ward AD, editors, Medical Imaging 2019: Digital Pathology. Bellingham. 2019. 109560Q. (Proceedings of SPIE). https://doi.org/10.1117/12.2511408