Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening

Preliminary Study

Eunkyung Kim, Hyo Eun Kim, Kyunghwa Han, Bong Joo Kang, Yu Mee Sohn, Ok Hee Woo, Chan Wha Lee

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

We assessed the feasibility of a data-driven imaging biomarker based on weakly supervised learning (DIB; an imaging biomarker derived from large-scale medical image data with deep learning technology) in mammography (DIB-MG). A total of 29,107 digital mammograms from five institutions (4,339 cancer cases and 24,768 normal cases) were included. After matching patients' age, breast density, and equipment, 1,238 and 1,238 cases were chosen as validation and test sets, respectively, and the remainder were used for training. The core algorithm of DIB-MG is a deep convolutional neural network; a deep learning algorithm specialized for images. Each sample (case) is an exam composed of 4-view images (RCC, RMLO, LCC, and LMLO). For each case in a training set, the cancer probability inferred from DIB-MG is compared with the per-case ground-truth label. Then the model parameters in DIB-MG are updated based on the error between the prediction and the ground-truth. At the operating point (threshold) of 0.5, sensitivity was 75.6% and 76.1% when specificity was 90.2% and 88.5%, and AUC was 0.903 and 0.906 for the validation and test sets, respectively. This research showed the potential of DIB-MG as a screening tool for breast cancer.

Original languageEnglish
Number of pages1
JournalScientific reports
Volume8
Issue number1
DOIs
Publication statusPublished - 2018 Feb 9

Fingerprint

Mammography
Early Detection of Cancer
Biomarkers
Learning
Breast Neoplasms
Area Under Curve
Neoplasms
Technology
Equipment and Supplies
Research

All Science Journal Classification (ASJC) codes

  • General

Cite this

Kim, Eunkyung ; Kim, Hyo Eun ; Han, Kyunghwa ; Kang, Bong Joo ; Sohn, Yu Mee ; Woo, Ok Hee ; Lee, Chan Wha. / Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening : Preliminary Study. In: Scientific reports. 2018 ; Vol. 8, No. 1.
@article{3e533870f4cf4404b1d2944c37a701a7,
title = "Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening: Preliminary Study",
abstract = "We assessed the feasibility of a data-driven imaging biomarker based on weakly supervised learning (DIB; an imaging biomarker derived from large-scale medical image data with deep learning technology) in mammography (DIB-MG). A total of 29,107 digital mammograms from five institutions (4,339 cancer cases and 24,768 normal cases) were included. After matching patients' age, breast density, and equipment, 1,238 and 1,238 cases were chosen as validation and test sets, respectively, and the remainder were used for training. The core algorithm of DIB-MG is a deep convolutional neural network; a deep learning algorithm specialized for images. Each sample (case) is an exam composed of 4-view images (RCC, RMLO, LCC, and LMLO). For each case in a training set, the cancer probability inferred from DIB-MG is compared with the per-case ground-truth label. Then the model parameters in DIB-MG are updated based on the error between the prediction and the ground-truth. At the operating point (threshold) of 0.5, sensitivity was 75.6{\%} and 76.1{\%} when specificity was 90.2{\%} and 88.5{\%}, and AUC was 0.903 and 0.906 for the validation and test sets, respectively. This research showed the potential of DIB-MG as a screening tool for breast cancer.",
author = "Eunkyung Kim and Kim, {Hyo Eun} and Kyunghwa Han and Kang, {Bong Joo} and Sohn, {Yu Mee} and Woo, {Ok Hee} and Lee, {Chan Wha}",
year = "2018",
month = "2",
day = "9",
doi = "10.1038/s41598-018-21215-1",
language = "English",
volume = "8",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening : Preliminary Study. / Kim, Eunkyung; Kim, Hyo Eun; Han, Kyunghwa; Kang, Bong Joo; Sohn, Yu Mee; Woo, Ok Hee; Lee, Chan Wha.

In: Scientific reports, Vol. 8, No. 1, 09.02.2018.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Applying Data-driven Imaging Biomarker in Mammography for Breast Cancer Screening

T2 - Preliminary Study

AU - Kim, Eunkyung

AU - Kim, Hyo Eun

AU - Han, Kyunghwa

AU - Kang, Bong Joo

AU - Sohn, Yu Mee

AU - Woo, Ok Hee

AU - Lee, Chan Wha

PY - 2018/2/9

Y1 - 2018/2/9

N2 - We assessed the feasibility of a data-driven imaging biomarker based on weakly supervised learning (DIB; an imaging biomarker derived from large-scale medical image data with deep learning technology) in mammography (DIB-MG). A total of 29,107 digital mammograms from five institutions (4,339 cancer cases and 24,768 normal cases) were included. After matching patients' age, breast density, and equipment, 1,238 and 1,238 cases were chosen as validation and test sets, respectively, and the remainder were used for training. The core algorithm of DIB-MG is a deep convolutional neural network; a deep learning algorithm specialized for images. Each sample (case) is an exam composed of 4-view images (RCC, RMLO, LCC, and LMLO). For each case in a training set, the cancer probability inferred from DIB-MG is compared with the per-case ground-truth label. Then the model parameters in DIB-MG are updated based on the error between the prediction and the ground-truth. At the operating point (threshold) of 0.5, sensitivity was 75.6% and 76.1% when specificity was 90.2% and 88.5%, and AUC was 0.903 and 0.906 for the validation and test sets, respectively. This research showed the potential of DIB-MG as a screening tool for breast cancer.

AB - We assessed the feasibility of a data-driven imaging biomarker based on weakly supervised learning (DIB; an imaging biomarker derived from large-scale medical image data with deep learning technology) in mammography (DIB-MG). A total of 29,107 digital mammograms from five institutions (4,339 cancer cases and 24,768 normal cases) were included. After matching patients' age, breast density, and equipment, 1,238 and 1,238 cases were chosen as validation and test sets, respectively, and the remainder were used for training. The core algorithm of DIB-MG is a deep convolutional neural network; a deep learning algorithm specialized for images. Each sample (case) is an exam composed of 4-view images (RCC, RMLO, LCC, and LMLO). For each case in a training set, the cancer probability inferred from DIB-MG is compared with the per-case ground-truth label. Then the model parameters in DIB-MG are updated based on the error between the prediction and the ground-truth. At the operating point (threshold) of 0.5, sensitivity was 75.6% and 76.1% when specificity was 90.2% and 88.5%, and AUC was 0.903 and 0.906 for the validation and test sets, respectively. This research showed the potential of DIB-MG as a screening tool for breast cancer.

UR - http://www.scopus.com/inward/record.url?scp=85054190424&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054190424&partnerID=8YFLogxK

U2 - 10.1038/s41598-018-21215-1

DO - 10.1038/s41598-018-21215-1

M3 - Article

VL - 8

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

ER -