Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems

Anya Burton; Graham Byrnes; Jennifer Stone; Rulla M Tamimi; John Heine; Celine Vachon; Vahit Ozmen; Ana Pereira; Maria Luisa Garmendia; Christopher Scott; John H Hipwell; Caroline Dickens; Joachim Schüz; Mustafa Erkin Aribal; Kimberly Bertrand; Ava Kwong; Graham G Giles; John Hopper; Beatriz Pérez Gómez; Marina Pollán; Soo-Hwang Teo; Shivaani Mariapun; Nur Aishah Mohd Taib; Martín Lajous; Ruy Lopez-Riduara; Megan Rice; Isabelle Romieu; Anath Arzee Flugelman; Giske Ursin; Samera Qureshi; Huiyan Ma; Eunjung Lee; Reza Sirous; Mehri Sirous; Jong Won Lee; Jisun Kim; Dorria Salem; Rasha Kamal; Mikael Hartman; Hui Miao; Kee-Seng Chia; Chisato Nagata; Sudhir Vinayak; Rose Ndumia; Carla H van Gils; Johanna O P Wanders; Beata Peplonska; Agnieszka Bukowska; Steve Allen; Sarah Vinnicombe; Sue Moss; Anna M Chiarelli; Linda Linton; Gertraud Maskarinec; Martin J Yaffe; Norman F Boyd; Isabel Dos-Santos-Silva; Valerie A McCormack

doi:10.1186/s13058-016-0787-0

Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems

Anya Burton, Graham Byrnes, Jennifer Stone, Rulla M Tamimi, John Heine, Celine Vachon, Vahit Ozmen, Ana Pereira, Maria Luisa Garmendia, Christopher Scott, John H Hipwell, Caroline Dickens, Joachim Schüz, Mustafa Erkin Aribal, Kimberly Bertrand, Ava Kwong, Graham G Giles, John Hopper, Beatriz Pérez Gómez, Marina PollánSoo-Hwang Teo, Shivaani Mariapun, Nur Aishah Mohd Taib, Martín Lajous, Ruy Lopez-Riduara, Megan Rice, Isabelle Romieu, Anath Arzee Flugelman, Giske Ursin, Samera Qureshi, Huiyan Ma, Eunjung Lee, Reza Sirous, Mehri Sirous, Jong Won Lee, Jisun Kim, Dorria Salem, Rasha Kamal, Mikael Hartman, Hui Miao, Kee-Seng Chia, Chisato Nagata, Sudhir Vinayak, Rose Ndumia, Carla H van Gils, Johanna O P Wanders, Beata Peplonska, Agnieszka Bukowska, Steve Allen, Sarah Vinnicombe, Sue Moss, Anna M Chiarelli, Linda Linton, Gertraud Maskarinec, Martin J Yaffe, Norman F Boyd, Isabel Dos-Santos-Silva, Valerie A McCormack

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

BACKGROUND: Inter-women and intra-women comparisons of mammographic density (MD) are needed in research, clinical and screening applications; however, MD measurements are influenced by mammography modality (screen film/digital) and digital image format (raw/processed). We aimed to examine differences in MD assessed on these image types.

METHODS: We obtained 1294 pairs of images saved in both raw and processed formats from Hologic and General Electric (GE) direct digital systems and a Fuji computed radiography (CR) system, and 128 screen-film and processed CR-digital pairs from consecutive screening rounds. Four readers performed Cumulus-based MD measurements (n = 3441), with each image pair read by the same reader. Multi-level models of square-root percent MD were fitted, with a random intercept for woman, to estimate processed-raw MD differences.

RESULTS: Breast area did not differ in processed images compared with that in raw images, but the percent MD was higher, due to a larger dense area (median 28.5 and 25.4 cm(2) respectively, mean √dense area difference 0.44 cm (95% CI: 0.36, 0.52)). This difference in √dense area was significant for direct digital systems (Hologic 0.50 cm (95% CI: 0.39, 0.61), GE 0.56 cm (95% CI: 0.42, 0.69)) but not for Fuji CR (0.06 cm (95% CI: -0.10, 0.23)). Additionally, within each system, reader-specific differences varied in magnitude and direction (p < 0.001). Conversion equations revealed differences converged to zero with increasing dense area. MD differences between screen-film and processed digital on the subsequent screening round were consistent with expected time-related MD declines.

CONCLUSIONS: MD was slightly higher when measured on processed than on raw direct digital mammograms. Comparisons of MD on these image formats should ideally control for this non-constant and reader-specific difference.

Original language	English
Article number	18:130
Journal	Breast Cancer Research
Volume	18
DOIs	https://doi.org/10.1186/s13058-016-0787-0
Publication status	Published - 19 Dec 2016

Keywords

Breast density, Image processing, Mammographic density assessment, Breast cancer, Methods

Access to Document

10.1186/s13058-016-0787-0

87-0Final published version, 778 KB

Cite this

Burton, A., Byrnes, G., Stone, J., Tamimi, R. M., Heine, J., Vachon, C., Ozmen, V., Pereira, A., Garmendia, M. L., Scott, C., Hipwell, J. H., Dickens, C., Schüz, J., Aribal, M. E., Bertrand, K., Kwong, A., Giles, G. G., Hopper, J., Pérez Gómez, B., ... McCormack, V. A. (2016). Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems. Breast Cancer Research, 18, Article 18:130. https://doi.org/10.1186/s13058-016-0787-0

@article{b35664d544964f7c9ac07b229e16441c,

title = "Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems",

abstract = "BACKGROUND: Inter-women and intra-women comparisons of mammographic density (MD) are needed in research, clinical and screening applications; however, MD measurements are influenced by mammography modality (screen film/digital) and digital image format (raw/processed). We aimed to examine differences in MD assessed on these image types.METHODS: We obtained 1294 pairs of images saved in both raw and processed formats from Hologic and General Electric (GE) direct digital systems and a Fuji computed radiography (CR) system, and 128 screen-film and processed CR-digital pairs from consecutive screening rounds. Four readers performed Cumulus-based MD measurements (n = 3441), with each image pair read by the same reader. Multi-level models of square-root percent MD were fitted, with a random intercept for woman, to estimate processed-raw MD differences.RESULTS: Breast area did not differ in processed images compared with that in raw images, but the percent MD was higher, due to a larger dense area (median 28.5 and 25.4 cm(2) respectively, mean √dense area difference 0.44 cm (95% CI: 0.36, 0.52)). This difference in √dense area was significant for direct digital systems (Hologic 0.50 cm (95% CI: 0.39, 0.61), GE 0.56 cm (95% CI: 0.42, 0.69)) but not for Fuji CR (0.06 cm (95% CI: -0.10, 0.23)). Additionally, within each system, reader-specific differences varied in magnitude and direction (p < 0.001). Conversion equations revealed differences converged to zero with increasing dense area. MD differences between screen-film and processed digital on the subsequent screening round were consistent with expected time-related MD declines.CONCLUSIONS: MD was slightly higher when measured on processed than on raw direct digital mammograms. Comparisons of MD on these image formats should ideally control for this non-constant and reader-specific difference.",

keywords = "Breast density, Image processing, Mammographic density assessment, Breast cancer, Methods",

author = "Anya Burton and Graham Byrnes and Jennifer Stone and Tamimi, {Rulla M} and John Heine and Celine Vachon and Vahit Ozmen and Ana Pereira and Garmendia, {Maria Luisa} and Christopher Scott and Hipwell, {John H} and Caroline Dickens and Joachim Sch{\"u}z and Aribal, {Mustafa Erkin} and Kimberly Bertrand and Ava Kwong and Giles, {Graham G} and John Hopper and {P{\'e}rez G{\'o}mez}, Beatriz and Marina Poll{\'a}n and Soo-Hwang Teo and Shivaani Mariapun and Taib, {Nur Aishah Mohd} and Mart{\'i}n Lajous and Ruy Lopez-Riduara and Megan Rice and Isabelle Romieu and Flugelman, {Anath Arzee} and Giske Ursin and Samera Qureshi and Huiyan Ma and Eunjung Lee and Reza Sirous and Mehri Sirous and Lee, {Jong Won} and Jisun Kim and Dorria Salem and Rasha Kamal and Mikael Hartman and Hui Miao and Kee-Seng Chia and Chisato Nagata and Sudhir Vinayak and Rose Ndumia and {van Gils}, {Carla H} and Wanders, {Johanna O P} and Beata Peplonska and Agnieszka Bukowska and Steve Allen and Sarah Vinnicombe and Sue Moss and Chiarelli, {Anna M} and Linda Linton and Gertraud Maskarinec and Yaffe, {Martin J} and Boyd, {Norman F} and Isabel Dos-Santos-Silva and McCormack, {Valerie A}",

year = "2016",

month = dec,

day = "19",

doi = "10.1186/s13058-016-0787-0",

language = "English",

volume = "18",

journal = "Breast Cancer Research",

issn = "1465-5411",

publisher = "BioMed Central",

}

Burton, A, Byrnes, G, Stone, J, Tamimi, RM, Heine, J, Vachon, C, Ozmen, V, Pereira, A, Garmendia, ML, Scott, C, Hipwell, JH, Dickens, C, Schüz, J, Aribal, ME, Bertrand, K, Kwong, A, Giles, GG, Hopper, J, Pérez Gómez, B, Pollán, M, Teo, S-H, Mariapun, S, Taib, NAM, Lajous, M, Lopez-Riduara, R, Rice, M, Romieu, I, Flugelman, AA, Ursin, G, Qureshi, S, Ma, H, Lee, E, Sirous, R, Sirous, M, Lee, JW, Kim, J, Salem, D, Kamal, R, Hartman, M, Miao, H, Chia, K-S, Nagata, C, Vinayak, S, Ndumia, R, van Gils, CH, Wanders, JOP, Peplonska, B, Bukowska, A, Allen, S, Vinnicombe, S, Moss, S, Chiarelli, AM, Linton, L, Maskarinec, G, Yaffe, MJ, Boyd, NF, Dos-Santos-Silva, I & McCormack, VA 2016, 'Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems', Breast Cancer Research, vol. 18, 18:130. https://doi.org/10.1186/s13058-016-0787-0

TY - JOUR

T1 - Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems

AU - Burton, Anya

AU - Byrnes, Graham

AU - Stone, Jennifer

AU - Tamimi, Rulla M

AU - Heine, John

AU - Vachon, Celine

AU - Ozmen, Vahit

AU - Pereira, Ana

AU - Garmendia, Maria Luisa

AU - Scott, Christopher

AU - Hipwell, John H

AU - Dickens, Caroline

AU - Schüz, Joachim

AU - Aribal, Mustafa Erkin

AU - Bertrand, Kimberly

AU - Kwong, Ava

AU - Giles, Graham G

AU - Hopper, John

AU - Pérez Gómez, Beatriz

AU - Pollán, Marina

AU - Teo, Soo-Hwang

AU - Mariapun, Shivaani

AU - Taib, Nur Aishah Mohd

AU - Lajous, Martín

AU - Lopez-Riduara, Ruy

AU - Rice, Megan

AU - Romieu, Isabelle

AU - Flugelman, Anath Arzee

AU - Ursin, Giske

AU - Qureshi, Samera

AU - Ma, Huiyan

AU - Lee, Eunjung

AU - Sirous, Reza

AU - Sirous, Mehri

AU - Lee, Jong Won

AU - Kim, Jisun

AU - Salem, Dorria

AU - Kamal, Rasha

AU - Hartman, Mikael

AU - Miao, Hui

AU - Chia, Kee-Seng

AU - Nagata, Chisato

AU - Vinayak, Sudhir

AU - Ndumia, Rose

AU - van Gils, Carla H

AU - Wanders, Johanna O P

AU - Peplonska, Beata

AU - Bukowska, Agnieszka

AU - Allen, Steve

AU - Vinnicombe, Sarah

AU - Moss, Sue

AU - Chiarelli, Anna M

AU - Linton, Linda

AU - Maskarinec, Gertraud

AU - Yaffe, Martin J

AU - Boyd, Norman F

AU - Dos-Santos-Silva, Isabel

AU - McCormack, Valerie A

PY - 2016/12/19

Y1 - 2016/12/19

N2 - BACKGROUND: Inter-women and intra-women comparisons of mammographic density (MD) are needed in research, clinical and screening applications; however, MD measurements are influenced by mammography modality (screen film/digital) and digital image format (raw/processed). We aimed to examine differences in MD assessed on these image types.METHODS: We obtained 1294 pairs of images saved in both raw and processed formats from Hologic and General Electric (GE) direct digital systems and a Fuji computed radiography (CR) system, and 128 screen-film and processed CR-digital pairs from consecutive screening rounds. Four readers performed Cumulus-based MD measurements (n = 3441), with each image pair read by the same reader. Multi-level models of square-root percent MD were fitted, with a random intercept for woman, to estimate processed-raw MD differences.RESULTS: Breast area did not differ in processed images compared with that in raw images, but the percent MD was higher, due to a larger dense area (median 28.5 and 25.4 cm(2) respectively, mean √dense area difference 0.44 cm (95% CI: 0.36, 0.52)). This difference in √dense area was significant for direct digital systems (Hologic 0.50 cm (95% CI: 0.39, 0.61), GE 0.56 cm (95% CI: 0.42, 0.69)) but not for Fuji CR (0.06 cm (95% CI: -0.10, 0.23)). Additionally, within each system, reader-specific differences varied in magnitude and direction (p < 0.001). Conversion equations revealed differences converged to zero with increasing dense area. MD differences between screen-film and processed digital on the subsequent screening round were consistent with expected time-related MD declines.CONCLUSIONS: MD was slightly higher when measured on processed than on raw direct digital mammograms. Comparisons of MD on these image formats should ideally control for this non-constant and reader-specific difference.

AB - BACKGROUND: Inter-women and intra-women comparisons of mammographic density (MD) are needed in research, clinical and screening applications; however, MD measurements are influenced by mammography modality (screen film/digital) and digital image format (raw/processed). We aimed to examine differences in MD assessed on these image types.METHODS: We obtained 1294 pairs of images saved in both raw and processed formats from Hologic and General Electric (GE) direct digital systems and a Fuji computed radiography (CR) system, and 128 screen-film and processed CR-digital pairs from consecutive screening rounds. Four readers performed Cumulus-based MD measurements (n = 3441), with each image pair read by the same reader. Multi-level models of square-root percent MD were fitted, with a random intercept for woman, to estimate processed-raw MD differences.RESULTS: Breast area did not differ in processed images compared with that in raw images, but the percent MD was higher, due to a larger dense area (median 28.5 and 25.4 cm(2) respectively, mean √dense area difference 0.44 cm (95% CI: 0.36, 0.52)). This difference in √dense area was significant for direct digital systems (Hologic 0.50 cm (95% CI: 0.39, 0.61), GE 0.56 cm (95% CI: 0.42, 0.69)) but not for Fuji CR (0.06 cm (95% CI: -0.10, 0.23)). Additionally, within each system, reader-specific differences varied in magnitude and direction (p < 0.001). Conversion equations revealed differences converged to zero with increasing dense area. MD differences between screen-film and processed digital on the subsequent screening round were consistent with expected time-related MD declines.CONCLUSIONS: MD was slightly higher when measured on processed than on raw direct digital mammograms. Comparisons of MD on these image formats should ideally control for this non-constant and reader-specific difference.

KW - Breast density, Image processing, Mammographic density assessment, Breast cancer, Methods

U2 - 10.1186/s13058-016-0787-0

DO - 10.1186/s13058-016-0787-0

M3 - Article

C2 - 27993168

SN - 1465-5411

VL - 18

JO - Breast Cancer Research

JF - Breast Cancer Research

M1 - 18:130

ER -

Mammographic density assessed on paired raw and processed digital images and on paired screen-film and digital images across three mammography systems

Abstract

Keywords

Access to Document

Fingerprint

Cite this