{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T17:54:37Z","timestamp":1781718877950,"version":"3.54.5"},"reference-count":19,"publisher":"Oxford University Press (OUP)","issue":"1183","license":[{"start":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T00:00:00Z","timestamp":1777420800000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000272","name":"National Institute for Health Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000272","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014653","name":"Manchester Biomedical Research Centre","doi-asserted-by":"publisher","award":["IS-BRC-1215-20007"],"award-info":[{"award-number":["IS-BRC-1215-20007"]}],"id":[{"id":"10.13039\/100014653","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Medical Research Council through the ALDRAM grant"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Mammographic density is associated with increased risk of developing breast cancer. Automated estimation of density in women below normal screening age would enable earlier risk stratification. We are piloting the use of low dose mammograms at 10% of standard dose combined with models that make accurate mammographic density estimates.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Three models were trained on a joint set (107 619) of standard dose mammograms with associated density scores and their simulated low dose counterparts such that the models made predictions on standard and low dose mammograms. A second set of models was trained separately on the standard and simulated low dose mammograms. All models were tested on a held-out set from the training data and an independent dataset with 294 pairs of standard and real low dose mammograms.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The root mean squared errors (RMSE) between model predictions and density scores on standard and simulated low dose images were 8.26 (8.16-8.36) and 8.27 (8.17-8.38) respectively. The RMSE between predictions on standard and simulated low dose images for the jointly trained models was 1.91 (1.88-1.96). The RMSE of the predictions on real low dose images compared to standard dose images was 3.79 (2.75-4.99).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Deep learning models make density predictions on low dose images with similar quality as on standard dose images. Automated analysis of low dose mammograms could contribute to accurate breast cancer risk estimation in younger women enabling stratification for further monitoring and preventative therapy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Advances in knowledge<\/jats:title>\n                    <jats:p>Mammographic density can be estimated in low dose mammograms with similar quality to standard dose mammograms.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bjr\/tqag090","type":"journal-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T03:03:40Z","timestamp":1777431820000},"page":"1397-1404","source":"Crossref","is-referenced-by-count":0,"title":["Deep learning models to predict mammographic density jointly on standard dose and low dose images"],"prefix":"10.1093","volume":"99","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3681-1342","authenticated-orcid":false,"given":"Steven","family":"Squires","sequence":"first","affiliation":[{"name":"University of Exeter Department of Computer Science, , The Queen's Drive , Exeter, EX4 4QJ,","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4545-8018","authenticated-orcid":false,"given":"Alistair","family":"Mackenzie","sequence":"additional","affiliation":[{"name":"Royal Surrey NHS Foundation Trust NCCPM, , 18 Frederick Sanger Road , Guildford, GU2 7YD,","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dafydd Gareth","family":"Evans","sequence":"additional","affiliation":[{"name":"University of Manchester Division of Evolution, Infection and Genomics, School of Biological Sciences, , Oxford Road , Manchester, M13 9PL,","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sacha J","family":"Howell","sequence":"additional","affiliation":[{"name":"University of Manchester Division of Cancer Sciences, , Wilmslow Road , Manchester, M20 4BX,","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Susan M","family":"Astley","sequence":"additional","affiliation":[{"name":"University of Manchester Division of Informatics, Imaging and Data Science, , Oxford Road , Manchester, M13 9PT,","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,4,28]]},"reference":[{"key":"2026061713112600600_tqag090-B1","doi-asserted-by":"publisher","first-page":"1224","DOI":"10.1093\/jnci\/djq239","article-title":"Breast tissue composition and susceptibility to breast cancer","volume":"102","author":"Boyd","year":"2010","journal-title":"J Natl Cancer Inst"},{"key":"2026061713112600600_tqag090-B2","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1148\/radiol.2251011667","article-title":"Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations","volume":"225","author":"Kolb","year":"2002","journal-title":"Radiology"},{"key":"2026061713112600600_tqag090-B3","doi-asserted-by":"publisher","first-page":"10","DOI":"10.1186\/s13058-018-0932-z","article-title":"A comparison of five methods of measuring mammographic density: a case-control study","volume":"20","author":"Astley","year":"2018","journal-title":"Breast Cancer Res"},{"key":"2026061713112600600_tqag090-B4","doi-asserted-by":"publisher","first-page":"693","DOI":"10.1148\/radiol.2016151749","article-title":"The impact of acquisition dose on quantitative breast density estimation with digital mammography: results from ACRIN PA 4006","volume":"280","author":"Chen","year":"2016","journal-title":"Radiology"},{"key":"2026061713112600600_tqag090-B5","doi-asserted-by":"publisher","first-page":"e078555","DOI":"10.1136\/bmjopen-2023-078555","article-title":"Feasibility and acceptability of offering breast cancer risk assessment to general population women aged 30\u201339 years: a mixed-methods study protocol","volume":"14","author":"Hindmarch","year":"2024","journal-title":"BMJ Open"},{"key":"2026061713112600600_tqag090-B6","author":"Squires","year":"2020"},{"key":"2026061713112600600_tqag090-B7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1117\/1.JMI.6.3.031405","article-title":"Prediction of reader estimates of mammographic density using convolutional neural networks","volume":"6","author":"Ionescu","year":"2019","journal-title":"J Med Imag"},{"key":"2026061713112600600_tqag090-B8","doi-asserted-by":"publisher","first-page":"44506","DOI":"10.1117\/1.JMI.11.4.044506","article-title":"Capability and reliability of deep learning models to make density predictions on low-dose mammograms","volume":"11","author":"Squires","year":"2024","journal-title":"J Med Imag"},{"key":"2026061713112600600_tqag090-B9","doi-asserted-by":"publisher","first-page":"45021","DOI":"10.1088\/2057-1976\/ad470b","article-title":"Breast density prediction from low and standard dose mammograms using deep learning: effect of image resolution and model training approach on prediction quality","volume":"10","author":"Squires","year":"2024","journal-title":"Biomed Phys Eng Express"},{"key":"2026061713112600600_tqag090-B10","doi-asserted-by":"publisher","first-page":"35030","DOI":"10.1088\/2057-1976\/accaea","article-title":"The effect of variable labels on deep learning models trained to predict breast density","volume":"9","author":"Squires","year":"2023","journal-title":"Biomed Phys Eng Express"},{"key":"2026061713112600600_tqag090-B11","first-page":"109","volume-title":"Psychology of Learning and Motivation","author":"Mccloskey","year":"1989"},{"key":"2026061713112600600_tqag090-B12","doi-asserted-by":"publisher","first-page":"943","DOI":"10.1158\/1940-6207.CAPR-11-0458","article-title":"Assessing individual breast cancer risk within the UK national health service breast screening program: a new paradigm for cancer prevention","volume":"5","author":"Evans","year":"2012","journal-title":"Cancer Prev Res (Phila)"},{"key":"2026061713112600600_tqag090-B13","doi-asserted-by":"publisher","first-page":"121901","DOI":"10.1118\/1.4900819","article-title":"Image simulation and a model of noise power spectra across a range of mammographic beam qualities","volume":"41","author":"Mackenzie","year":"2014","journal-title":"Med Phys"},{"key":"2026061713112600600_tqag090-B14","doi-asserted-by":"publisher","first-page":"33502","DOI":"10.1117\/1.JMI.8.3.033502","article-title":"Validation of a mammographic image quality modification algorithm using 3D-printed breast phantoms","volume":"8","author":"Boita","year":"2021","journal-title":"J Med Imag"},{"key":"2026061713112600600_tqag090-B15","first-page":"770","author":"He","year":"2016"},{"key":"2026061713112600600_tqag090-B16","first-page":"4700","author":"Huang","year":"2017"},{"key":"2026061713112600600_tqag090-B17","doi-asserted-by":"publisher","first-page":"24502","DOI":"10.1117\/1.JMI.10.2.024502","article-title":"Automatic assessment of mammographic density using a deep transfer learning method","volume":"10","author":"Squires","year":"2023","journal-title":"J Med Imag"},{"key":"2026061713112600600_tqag090-B18","volume-title":"ImageNet: A Large-Scale Hierarchical Image Database","author":"Deng","year":"2009"},{"key":"2026061713112600600_tqag090-B19","author":"Kingma","year":"2015"}],"container-title":["British Journal of Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bjr\/advance-article-pdf\/doi\/10.1093\/bjr\/tqag090\/68182071\/tqag090.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bjr\/article-pdf\/99\/1183\/1397\/68182071\/tqag090.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bjr\/article-pdf\/99\/1183\/1397\/68182071\/tqag090.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T17:11:33Z","timestamp":1781716293000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bjr\/article\/99\/1183\/1397\/8664399"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,28]]},"references-count":19,"journal-issue":{"issue":"1183","published-online":{"date-parts":[[2026,4,28]]},"published-print":{"date-parts":[[2026,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bjr\/tqag090","relation":{},"ISSN":["0007-1285","1748-880X"],"issn-type":[{"value":"0007-1285","type":"print"},{"value":"1748-880X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,7]]},"published":{"date-parts":[[2026,4,28]]}}}