Deep transfer learning for detection of breast arterial calcifications on mammograms: a comparative study.
Artificial intelligence
Breast arterial calcification
Cardiovascular diseases
Deep learning
Mammography
Journal
European radiology experimental
ISSN: 2509-9280
Titre abrégé: Eur Radiol Exp
Pays: England
ID NLM: 101721752
Informations de publication
Date de publication:
15 Jul 2024
15 Jul 2024
Historique:
received:
16
02
2024
accepted:
03
05
2024
medline:
15
7
2024
pubmed:
15
7
2024
entrez:
14
7
2024
Statut:
epublish
Résumé
Breast arterial calcifications (BAC) are common incidental findings on routine mammograms, which have been suggested as a sex-specific biomarker of cardiovascular disease (CVD) risk. Previous work showed the efficacy of a pretrained convolutional network (CNN), VCG16, for automatic BAC detection. In this study, we further tested the method by a comparative analysis with other ten CNNs. Four-view standard mammography exams from 1,493 women were included in this retrospective study and labeled as BAC or non-BAC by experts. The comparative study was conducted using eleven pretrained convolutional networks (CNNs) with varying depths from five architectures including Xception, VGG, ResNetV2, MobileNet, and DenseNet, fine-tuned for the binary BAC classification task. Performance evaluation involved area under the receiver operating characteristics curve (AUC-ROC) analysis, F The dataset exhibited a BAC prevalence of 194/1,493 women (13.0%) and 581/5,972 images (9.7%). Among the retrained models, VGG, MobileNet, and DenseNet demonstrated the most promising results, achieving AUC-ROCs > 0.70 in both training and independent testing subsets. In terms of testing F Deep transfer learning showed promise in automated BAC detection on mammograms, where relatively shallow networks demonstrated superior performances requiring shorter training times and reduced resources. Deep transfer learning is a promising approach to enhance reporting BAC on mammograms and facilitate developing efficient tools for cardiovascular risk stratification in women, leveraging large-scale mammographic screening programs. • We tested different pretrained convolutional networks (CNNs) for BAC detection on mammograms. • VGG and MobileNet demonstrated promising performances, outperforming their deeper, more complex counterparts. • Visual explanations using Grad-CAM++ highlighted VGG16's superior performance in localizing BAC.
Identifiants
pubmed: 39004645
doi: 10.1186/s41747-024-00478-6
pii: 10.1186/s41747-024-00478-6
doi:
Types de publication
Journal Article
Comparative Study
Langues
eng
Sous-ensembles de citation
IM
Pagination
80Informations de copyright
© 2024. The Author(s).
Références
Tsao CW, Aday AW, Almarzooq ZI et al (2023) Heart disease and stroke statistics—2023 update: a report from the American Heart Association. Circulation. 147:e93–e621
Timmis A, Vardas P, Townsend N et al (2022) European Society of Cardiology: cardiovascular disease statistics 2021. Eur Heart J. 43:716–99
doi: 10.1093/eurheartj/ehab892
pubmed: 35016208
Iribarren C, Chandra M, Lee C et al (2022) Breast arterial calcification: a novel cardiovascular risk enhancer among postmenopausal women. Circ Cardiovasc Imaging. 15:e013526
Suh JW, La Yun B (2018) Breast arterial calcification: a potential surrogate marker for cardiovascular disease. J Cardiovasc Imaging 26:125
doi: 10.4250/jcvi.2018.26.e20
pubmed: 30310879
pmcid: 6160812
Rotter MA, Schnatz PF, Currier AA, O’Sullivan DM (2008) Breast arterial calcifications (BACs) found on screening mammography and their association with cardiovascular disease. Menopause. 15:276–81
Moshyedi AC, Puthawala AH, Kurland RJ, O’Leary DH (1995) Breast arterial calcification: association with coronary artery disease Work in progress. Radiology 194:181–3
doi: 10.1148/radiology.194.1.7997548
pubmed: 7997548
Chadashvili T, Litmanovich D, Hall F, Slanetz PJ (2016) Do breast arterial calcifications on mammography predict elevated risk of coronary artery disease? Eur J Radiol 85:1121–4
doi: 10.1016/j.ejrad.2016.03.006
pubmed: 27161061
Magni V, Capra D, Cozzi A et al (2023) Mammography biomarkers of cardiovascular and musculoskeletal health: a review. Maturitas 167:75–81
doi: 10.1016/j.maturitas.2022.10.001
pubmed: 36308974
Bui QM, Daniels LB (2019) A review of the role of breast arterial calcification for cardiovascular risk stratification in women. Circulation 139:1094–101
doi: 10.1161/CIRCULATIONAHA.118.038092
pubmed: 30779650
Minssen L, Dao TH, Quang AV et al (2022) Breast arterial calcifications on mammography: a new marker of cardiovascular risk in asymptomatic middle age women? Eur Radiol 32:4889–97
doi: 10.1007/s00330-022-08571-3
pubmed: 35147775
Hendriks EJE, de Jong PA, van der Graaf Y, Mali WPThM, van der Schouw YT, Beulens JWJ (2015) Breast arterial calcifications: a systematic review and meta-analysis of their determinants and their association with cardiovascular events. Atherosclerosis 239:11–20
doi: 10.1016/j.atherosclerosis.2014.12.035
pubmed: 25568948
Zazzeroni L, Faggioli G, Pasquinelli G (2018) Mechanisms of arterial calcification: the role of matrix vesicles. Eur J Vasc Endovasc Surg 55:425–32
doi: 10.1016/j.ejvs.2017.12.009
pubmed: 29371036
Galiano NG, Eiro N, Martín A, Fernández-Guinea O, Martínez C del B, Vizoso FJ (2022) Relationship between arterial calcifications on mammograms and cardiovascular events: a twenty-three year follow-up retrospective cohort study. Biomedicines. 10:3227
Hendriks EJE, Beulens JWJ, Mali WPThM et al (2015) Breast arterial calcifications and their association with incident cardiovascular disease and diabetes. J Am Coll Cardiol 65:859–60
doi: 10.1016/j.jacc.2014.12.015
pubmed: 25720630
Iribarren C, Go AS, Tolstykh I, Sidney S, Johnston SC, Spring DB (2004) Breast vascular calcification and risk of coronary heart disease, stroke, and heart failure. J Womens Health 13:381–9
doi: 10.1089/154099904323087060
Margolies L, Salvatore M, Hecht HS et al (2016) Digital mammography and screening for coronary artery disease. JACC Cardiovasc Imaging 9:350–60
doi: 10.1016/j.jcmg.2015.10.022
pubmed: 27053465
Lee SC, Phillips M, Bellinge J, Stone J, Wylie E, Schultz C (2020) Is breast arterial calcification associated with coronary artery disease?—A systematic review and meta-analysis. PLoS One 15:e0236598
doi: 10.1371/journal.pone.0236598
pubmed: 32722699
pmcid: 7386618
Mantas D, Markopoulos C (2016) Screening mammography: usefulness beyond early detection of breast cancer. Atherosclerosis 248:1
doi: 10.1016/j.atherosclerosis.2016.02.019
pubmed: 26970928
Trimboli RM, Codari M, Cozzi A et al (2021) Semiquantitative score of breast arterial calcifications on mammography (BAC-SS): intra- and inter-reader reproducibility. Quant Imaging Med Surg 11:2019–27
doi: 10.21037/qims-20-560
pubmed: 33936983
pmcid: 8047367
Trimboli RM, Codari M, Bert A et al (2018) Breast arterial calcifications on mammography: intra- and inter-observer reproducibility of a semi-automatic quantification tool. Radiol Med 123:168–73
doi: 10.1007/s11547-017-0827-6
pubmed: 29086382
Cheng JZ, Chen CM, Cole EB, Pisano ED, Shen D (2012) Automated delineation of calcified vessels in mammography by tracking with uncertainty and graphical linking techniques. IEEE Trans Med Imaging 31:2143–55
doi: 10.1109/TMI.2012.2215880
pubmed: 22949053
Trimboli RM, Codari M, Guazzi M, Sardanelli F (2019) Screening mammography beyond breast cancer: breast arterial calcifications as a sex-specific biomarker of cardiovascular risk. Eur J Radiol 119:108636
doi: 10.1016/j.ejrad.2019.08.005
pubmed: 31493727
Wang J, Ding H, Bidgoli FA et al (2017) Detecting cardiovascular disease from mammograms with deep learning. IEEE Trans Med Imaging 36:1172–81
doi: 10.1109/TMI.2017.2655486
pubmed: 28113340
pmcid: 5522710
Alghamdi M, Abdel-Mottaleb M, Collado-Mesa F (2020) DU-Net: convolutional network for the detection of arterial calcifications in mammograms. IEEE Trans Med Imaging 39:3240–9
doi: 10.1109/TMI.2020.2989737
pubmed: 32324546
Guo X, O’Neill WC, Vey B et al (2021) SCU-Net: A deep learning method for segmentation and quantification of breast arterial calcifications on mammograms. Med Phys 48:5851–61
doi: 10.1002/mp.15017
pubmed: 34328661
Pan SJ, Yang Q (2010) A survey on transfer learning. Vol. 22, IEEE Transactions on Knowledge and Data Engineering. p. 1345–59
Tajbakhsh N, Shin JY, Gurudu SR et al (2017) Convolutional neural networks for medical image analysis: full training or fine tuning? Available from: http://arxiv.org/abs/1706.00712
Mobini N, Codari M, Riva F et al (2023) Detection and quantification of breast arterial calcifications on mammograms: a deep learning approach. Eur Radiol 32:6746–55
Fernández A, García S, Galar M, Prati RC, Krawczyk B, Herrera F (2018) Learning from imbalanced data sets. Springer International Publishing, Cham
doi: 10.1007/978-3-319-98074-4
Fujiwara K, Huang Y, Hori K et al (2020) Over- and under-sampling approach for extremely imbalanced and small minority data problem in health record analysis. Front Public Health 8:178
Otsu N (1979) A threshold selection method from gray-level histograms. IEEE Trans Syst Man Cybern 9:62–6
doi: 10.1109/TSMC.1979.4310076
Deepa S, Bharathi VS (2013) Efficient ROI segmentation of digital mammogram images using Otsu’s N thresholding method. Int J Eng Res Technol 2:1–6
Chollet F (2016) Xception: deep learning with depthwise separable convolutions. Available from: http://arxiv.org/abs/1610.02357
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. Available from: http://arxiv.org/abs/1409.1556
He K, Zhang X, Ren S, Sun J (2016) Identity mappings in deep residual networks. Available from: http://arxiv.org/abs/1603.05027
Howard AG, Zhu M, Chen B et al (2017) MobileNets: efficient convolutional neural networks for mobile vision applications. Available from: http://arxiv.org/abs/1704.04861
Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC (2018) MobileNetV2: inverted residuals and linear bottlenecks. Available from: http://arxiv.org/abs/1801.04381
Huang G, Liu Z, van der Maaten L, Weinberger KQ (2016) Densely connected convolutional networks. Available from: http://arxiv.org/abs/1608.06993
Deng J, Dong W, Socher R, Li LJ, Kai Li, Li Fei-Fei (2009) ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE; p. 248–55
Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. International Conference for Learning Representations. Available from: http://arxiv.org/abs/1412.6980
Loshchilov I, Hutter F (2016) SGDR: stochastic gradient descent with warm restarts. Available from: http://arxiv.org/abs/1608.03983
Perez L, Wang J (2017) The effectiveness of data augmentation in image classification using deep learning. Available from: https://doi.org/10.48550/arXiv.1712.04621
Mikolajczyk A, Grochowski M (2018) Data augmentation for improving deep learning in image classification problem. In: 2018 International Interdisciplinary PhD Workshop (IIPhDW). IEEE, Świnouście, p 117–22
Di Leo G, Sardanelli F (2020) Statistical significance: p value, 0.05 threshold, and applications to radiomics—reasons for a conservative approach. Eur Radiol Exp. 4:1–8
Chattopadhay A, Sarkar A, Howlader P, Balasubramanian VN (2018) Grad-CAM++: generalized gradient-based visual explanations for deep convolutional networks. In: 2018 IEEE Winter Conference on Applications of Computer Vision (WACV). Lake Tahoe, IEEE, p. 839–47
Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-CAM: visual explanations from deep networks via gradient-based localization. In: 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 2017. IEEE; p. 618–26
Baselli G, Codari M, Sardanelli F (2020) Opening the black box of machine learning in radiology: can the proximity of annotated cases be a way? Eur Radiol Exp. 4:30
Bressem KK, Adams LC, Erxleben C, Hamm B, Niehues SM, Vahldiek JL (2020) Comparing different deep learning architectures for classification of chest radiographs. Sci Rep. 10:13590
Tsochatzidis L, Costaridou L, Pratikakis I (2019) Deep learning for breast cancer diagnosis from mammograms — a comparative study. J Imaging. 5:37