دورية أكاديمية
Capability and reliability of deep learning models to make density predictions on low-dose mammograms.
العنوان: | Capability and reliability of deep learning models to make density predictions on low-dose mammograms. |
---|---|
المؤلفون: | Squires S; University of Manchester, School of Health Sciences, Division of Imaging, Informatics and Data Sciences, Faculty of Biology, Medicine and Health, Manchester, United Kingdom., Mackenzie A; NCCPM, Royal Surrey NHS Foundation Trust, Guildford, United Kingdom., Evans DG; University of Manchester, School of Biological Sciences, Division of Evolution, Infection and Genomics, Manchester, Greater Manchester, United Kingdom., Howell SJ; University of Manchester, Division of Cancer Sciences, Manchester, United Kingdom., Astley SM; University of Manchester, School of Health Sciences, Division of Imaging, Informatics and Data Sciences, Faculty of Biology, Medicine and Health, Manchester, United Kingdom. |
المصدر: | Journal of medical imaging (Bellingham, Wash.) [J Med Imaging (Bellingham)] 2024 Jul; Vol. 11 (4), pp. 044506. Date of Electronic Publication: 2024 Aug 06. |
نوع المنشور: | Journal Article |
اللغة: | English |
بيانات الدورية: | Publisher: Society of Photo-Optical Instrumentation Engineers Country of Publication: United States NLM ID: 101643461 Publication Model: Print-Electronic Cited Medium: Print ISSN: 2329-4302 (Print) Linking ISSN: 23294302 NLM ISO Abbreviation: J Med Imaging (Bellingham) Subsets: PubMed not MEDLINE |
أسماء مطبوعة: | Original Publication: Bellingham, Wash. : Society of Photo-Optical Instrumentation Engineers |
مستخلص: | Purpose: Breast density is associated with the risk of developing cancer and can be automatically estimated using deep learning models from digital mammograms. Our aim is to evaluate the capacity and reliability of such models to predict density from low-dose mammograms taken to enable risk estimates for younger women. Approach: We trained deep learning models on standard-dose and simulated low-dose mammograms. The models were then tested on a mammography dataset with paired standard- and low-dose images. The effect of different factors (including age, density, and dose ratio) on the differences between predictions on standard and low doses is analyzed. Methods to improve performance are assessed, and factors that reduce the model quality are demonstrated. Results: We showed that, although many factors have no significant effect on the quality of low-dose density prediction, both density and breast area have an impact. The correlation between density predictions on low- and standard-dose images of breasts with the largest breast area is 0.985 (0.949 to 0.995), whereas that with the smallest is 0.882 (0.697 to 0.961). We also demonstrated that averaging across craniocaudal-mediolateral oblique (CC-MLO) images and across repeatedly trained models can improve predictive performance. Conclusions: Low-dose mammography can be used to produce density and risk estimates that are comparable to standard-dose images. Averaging across CC-MLO and model predictions should improve this performance. The model quality is reduced when making predictions on denser and smaller breasts. (© 2024 Society of Photo-Optical Instrumentation Engineers (SPIE).) |
References: | IEEE Trans Med Imaging. 2016 May;35(5):1322-1331. (PMID: 26915120) J Med Imaging (Bellingham). 2019 Jul;6(3):031405. (PMID: 30746393) Diagn Interv Imaging. 2017 Mar;98(3):179-190. (PMID: 28131457) Cancer Prev Res (Phila). 2012 Jul;5(7):943-51. (PMID: 22581816) Biomed Phys Eng Express. 2024 May 15;10(4):. (PMID: 38701765) J Med Imaging (Bellingham). 2023 Mar;10(2):024502. (PMID: 37034359) Breast Cancer Res. 2018 Feb 05;20(1):10. (PMID: 29402289) Biomed Phys Eng Express. 2023 Apr 19;9(3):. (PMID: 37023727) Cancer Epidemiol Biomarkers Prev. 2017 Dec;26(12):1722-1729. (PMID: 28903991) BMJ Open. 2024 Jan 10;14(1):e078555. (PMID: 38199637) J Natl Cancer Inst. 2011 May 4;103(9):744-52. (PMID: 21483019) J Med Imaging (Bellingham). 2021 May;8(3):033502. (PMID: 34026921) Br J Radiol. 2002 Apr;75(892):362-70. (PMID: 12000696) Acta Oncol. 2018 May;57(5):595-603. (PMID: 29164969) Fam Cancer. 2014 Mar;13(1):13-21. (PMID: 23733252) Med Phys. 2014 Dec;41(12):121901. (PMID: 25471961) |
فهرسة مساهمة: | Keywords: artificial intelligence; cancer risk; deep learning; low-dose mammography; machine learning; mammographic density |
تواريخ الأحداث: | Date Created: 20240808 Latest Revision: 20240809 |
رمز التحديث: | 20240809 |
مُعرف محوري في PubMed: | PMC11301609 |
DOI: | 10.1117/1.JMI.11.4.044506 |
PMID: | 39114539 |
قاعدة البيانات: | MEDLINE |
كن أول من يترك تعليقا!