Article Sidebar

PDF
Date Published: 25/03/2025
Online Published: 25/03/2025
Abstract Views: 238
Views PDF: 119
DOI: https://doi.org/10.58490/ctump.2025i85.3398
Issue
No. 85 (2025)
Section
Article
How to Cite
Vo, T. T., Huynh, N. T., Vu, T. L., Nguyen, T. N. Q., & Do, V. C. N. (2025). PRELIMINARY RESULTS OF AUTOMATIC BREAST DENSITY CLASSIFICATION ON MAMOGRAPHY BY ARTIFICIAL INTELLIGENCE AT THONG NHAT HOSPITAL. Cantho Journal of Medicine and Pharmacy, 85, 179-187. https://doi.org/10.58490/ctump.2025i85.3398

PRELIMINARY RESULTS OF AUTOMATIC BREAST DENSITY CLASSIFICATION ON MAMOGRAPHY BY ARTIFICIAL INTELLIGENCE AT THONG NHAT HOSPITAL

Thanh Toan Vo1, Nguyen Thuan Huynh1,, Tri Loc Vu2, Thuy Nguyet Quynh Nguyen3, Vo Cong Nguyen Do1
1 Thong Nhat Hospital
2 Tan Tao University
3 University of Science - VNUHCM

Main Article Content

Abstract

Background: Mammography is the main method for screening and early diagnosis of breast cancer. Application of artificial intelligence in classification is necessary to bring benefits to patients and reduce inconsistency in diagnosis. Objectives: The research will build a Thong Nhat hospital dataset and combine it with open datasets to create a mammography classification model for Thong Nhat hospital. Materials and methods: All mammogram images of women screened for breast cancer at Thong Nhat Hospital were selected and labeled by 2 doctors (double blind). Their images combined with previously published dataset participate in model training. Results: Dataset included 13395 mammograms of 5506 women (698 patients from Thong Nhat hospital with an mean age were 48.7 ± 12.6). On the test dataset, the model achieved an overall accuracy of 76.8 %, in good agreement with physicians. Specifically, the sensitivity gradually increases from group A with 31.8% to group D with 92.2%. On the contrary, the specificity decreased from group A with 100% to group D with 83.5%. Conclusion: These results demonstrate the potential of applying artificial intelligence to support breast density assessment in breast cancer screening.

Article Details

Keywords

Artificial Intelligence, Breast Cancer, breast density

References

1. Bray, F., et al., Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, 2024. 74(3): p. 229-263, DOI: doi.org/10.3322/caac.21834.
2. Sickles, et al., Breast Imaging Reporting and Data System, in ACR BI-RADS® Atlas. 2013, American College of Radiology.
3. Freer, P.E., Mammographic breast density: impact on breast cancer risk and implications for screening. Radiographics, 2015. 35(2): p. 302-15.DOI: 10.1148/rg.352140106.
4. Nguyen, H.T., et al., VinDr-Mammo: A large-scale benchmark dataset for computer-aided diagnosis in full-field digital mammography. Scientific Data, 2023. 10(1): p. 277.DOI:
10.1038/s41597-023-02100-7.
5. Carr, C., et al. RSNA Screening Mammography Breast Cancer Detection. 2022; Available from: https://kaggle.com/competitions/rsna-breast-cancer-detection.
6. Kopans, D., et al. Digital Database for Screening Mammography (DDSM). 2000 2000; http://www.eng.usf.edu/cvprg/mammography/database.html.
7. Lehman, C.D., et al., Mammographic Breast Density Assessment Using Deep Learning:
Clinical Implementation. Radiology, 2019. 290(1): p. 52-58.DOI: 10.1148/radiol.2018180694.
8. Magni, V., et al., Development and Validation of an AI-driven Mammographic Breast Density Classification Tool Based on Radiologist Consensus. Radiology: Artificial Intelligence, 2022. 4(2): p. e210199.DOI: 10.1148/ryai.210199.
9. Mohamed, A.A., et al., A deep learning method for classifying mammographic breast density categories. Med Phys, 2018. 45(1): p. 314-321.DOI: 10.1002/mp.12683.
10. Sprague, B.L., et al., Variation in Mammographic Breast Density Assessments Among Radiologists in Clinical Practice: A Multicenter Observational Study. Ann Intern Med, 2016. 165(7): p. 457-464.DOI: 10.7326/m15-2934.
11. Chan, H.P. and M.A. Helvie, Deep Learning for Mammographic Breast Density Assessment and Beyond. Radiology, 2019. 290(1): p. 59-60.DOI: 10.1148/radiol.2018182116.