Abstract

Breast cancer is one of the most prevalent types of cancer in women worldwide. Regarding deaths caused by cancer, breast cancer is the second leading cause of cancer-related mortality. Fine-needle aspiration (FNA) is a method for detecting breast cancer at an early stage; however, it has limitations. These limitations include a small number of samples, which could be one of the factors for diagnostic errors, and the level of experience of the person performing the procedure. The application of machine learning to solve some problems in the healthcare sector includes cancer detection using machine learning algorithms such as XGBoost and LightGBM. The two learning algorithms, XGBoost and LightGBM, are efficient machine learning algorithms that differ in terms of how they learn:level-wise learning for XGBoost and leaf-wise learning for LightGBM. We compared both XGBoost and LightGBM in terms of accuracy, sensitivity, and specificity to determine which would perform best in classification. From the results obtained in the experiments, XGBoost performed better, with an average accuracy, sensitivity, and specificity of 97.03 %, 97.40%, and 96.81 %, respectively. The average accuracy, sensitivity, and specificity of LightGBM were 95.59%, 94.70%, and 96.10%, respectively.

References

• World Health Organization (WHO). Global Breast Cancer Initiative (GBCC). Available at: https://www.who.int/initiatives/global-breast-cancer-initiative
• Hu, K. et al. Global patterns and trends in breast cancer incidence and mortality according to sociodemographic indices: an observational study based on the global burden of disease. BMJ Open, 9, e028461 (2019).
• LIV Hospital. 12 Essential Facts About Breast Cancer: Global Statistics, Risks, and Outcomes. Available at: https://int.livhospital.com/12-essential-facts-about-boob-cancer-global-statistics-risks-and-outcomes/
• International Agency for Research on Cancer (IARC). Cancer Over Time. Available at: https://gco.iarc.who.int/overtime/en/about
• World Cancer Research Fund International (WCRFI). Breast cancer statistics. Available at: https://www.wcrf.org/preventing-cancer/cancer-statistics/breast-cancer-statistics/
• Arnold, M. et al. Current and future burdens of breast cancer: Global statistics for 2020 and 2040. The Breast, 66, 15–23 (2022). PMC9465273.
• Painless mammogram. Early Detection of Breast Cancer. Available at: https://thepainlessmammogram.com/early-detection-in-breast-cancer/
• Unger-Saldaña K. Challenges to the early diagnosis and treatment of breast cancer in developing countries. World Journal of Clinical Oncology, 5(3), 465–477 (2014). PMC3437146.
• Frontiers in Oncology. Advanced Diagnostic Techniques and Wearable Technologies for Breast Cancer Detection. Front. Oncol. (2025). doi: 10.3389/fonc.2025.1587517.
• Ibid. [Same as Ref. 9].
• Cleveland Clinic. Fine-needle aspiration (FNA). Available at: https://my.clevelandclinic.org/health/diagnostics/17872-fine-needle-aspiration-fna
• Willems, S.M. et al. Diagnostic cytopathology of fine-needle aspiration biopsies of breast lesions. Journal of Clinical Pathology (2016). PMC5019018.
• Moschetta, M. et al. Role of fine-needle aspiration in breast lesions. Diagnostic and Interventional Radiology, 20(4), 300–306 (2014). PMC4063267.
• Computers Biol. Medicine Human Error in Manual Breast Cancer Analysis. Comput Biol Med (2025). doi: 10.1016/j.compbiomed.2025.01.015.
• Hoff, S.R. et al. Missed and true intervals and screen-detected breast cancers in a population-based screening program. Academic Radiology (2018). PMC5980911.
• Ibid. [Same as Ref. 15].
• Richards MA et al. Influence of delay in the presentation and treatment of symptomatic breast cancer on survival. British Journal of Cancer, 85(4), 478–480. (1999). PMC3332709.
• Ong, M.S. & Mandl, K.D. National expenditure for false-positive mammograms and overdiagnosis of breast cancer. Ann Intern Medicine (2015). PMC3079487.
• Li Y et al. Machine learning models for predicting distant metastasis in breast cancer using clinical blood markers and pathological data. BMC Cancer (2024). PMC11211559.
• Darbandi, N. et al. Advancements in AI for Breast Cancer Diagnosis: A 2024–2025 Systematic Review Update. In: AI Revolution: Research, Ethics and Society, pp. 157–165 (2026). doi: 10.1007/978-3-032-13056-3_13.
• Yilmaz, A. et al. Comparison of classification success rates of different machine learning methods for breast cancer diagnosis. Diagnostics, 13(4), 651 (2023). PMC9924317.
• Nahid, A.A. & Kong, Y. Classification Prediction of Breast Cancer Using Machine Learning. Neural Comput Appl (2023). PMC9848804.
• Anitha, P., Surya, V., Geetha, R., Loganathan, E., Singaravel, G., & Prakash, N. (2025). High-accuracy breast cancer classification utilizing XGBoost and support vector machine. Https://doi.org/10.1109/icdici66477.2025.11135150
• Kanimozhi, J. K., Bavya, G., Kalyanasundaram, P., Anitha, P., Geetha, R., & Loganathan, E. (2025). Evaluating the effectiveness of Random Forest + XGBoost versus support vector machines in achieving high accuracy and precision for breast cancer classification. Https://doi.org/10.1109/icscsa66339.2025.11170741
• Permana, A., & Romdendine, M. F. (2024). The ensemble learning approach revealed significant clinical attributes in real-world breast cancer cases. G-Tech. Https://doi.org/10.33379/gtech.v8i2.4044
• Athisayamani, S., Tamilazhagan, S., Singh, A. R., Hwang, J.-Y., & Joshi, G. P. (2025). A novel double machine learning approach for the early detection of breast cancer. Scientific Reports 15, 22971.
• Khan, M. Z., Srivastava, P., Aljubayri, I., Pattanaik, P., Ahmad, J., & Chelabi, K. (2024). Optimized breast cancer detection using genetic algorithms and ensemble machine learning models. Journal Electrical Systems.
• Karamti, H., Alharthi, R., Umer, M., Shaiba, H., Ishaq, A., Abuzinadah, N., Alsubai, S., & Ashraf, I. (2024). Breast cancer detection employing a stacked ensemble model with convolutional features. Computational Biology and Medicine.
• Pinheiro, J. M. H., & Becker, M. (2024). Breast cancer classification using gradient boosting algorithms with explainability. ArXiv preprint.
• Sait, A. R. W., & Nagaraj, R. (2024). Enhanced LightGBM-based breast cancer detection using mammographic images. Diagnostics, 14(2), 227.
• Ghosh, P., & Chatterjee, D. (2024). Comparative analysis of machine learning algorithms for breast cancer classification. BioRxiv.