Robust Deep Ensemble Learning for Mammographic Lesion Classification on the INbreast and MIAS Datasets Using Focal Loss and Misclassification-Based Refinement
Received: 24 July 2025 | Revised: 29 August 2025 and 14 September 2025 | Accepted: 24 September 2025 | Online: 14 October 2025
Corresponding author: Kavita P. Shinde
Abstract
To increase patient survival rates, breast cancer must be detected early and, most importantly, accurately. Although mammography analysis using deep learning has advanced, generalizability is still hampered by issues including dataset variability, class imbalance, and inter-class similarity. Using pre-trained DenseNet121, InceptionV3, and EfficientNetB0 models, this study presents a unique and robust ensemble-based deep learning architecture that was independently refined on mammography images from the INbreast and Mammographic Image Analysis Society (MIAS) datasets. The pipeline incorporates Contrast Limited Adaptive Histogram Equalization (CLAHE) for preprocessing, employs focal loss to mitigate class imbalance, and utilizes a two-stage refinement strategy in which misclassified samples are reintroduced for model retraining. This approach achieves significant improvements, reaching up to 99% and 97% classification accuracy on the INbreast and MIAS datasets, respectively. The ensemble model with misclassification refinement demonstrates notable robustness and generalization capability. The same pipeline was applied independently to both datasets to perform cross-dataset validation, confirming its adaptability to diverse mammographic characteristics. Accuracy, precision, recall, F1-score, and misclassification analysis were employed to evaluate the method, making it suitable for real-world clinical applications. The proposed framework holds strong potential for deployment in clinical decision support systems and lays a solid foundation for future research on small and heterogeneous medical datasets.
Keywords:
Contrast Limited Adaptive Histogram Equalization (CLAHE), focal loss, ensemble learningDownloads
References
H. Sung et al., "Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries," CA: A Cancer Journal for Clinicians, vol. 71, no. 3, pp. 209–249, 2021. DOI: https://doi.org/10.3322/caac.21660
X. Jing et al., "Automated Breast Density Assessment in MRI Using Deep Learning and Radiomics: Strategies for Reducing Inter-Observer Variability," Journal of Magnetic Resonance Imaging, vol. 60, no. 1, pp. 80–91, 2024. DOI: https://doi.org/10.1002/jmri.29058
A. F. A. Alshamrani and F. Saleh Zuhair Alshomrani, "Optimizing Breast Cancer Mammogram Classification Through a Dual Approach: A Deep Learning Framework Combining ResNet50, SMOTE, and Fully Connected Layers for Balanced and Imbalanced Data," IEEE Access, vol. 13, pp. 4815–4826, 2025. DOI: https://doi.org/10.1109/ACCESS.2024.3524633
F. Azour and A. Boukerche, "An Efficient Transfer and Ensemble Learning Based Computer Aided Breast Abnormality Diagnosis System," IEEE Access, vol. 11, pp. 21199–21209, 2023. DOI: https://doi.org/10.1109/ACCESS.2022.3192857
S. Demi̇Rel, A. Urfali, Ö. F. Bozkir, A. Çeli̇Kten, A. Budak, and H. Karataş, "Improving Mass Detection in Mammography Using Focal Loss Based RetinaNet," Turkish Journal of Forecasting, vol. 07, no. 1, pp. 1–9, Oct. 2023. DOI: https://doi.org/10.34110/forecasting.1326245
Y. Gao, J. Lin, Y. Zhou, and R. Lin, "The application of traditional machine learning and deep learning techniques in mammography: a review," Frontiers in Oncology, vol. 13, Aug. 2023, Art. no. 1213045. DOI: https://doi.org/10.3389/fonc.2023.1213045
M. A. Al-antari, M. A. Al-masni, and T. S. Kim, "Deep Learning Computer-Aided Diagnosis for Breast Lesion in Digital Mammogram," in Deep Learning in Medical Image Analysis, vol. 1213, G. Lee and H. Fujita, Eds. Springer International Publishing, 2020, pp. 59–72. DOI: https://doi.org/10.1007/978-3-030-33128-3_4
Y. Khourdifi, A. El Alami, M. Zaydi, Y. Maleh, and O. Er-Remyly, "Early Breast Cancer Detection Based on Deep Learning: An Ensemble Approach Applied to Mammograms," BioMedInformatics, vol. 4, no. 4, pp. 2338–2373, Dec. 2024. DOI: https://doi.org/10.3390/biomedinformatics4040127
H. Avcı and J. Karakaya, "A Novel Medical Image Enhancement Algorithm for Breast Cancer Detection on Mammography Images Using Machine Learning," Diagnostics, vol. 13, no. 3, Jan. 2023, Art. no. 348. DOI: https://doi.org/10.3390/diagnostics13030348
Z. Liu, J. Peng, X. Guo, S. Chen, and L. Liu, "Breast cancer classification method based on improved VGG16 using mammography images," Journal of Radiation Research and Applied Sciences, vol. 17, no. 2, Jun. 2024, Art. no. 100885. DOI: https://doi.org/10.1016/j.jrras.2024.100885
P. Aliniya, M. Nicolescu, M. Nicolescu, and G. Bebis, "Improved Loss Function for Mass Segmentation in Mammography Images Using Density and Mass Size," Journal of Imaging, vol. 10, no. 1, Jan. 2024, Art. no. 20. DOI: https://doi.org/10.3390/jimaging10010020
M. M. Dhas and N. S. Singh,"Breast Cancer Diagnosis Using Majority Voting Ensemble Classifier Approach," Journal of Soft Computing and Data Mining, vol. 5, no. 1, Jun. 2024. DOI: https://doi.org/10.30880/jscdm.2024.05.01.013
M. Khaled, F. Touazi, and D. Gaceb, "Improving Breast Cancer Diagnosis in Mammograms with Progressive Transfer Learning and Ensemble Deep Learning," Arabian Journal for Science and Engineering, vol. 50, no. 10, pp. 7697–7720, May 2025. DOI: https://doi.org/10.1007/s13369-024-09428-1
"INbreast Dataset." Kaggle, [Online]. Available: https://www.kaggle.com/datasets/ramanathansp20/inbreast-dataset.
L. Shen, L. R. Margolies, J. H. Rothstein, E. Fluder, R. McBride, and W. Sieh, "Deep Learning to Improve Breast Cancer Detection on Screening Mammography," Scientific Reports, vol. 9, no. 1, Aug. 2019, Art. no. 12495. DOI: https://doi.org/10.1038/s41598-019-48995-4
J. Suckling, "The mammographic images analysis society digital mammogram database," in Exerpta Medica. International Congress Series, 1994, vol. 1069, pp. 375–378.
S. Aymaz, "A new framework for early diagnosis of breast cancer using mammography images," Neural Computing and Applications, vol. 36, no. 4, pp. 1665–1680, Feb. 2024. DOI: https://doi.org/10.1007/s00521-023-09156-x
P. Oza, P. Sharma, S. Patel, F. Adedoyin, and A. Bruno, "Image Augmentation Techniques for Mammogram Analysis," Journal of Imaging, vol. 8, no. 5, May 2022, Art. no. 141. DOI: https://doi.org/10.3390/jimaging8050141
R. K. Pattanaik, S. Mishra, M. Siddique, T. Gopikrishna, and S. Satapathy, "Breast Cancer Classification from Mammogram Images Using Extreme Learning Machine-Based DenseNet121 Model," Journal of Sensors, vol. 2022, pp. 1–12, Dec. 2022. DOI: https://doi.org/10.1155/2022/2731364
W. M. Salama and M. H. Aly, "Deep learning in mammography images segmentation and classification: Automated CNN approach," Alexandria Engineering Journal, vol. 60, no. 5, pp. 4701–4709, Oct. 2021. DOI: https://doi.org/10.1016/j.aej.2021.03.048
S. Hu, M. Zhou, S. Jin, J. Peng, and Y. Wang, "A Comparative Study of Different Deep Learning Models Over Diverse Mammography Datasets," in 2024 IEEE International Conference on Medical Artificial Intelligence (MedAI), Chongqing, China, Nov. 2024, pp. 168–173. DOI: https://doi.org/10.1109/MedAI62885.2024.00029
A. Darakh, A. Shah, and P. Oza, "Exploring the Benefits of Data Augmentation for Breast Cancer Classification using Transfer Learning," in Information Systems for Intelligent Systems, vol. 379, C. S. In, N. D. Londhe, N. Bhatt, and M. Kitsing, Eds. Springer Nature Singapore, 2024, pp. 509–520. DOI: https://doi.org/10.1007/978-981-99-8612-5_41
A. Saber, A. G. Hussien, W. A. Awad, A. Mahmoud, and A. Allakany, "Adapting the pre-trained convolutional neural networks to improve the anomaly detection and classification in mammographic images," Scientific Reports, vol. 13, no. 1, Sep. 2023, Art. no. 14877. DOI: https://doi.org/10.1038/s41598-023-41633-0
E. Justaniah, G. Aldabbagh, A. Alhothali, and N. Abourokbah, "Classifying Breast Density from Mammogram with Pretrained CNNs and Weighted Average Ensembles," Applied Sciences, vol. 12, no. 11, May 2022, Art. no. 5599. DOI: https://doi.org/10.3390/app12115599
J. J. Puspo, "A Novel Approach to Classify Breast Cancer using Transfer Learning," in 2024 27th International Conference on Computer and Information Technology (ICCIT), Cox's Bazar, Bangladesh, Dec. 2024, pp. 3354–3359. DOI: https://doi.org/10.1109/ICCIT64611.2024.11021716
M. S. Darweesh et al., "Early breast cancer diagnostics based on hierarchical machine learning classification for mammography images," Cogent Engineering, vol. 8, no. 1, Jan. 2021, Art. no. 1968324. DOI: https://doi.org/10.1080/23311916.2021.1968324
M. T. Ahad, S. Mustofa, F. Ahmed, Y. R. Emon, and A. D. Anu, "A study on Deep Convolutional Neural Networks, Transfer Learning and Ensemble Model for Breast Cancer Detection." arXiv, 2024. DOI: https://doi.org/10.2139/ssrn.4828934
N. A. Samee, G. Atteia, S. Meshoul, M. A. Al-antari, and Y. M. Kadah, "Deep Learning Cascaded Feature Selection Framework for Breast Cancer Classification: Hybrid CNN with Univariate-Based Approach," Mathematics, vol. 10, no. 19, Oct. 2022, Art. no. 3631. DOI: https://doi.org/10.3390/math10193631
S. Khan, F. T. Zahra, A. Asif, S. Arif, and U. Fatima, "Comparative Study of Transfer Learning with VGG16 and VGG19 for Breast Cancer Classification," in 2024 26th International Multi-Topic Conference (INMIC), Karachi, Pakistan, Dec. 2024, pp. 1–6. DOI: https://doi.org/10.1109/INMIC64792.2024.11004305
D. Saraswathi and E. Srinivasan, "Mammogram Analysis using League Championship Algorithm Optimized Ensembled FCRN Classifier," Indonesian Journal of Electrical Engineering and Computer Science, vol. 5, no. 2, pp. 451–461, Feb. 2017. DOI: https://doi.org/10.11591/ijeecs.v5.i2.pp451-461
Downloads
How to Cite
License
Copyright (c) 2025 Kavita P. Shinde, Uday V. Kulkarni
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
