An improved SMOTE-based hybrid resampling (pipelining) framework for imbalance-aware breast cancer diagnosis

Authors

  • Yusuf Abubakar Kutigi Department of Computer Science, Federal University of Technology, Minna, Nigeria
  • Abdullahi Muhammad Bashir Department of Data Science, Federal University of Technology, Minna, Nigeria
  • Mohammed Abdulmalik Danlami Department of Computer Science, Federal University of Technology, Minna, Nigeria
  • Adabara Nasiru Usman Department of Microbiology, Federal University of Technology, Minna, Nigeria

Keywords:

SMOTE, ADASYN, Class imbalance, Breast cancer diagnosis, Data resampling

Abstract

Class imbalance remains a persistent challenge in medical diagnosis, particularly in breast cancer datasets, which often leads to models that overlook malignant cases. While the Synthetic Minority Oversampling Technique (SMOTE) and its variants address this issue, they frequently introduce noise and fail to preserve the intrinsic data structure. This study proposes a sequentially integrated resampling framework combining Adaptive Synthetic Sampling (ADASYN), Cluster-based oversampling and density-based filtering to enhance data quality and classification performance. The framework was evaluated using extreme gradient boosting (XGBoost) on the Wisconsin Diagnostic Breast Cancer (WDBC) dataset, which contains 569 observations and 30 real-valued diagnostic features derived from digitized images of fine-needle aspirates (FNA) of breast masses. The original data also contains an identification variable and a diagnosis variable. The identification variable was excluded from model development because it has no diagnostic meaning, while the diagnosis variable was retained as the target variable and encoded as malignant = 1 and benign = 0. Thus, only the 30 continuous diagnostic features were used as predictor variables. The dataset was split into training and testing sets. The proposed improved SMOTE achieved an accuracy of 96.49%, recall of 97.22%, precision of 97.22%, F1 score of 97.22% and area under the receiver operating characteristic curve (ROC-AUC) of 96.23%, achieving higher threshold-dependent classification performance while yielding a lower ROC-AUC than the evaluated baseline resampling approaches using the same XGBoost classifier.

0 0

References

[1] S. Susan & A. Kumar, ``The balancing trick: Optimized sampling of imbalanced datasets---a brief survey of the recent state of the art'', Engineering Reports 3 (2021) e12298. https://doi.org/10.1002/eng2.12298.

[2] M. Irfan, Z. Mushtaq, N. A. Khan, S. N. F. Mursal, S. Rahman, M. A. Magzoub, M. A. Latif, F. Althobiani, I. Khan & G. Abbas, ``A scalo gram-based CNN ensemble method with density-aware SMOTE oversampling for improving bearing fault diagnosis'', IEEE Access 11 (2023) 127783. https://doi.org/10.1109/ACCESS.2023.3332243.

[3] A. A. Khan, O. Chaudhari & R. Chandra, ``A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation'', Expert Systems with Applications 244 (2024) 122778. https://doi.org/10.1016/j.eswa.2023.122778.

[4] T. Chen & C. Guestrin, XGBoost: A scalable tree boosting system, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, USA, 2016, pp. 785--794. https://doi.org/10.1145/2939672.2939785.

[5] N. V. Chawla, K. W. Bowyer, L. O. Hall & W. P. Kegelmeyer, ``SMOTE: Synthetic minority over-sampling technique'', Journal of Artificial Intelligence Research 16 (2002) 321. https://doi.org/10.1613/jair.953.

[6] F. Gurcan & A. Soylu, ``Synthetic boosted resampling using deep generative adversarial networks: A novel approach to improve cancer prediction from imbalanced datasets'', Cancers 16 (2024) 4046. https://doi.org/10.3390/cancers16234046.

[7] T. Wongvorachan, S. He & O. Bulut, ``A comparison of undersampling, oversampling, and SMOTE methods for dealing with imbalanced classification in educational data mining'', Information 14 (2023) 54. https://doi.org/10.3390/info14010054.

[8] O. Njama-Abang, D. U. Ashishie & P. T. Bukie, ``Addressing class imbalance in lassa fever epidemic data, using machine learning: A case study with SMOTE and random forest'', Journal of the Nigerian Society of Physical Sciences 7 (2025) 2586. https://doi.org/10.46481/jnsps.2025.2586.

[9] Y. Li, Y. Yang, P. Song, L. Duan & R. Ren, ``An improved SMOTE algorithm for enhanced imbalanced data classification by expanding sample generation space'', Scientific Reports 15 (2025) 23521. https://doi.org/10.1038/s41598-025-09506-w.

[10] E. Taccaliti & J. S. Aguilar-Ruiz, ``Improving classification on imbalanced genomic data via KDE-based synthetic sampling'', BioData Mining 18 (2025) 60. https://doi.org/10.1186/s13040-025-00474-5.

[11] J. Lyu, J. Yang, Z. Su & Z. Zhu, ``LD-SMOTE: A novel local density estimation-based oversampling method for imbalanced datasets'', Symmetry 17 (2025) 160. https://doi.org/10.3390/sym17020160.

[12] H. He, Y. Bai, E. A. Garcia & S. Li, ADASYN: Adaptive synthetic sampling approach for imbalanced learning, 2008 IEEE International Joint Conference on Neural Networks, Hong Kong, China, 2008, pp. 1322--1328. https://doi.org/10.1109/IJCNN.2008.4633969.

[13] G. Douzas & F. Bacao, ``Effective data generation for imbalanced learning using conditional generative adversarial networks'', Expert Systems with Applications 91 (2018) 464. https://doi.org/10.1016/j.eswa.2017.09.030.

[14] W. H. Wolberg, W. N. Street & O. L. Mangasarian, ``Image analysis and machine learning applied to breast cancer diagnosis and prognosis'', Analytical and Quantitative Cytology and Histology 17 (1995) 77. https://pubmed.ncbi.nlm.nih.gov/7612134/.

[15] T. A. Assegie, A. O. Salau, K. Sampath, R. Govindarajan, S. Murugan & B. Lakshmi, ``Evaluation of adaptive synthetic resampling technique for imbalanced breast cancer identification'', Procedia Computer Science 235 (2024) 1000. https://doi.org/10.1016/j.procs.2024.04.095.

[16] C. P. Hardiyanti, & S. Sarina, ``Optimizing breast cancer classification using SMOTE, Boruta, and XGBoost'', Science in Information Technology Letters 6 (2025) 16. https://doi.org/10.31763/sitech.v6i1.2109.

FIG1

Published

2026-11-18

How to Cite

An improved SMOTE-based hybrid resampling (pipelining) framework for imbalance-aware breast cancer diagnosis. (2027). Journal of the Nigerian Society of Physical Sciences, 9(1), 3495. https://doi.org/10.46481/jnsps.2027.3495

How to Cite

An improved SMOTE-based hybrid resampling (pipelining) framework for imbalance-aware breast cancer diagnosis. (2027). Journal of the Nigerian Society of Physical Sciences, 9(1), 3495. https://doi.org/10.46481/jnsps.2027.3495