Breast Cancer Classification Using Machine Learning

Abstract

Breast cancer has been one of the leading forms of cancer that afflict women and has made it a significant health issue across the world. The chances of success+6ful treatment are very high when the disease is detected at an early stage. Nevertheless, there are cases when manual diagnosis can be time-consuming, and it can be subject to the experience of the medical professionals. Due to this fact, the necessity of intelligent systems that will assist doctors make their decisions faster and more precise is increasing too. The present project is devoted to the issue of breast cancer classification with the help of machine learning techniques as a diagnostic aid (Ref.no.1). The system utilizes medical data sets that have significant characteristics of cell nuclei observed in the samples of breast masses. These characteristics explain characteristics of size, shape, texture, and smoothness of the cells, and this comes in handy to identify whether the tumor is benign (non-cancerous) or malignant (cancerous). The data undergoes various preprocessing stages such as dealing with missing values, eliminating noise, and normalization of the values before the application of machine learning algorithms so as to enhance the performance of the models (Ref.no.1,11). The feature selection is also performed in order to select the most pertinent characteristics that affect the prediction. The various machine learning algorithms like Logistic Regression, Decision Tree, Support Vector machine are applied and trained with the prepared datasets. Such models are trained to acquire patterns out of the current data and make predictions as to the type of new or unseen data. Evaluation metrics such as accuracy, precision, recall, and F1-score are used to determine the performance of every model in order to make sure that the system offers reliable results. Comparing these algorithms helps identify the most effective method for breast cancer classification. The project demonstrates how AI can support doctors by reducing errors, speeding up diagnosis, and enabling cost-effective early detection, ultimately improving patient care. The core objective of this project is to highlight the transformative role of artificial intelligence in healthcare (Ref.no.2). Ultimately, this study reinforces the idea that the integration of medical expertise with intelligent technologies can significantly improve patient care and treatment planning, paving the way for smarter and more reliable healthcare systems (Ref.no3).

Introduction

Breast cancer is a widespread and potentially life-threatening disease, with early diagnosis being critical to improving survival rates and reducing treatment burdens. Conventional diagnostic methods, including mammography, biopsy, ultrasound, and histopathology, are effective but heavily rely on expert interpretation, which can lead to errors, delays, and inconsistencies. This has created a need for automated, reliable tools to support physicians.

Machine Learning (ML) offers a promising solution by analyzing large datasets of cell features—such as size, shape, texture, smoothness, compactness, and other nuclear characteristics—to distinguish between benign (non-cancerous) and malignant (cancerous) tumors. Various ML algorithms, including Logistic Regression, Decision Trees, Support Vector Machines (SVM), K-Nearest Neighbors (KNN), ensemble methods like Random Forest and Gradient Boosting, and deep learning approaches (e.g., CNNs and ANN), have been studied for breast cancer classification. Ensemble and deep learning models often achieve higher accuracy but may require more computational resources.

Data preprocessing, feature selection, and careful evaluation using metrics like accuracy, precision, recall, and F1-score are essential to enhance model performance. Recall is particularly emphasized in medical diagnosis to avoid missing cancer cases. Studies show ML can augment medical expertise by improving diagnostic consistency, speed, and accuracy, though it cannot fully replace human professionals.

The proposed methodology involves:

1. Dataset preparation – collecting, cleaning, and normalizing breast cell data.

2. Feature selection – retaining the most relevant cell characteristics.

3. Model training and testing – using algorithms like Logistic Regression, Decision Tree, and SVM.

4. Evaluation – comparing models with accuracy, precision, recall, and F1-score to select the most effective classifier.

The dataset consists of breast cell samples with numeric features describing cell nuclei and labels indicating benign or malignant tumors. The ultimate goal is a machine-learning-based tool that supports early screening, aids decision-making, and improves patient outcomes.

Conclusion

In my class project, I studied the way machine learning is applicable and can be used to classify breast cancer, which will assist in early and proper diagnoses. Breast cancer continues to be among the most significant health challenges globally, and the ability to detect it early is a crucial factor to improve its survival and treatment outcomes. My paper reveals that ML can become a handy tool in the field of medicine, contributing to the classification of the tumor as benign or malignant at the cellular level. I was systematic and began by collecting, cleaning data, then passing through feature selection, model training, and lastly, performance evaluation. The preprocessing of the dataset also enhanced its quality by eliminating unnecessary information and normalizing the values of the features. The feature selection also made the model sharper, in that it identified the most relevant attributes that actually contribute to true classification, by ensuring that the algorithms are learning something meaningful. I have tried various classification algorithms such as the Logistic Regression, Decision Tree and Support Vector Machine, and compared them. All models managed to categorize tumors rather well, which proved that ML methods can be relied upon to make such medical forecasts. I calculated accuracy, precision, recall, and F1-score; recall was particularly significant since in a clinical environment, it is very important to correctly label malignant cases. The comparison made possible choosing which algorithm would do well in classifying breast cancer in relation to its performance and consistency. The results of this project imply that machine learning has the potential to reduce human error, accelerate the analysis process, and support health professionals to make decisions. It is not supposed to take the place of doctors, but it should become an extra diagnostic partner. Medical expertise and clever computational procedures may be combined to provide more reliable and efficient medical solutions. Ultimately, the project demonstrates the application of machine learning in healthcare in real-life, and it has the potential to improve the early detection of breast cancer. As future research, bigger sample populations and closer links with real-time medical systems are studied, such models may be even more accurate and useful in daily practice. This paper resonates closely with me on how significant the correct data processing and data handling can be in medical applications that use machine learning. We completely depended on the performance of the models to not only depend on the algorithms we chose but also the cleanliness of the input data. Normalization, abandoning irrelevant attributes, and selecting appropriate attributes, among other things, contributed significantly to the accuracy.

References

[1] Arravalli, S., et al. (2025). Explainable machine learning techniques for breast cancer classification. Scientific Reports, Springer Nature. [2] Toma, M., et al. (2023). Breast cancer detection based on simplified deep learning techniques using histopathology images. Radio Science, Wiley. [3] Sureshkumar, S., et al. (2024). Hybrid CNN and extreme learning machine for breast cancer diagnosis. Journal of Personalized Medicine, MDPI. [4] Kaddes, M., et al. (2025). CNN–LSTM hybrid deep learning approach for breast cancer detection. Scientific Reports, Springer Nature. [5] Islam, M. R., et al. (2024). Explainable machine learning models for breast cancer prediction. Diagnostics, MDPI. [6] Ali, S., et al. (2023). Meta-learning ensemble framework for breast cancer classification. Diagnostics, MDPI. [7] Khalid, S., et al. (2023). Comparative analysis of machine learning classifiers for breast cancer diagnosis. Computers in Biology and Medicine, Elsevier. [8] Houfani, D., Slatnia, S., Kazar, O., & Zerhouni, N. (2020). Breast cancer classification using machine learning techniques: A comparative study. Medical Technologies Journal, 4(2). [9] Díaz, J., et al. (2024). Artificial intelligence systems for breast cancer screening: A systematic review. Journal of Clinical Medicine, MDPI. [10] Litjens, G., et al. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, Elsevier. [11] Shen, D., Wu, G., & Suk, H. I. (2017). Deep learning in medical image analysis. Annual Review of Biomedical Engineering. [12] Esteva, A., et al. (2019). A guide to deep learning in healthcare. Nature Medicine, Springer Nature. [13] Ravi, D., et al. (2017). Deep learning for health informatics. IEEE Journal of Biomedical and Health Informatics. [14] Ahmed, L., et al. (2020). Breast cancer detection using machine learning techniques. IEEE Access. [15] Chougrad, H., Zouaki, H., & Alheyane, O. (2018). Deep convolutional neural networks for breast cancer screening. Computer Methods and Programs in Biomedicine, Elsevier. [16] Spanhol, F. A., et al. (2016). Breast cancer histopathology image classification using convolutional neural networks. IJCNN. [17] Han, Z., et al. (2021). Breast cancer multi-classification from histopathology images using deep learning. Pattern Recognition, Elsevier. [18] Kumar, A., & Kaur, A. (2022). Machine learning based decision support system for breast cancer diagnosis. International Journal of Medical Informatics, Elsevier. [19] Yasmin, M., et al. (2021). Explainable artificial intelligence for medical diagnosis. Artificial Intelligence Review, Springer. [20] Nawaz, H., et al. (2022). Breast cancer detection using ensemble learning methods. Computers in Biology and Medicine, Elsevier. [21] Araújo, T., et al. (2017). Classification of breast cancer histology images using CNNs. PLoS ONE. [22] Wang, H., et al. (2020). Deep learning-based breast cancer diagnosis using histopathology images. IEEE Transactions on Medical Imaging. [23] Bardou, D., et al. (2018). Classification of breast cancer based on histology images using deep neural networks. Bioinformatics, Oxford. [24] Rakhlin, A., et al. (2018). Deep convolutional neural networks for breast cancer histology image analysis. ICPR. [25] Tiwari, A., et al. (2021). Performance analysis of machine learning algorithms for breast cancer prediction. Procedia Computer Science, Elsevier.

Copyright

Copyright © 2026 Mr. Rahul Ranjan, Mohd Rizwan Ansari, Nipunika Singh, Avida Khan. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.