Automated Breast Cancer Detection Using 1D-CNN with Probability Calibration and Threshold Optimization

 In this study we present a 1-Dimensional Convolutional Neural Network (1D-CNN) for automated breast cancer diagnosis based on the well-known Wisconsin Breast Cancer dataset. The model created meaningful representations directly from the feature space without manual feature engineering by converting tabular features to a suitable format for convolutional learning. Results of experiments show the robustness of the proposed approach and the test accuracy of 98% and AUC 0.9987 revealed excellent discriminative power between malignant and benign tumors. The confusion matrix and classification report also validated that this model retains balance between sensitivity and specificity almost accurately with overlapped misclassification instances. Besides predictive power, other analyses including precision-recall curve, trend of F1-score vs. threshold and reliability diagram indicated that the model is not only accurate but also well-calibrated and stable across thresholds. This specificity makes it ideal for clinical applications that will require early detection, as well as a low false-positive rate. Conclusion: The results emphasize that 1D-CNNs are consistent predictive models of breast cancer on tabular medical datasets and can be used as an assistive tool to systems software to be deployed on clinical decision-support systems leading to better early diagnosis and treatment. Incorporating the proposed interpretation within methods that address the explainability aspects of deep learning could help the clinicians trust the model outputs while extending its examination to multimodal datasets in the future.

1.                Saharan, S., Wani, N. A., Chatterji, S., Kumar, N., &Almuhaideb, A. M. (2025). A Deep Learning and Explainable Artificial Intelligence based Scheme for Breast Cancer Detection. Scientific Reports, 15(1), 32125. 2.                Murugan, T. K., Karthikeyan, P., &Sekar, P. (2025). Efficient breast cancer detection using neural networks and explainable artificial intelligence. Neural Computing and Applications, 37(5), 3759-3776. 3.                Srinivasu, P. N., Jaya Lakshmi, G., Gudipalli, A., Narahari, S. C., Shafi, J., Woźniak, M., & Ijaz, M. F. (2024). XAI-driven CatBoost multi-layer perceptron neural network for analyzing breast cancer. Scientific Reports, 14(1), 28674. 4.                Wani, N. A., Kumar, R., &Bedi, J. (2024). Harnessing fusion modeling for enhanced breast cancer classification through interpretable artificial intelligence and in-depth explanations. Engineering Applications of Artificial Intelligence, 136, 108939. 5.                SalekShahabi, M. (2026). Artificial Intelligence for Early Detection and Diagnosis of Breast Cancer: A Systematic Review of Machine Learning and Deep Learning Approaches. Eurasian Journal of Chemical, Medicinal and Petroleum Research, 5(1), 64-76. 6.                Dutta, M., Hasan, K. M. M., Akter, A., Rahman, M. H., &Assaduzzaman, M. (2024). An interpretable machine learning-based breast cancer classification using XGBoost, SHAP, and LIME. Bulletin of Electrical Engineering and Informatics, 13(6), 4306-4315. 7.                Kumar, S., Singh, J., Ravi, V., Singh, P., Al Mazroa, A., Diwakar, M., & Gupta, I. (2024). Utilizing Multi-layer Perceptron for Esophageal Cancer Classification Through Machine Learning Methods. The Open Public Health Journal, 17(1). 8.                Chhetri, B., & Kumar, B. V. (2025). Bridging Accuracy and Interpretability: Deep Learning with XAI for Breast Cancer Detection. arXiv preprint arXiv:2510.21780. 9.                Sharafaddini, A. M., Esfahani, K. K., & Mansouri, N. (2025). Deep learning approaches to detect breast cancer: a comprehensive review. Multimedia Tools and Applications, 84(21), 24079-24190. 10.             Maheswari, B. U., Aaditi, A., Avvaru, A., Tandon, A., & de Prado, R. P. (2024, April). Interpretable machine learning model for breast cancer prediction using LIME and SHAP. In 2024 IEEE 9th International Conference for Convergence in Technology (I2CT) (pp. 1-6). IEEE.