Prostate tumors are tumors that grow in or around the prostate gland of the male reproductive system. these tumors are classified into two, that is: benign and malignant. Malignant tissues are cancerous and more dangerous as they can easily spread to other soft tissues that are easily attacked by cancer cells, such as the lungs and liver. In contrast, benign tumors are not as dangerous and can be surgically removed and never regrow. The cancerous tumors has a high mortality rate, according to GLOBOCAN 2020, there were estimated 19.3 million cases with 10 million registered deaths over the same year. Current diagnosis of prostate tumor still lacks maximum precision, with the PSA and invasive biopsy methods suffering from most false positives and negatives. The study suggested a ML algorithm blending technique from three base models, SVM, RF, and XGBoost. For maximum accuracy and overfitting, the study performed SMOTE for class balancing and correlation elimination techniques. The blended model outperformed the base model with an accuracy of 0.981 at 0.05 confidence level, followed by RF and XGboost at 0.971. In other performance metrics, blended model had the highest sensitivity and specificity of 0.979 and 0.981, respectively. XGBoost and RF shared almost the same sensitivity of 0.97 and a higher specificity of 0.98. SVM had an overall low performance and was not recommended for such a task as a standalone model. The study recommends the incorporation of a blended model over each performance, although Random Forest and XGBoost are still viable for application.
| Published in | American Journal of Mathematical and Computer Modelling (Volume 11, Issue 1) |
| DOI | 10.11648/j.ajmcm.20261101.13 |
| Page(s) | 30-38 |
| Creative Commons |
This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited. |
| Copyright |
Copyright © The Author(s), 2026. Published by Science Publishing Group |
SVM, SMOTE, RF, GLOBOCAN, PSA, RF, XGBOOST
| [1] | Barrett, J. E., & Moore, S. E. (2019). Multiparametric MRI in prostate cancer diagnosis and management. Clinical Radiology, 74(11), 841–849. |
| [2] | Bray, F., Laversanne, M., Sung, H., Ferlay, J., Siegel, R. L., Soerjomataram, I., & Jemal, A. (2024). Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 74(3), 229-263. |
| [3] | Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. |
| [4] | Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. |
| [5] | Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). ACM. |
| [6] | Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273–297. |
| [7] | Cruz, J. A., & Wishart, D. S. (2007). Applications of machine learning in cancer prediction and prognosis. Cancer Informatics, 2, 59–77. |
| [8] | Raschka, S. (2018). Model stacking and blending: A practical guide. arXiv preprint arXiv: 1810.01806. |
| [9] | Wolpert, D. H. (1992). Stacked generalization. Neural Networks, 5(2), 241–259. |
| [10] | World Health Organization. (2020). GLOBOCAN 2020: Global Cancer Observatory. International Agency for Research on Cancer (IARC). Retrieved from |
| [11] | Qi, X., Wang, K., Feng, B., Sun, X., Yang, J., Hu, Z., Zhang, M., Lv, C., Jin, L., Zhou, L., et al. (2023). Comparison of machine learning models based on multi-parametric magnetic resonance imaging and ultrasound videos for the prediction of prostate cancer. Frontiers in Oncology, 13: 1157949. |
| [12] | Singh, D., Kumar, V., Das, C. J., Singh, A., and Mehndiratta, A. (2022). Machine learning-based analysis of a semi-automated pi-rads v2. 1 scoring for prostate can- cer. Frontiers in Oncology, 12: 961985. |
| [13] | Tan, Y. G., Pang, L., Khalid, F., Poon, R., Huang, H. H., Chen, K., Tay, K. J., Lau, W., Cheng, C., Ho, H., et al. (2020). Local and systemic morbidities of denovo metastatic prostate cancer in singapore: insight from 685 consecutive patients from a large prospective uro-oncology registry. BMJ open, 10(2): e034331. |
APA Style
Odongo, K. A., Wamwea, C., Mwelu, S. (2026). Classification of Prostate Tumors for Effective Diagnosis and Treatment. American Journal of Mathematical and Computer Modelling, 11(1), 30-38. https://doi.org/10.11648/j.ajmcm.20261101.13
ACS Style
Odongo, K. A.; Wamwea, C.; Mwelu, S. Classification of Prostate Tumors for Effective Diagnosis and Treatment. Am. J. Math. Comput. Model. 2026, 11(1), 30-38. doi: 10.11648/j.ajmcm.20261101.13
Share This Article
Twitter
Linked In
Facebook
Copy
Download@article{10.11648/j.ajmcm.20261101.13,
author = {Kofuna Alfayo Odongo and Charity Wamwea and Susan Mwelu},
title = {Classification of Prostate Tumors for Effective Diagnosis and Treatment
},
journal = {American Journal of Mathematical and Computer Modelling},
volume = {11},
number = {1},
pages = {30-38},
doi = {10.11648/j.ajmcm.20261101.13},
url = {https://doi.org/10.11648/j.ajmcm.20261101.13},
eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajmcm.20261101.13},
abstract = {Prostate tumors are tumors that grow in or around the prostate gland of the male reproductive system. these tumors are classified into two, that is: benign and malignant. Malignant tissues are cancerous and more dangerous as they can easily spread to other soft tissues that are easily attacked by cancer cells, such as the lungs and liver. In contrast, benign tumors are not as dangerous and can be surgically removed and never regrow. The cancerous tumors has a high mortality rate, according to GLOBOCAN 2020, there were estimated 19.3 million cases with 10 million registered deaths over the same year. Current diagnosis of prostate tumor still lacks maximum precision, with the PSA and invasive biopsy methods suffering from most false positives and negatives. The study suggested a ML algorithm blending technique from three base models, SVM, RF, and XGBoost. For maximum accuracy and overfitting, the study performed SMOTE for class balancing and correlation elimination techniques. The blended model outperformed the base model with an accuracy of 0.981 at 0.05 confidence level, followed by RF and XGboost at 0.971. In other performance metrics, blended model had the highest sensitivity and specificity of 0.979 and 0.981, respectively. XGBoost and RF shared almost the same sensitivity of 0.97 and a higher specificity of 0.98. SVM had an overall low performance and was not recommended for such a task as a standalone model. The study recommends the incorporation of a blended model over each performance, although Random Forest and XGBoost are still viable for application.
},
year = {2026}
}
TY - JOUR T1 - Classification of Prostate Tumors for Effective Diagnosis and Treatment AU - Kofuna Alfayo Odongo AU - Charity Wamwea AU - Susan Mwelu Y1 - 2026/02/04 PY - 2026 N1 - https://doi.org/10.11648/j.ajmcm.20261101.13 DO - 10.11648/j.ajmcm.20261101.13 T2 - American Journal of Mathematical and Computer Modelling JF - American Journal of Mathematical and Computer Modelling JO - American Journal of Mathematical and Computer Modelling SP - 30 EP - 38 PB - Science Publishing Group SN - 2578-8280 UR - https://doi.org/10.11648/j.ajmcm.20261101.13 AB - Prostate tumors are tumors that grow in or around the prostate gland of the male reproductive system. these tumors are classified into two, that is: benign and malignant. Malignant tissues are cancerous and more dangerous as they can easily spread to other soft tissues that are easily attacked by cancer cells, such as the lungs and liver. In contrast, benign tumors are not as dangerous and can be surgically removed and never regrow. The cancerous tumors has a high mortality rate, according to GLOBOCAN 2020, there were estimated 19.3 million cases with 10 million registered deaths over the same year. Current diagnosis of prostate tumor still lacks maximum precision, with the PSA and invasive biopsy methods suffering from most false positives and negatives. The study suggested a ML algorithm blending technique from three base models, SVM, RF, and XGBoost. For maximum accuracy and overfitting, the study performed SMOTE for class balancing and correlation elimination techniques. The blended model outperformed the base model with an accuracy of 0.981 at 0.05 confidence level, followed by RF and XGboost at 0.971. In other performance metrics, blended model had the highest sensitivity and specificity of 0.979 and 0.981, respectively. XGBoost and RF shared almost the same sensitivity of 0.97 and a higher specificity of 0.98. SVM had an overall low performance and was not recommended for such a task as a standalone model. The study recommends the incorporation of a blended model over each performance, although Random Forest and XGBoost are still viable for application. VL - 11 IS - 1 ER -
Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
Department of Statistics and Actuarial Sciences, Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya
Information