Email this article Enhancing Survival Prediction in Cervical Cancer: A Hybrid Recurrent Neural Network Approach with Data Value Metric-Based Genetic Optimization
Abstract
This study develops an enhanced cervical cancer survival prediction framework using a hybrid machine learning approach that integrates a Genetic Algorithm optimized with a Data Value Metric for feature selection, a Recurrent Neural Network, and a Cox Proportional Hazards model to improve prognostic accuracy and clinical decision support. Clinical data from 100 cervical cancer patients at the University of Benin Teaching Hospital were used, with data augmentation applied to address sample size limitations and reduce overfitting. The hybrid model was trained on the expanded dataset, where the Data Value Metric served as the fitness function for optimal feature selection, while the Cox model estimated hazard ratios. Results show that the proposed RNN GA DVM framework outperformed existing Probabilistic Neural Network models, achieving 72 percent accuracy, an F measure of 0.82, and a concordance index of 0.75. Survival analysis identified age at diagnosis, chemotherapy, and years post menopause as significant prognostic factors, with chemotherapy reducing hazard by approximately 58 percent. Although data augmentation improved model robustness, the single institution dataset of 100 patients limits generalizability, motivating future multi centre validation. The framework offers practical clinical value as a decision support tool for oncologists by enabling more accurate survival forecasting and risk stratification, while highlighting key treatment and demographic predictors relevant to patient management. Societally, it contributes to improved cervical cancer outcomes in resource constrained settings by supporting earlier and more reliable prognostic assessment. The study is novel in its use of a Data Value Metric driven Genetic Algorithm for feature selection within a hybrid RNN and Cox Proportional Hazards architecture, offering a balanced combination of predictive performance and interpretability for survival analysis.
Full Text:
PDFReferences
A. E. Okpako and P. O. Asagba, “An improved framework for diagnosing confusability disease using neutrosophic based neural network,” Neutrosophic Sets and Systems, vol. 16, no. 1, p. 7, 2017.
I. A. Oludare, J. Aman, E. O. Abiodun, V. D. Kemi, N. A. Nachaat, and A. Humaira, “State-of-the-art in artificial neural network application: A survey,” Heliyon, vol. 4, no. 11, Art. no. e00938, 2018.
F. Jiang et al., “Artificial intelligence in healthcare: Past, present and future,” Stroke and Vascular Neurology, vol. 2, no. 4, pp. 230–243, 2017.
M. Schiffman and D. Solomon, “Cervical-cancer screening with human papillomavirus and cytologic cotesting,” New England Journal of Medicine, vol. 369, no. 24, pp. 2324–2331, 2013. https://doi.org/10.1056/NEJMcp1210379
World Health Organization, “Cervical cancer,” 2019. https://www.who.int/health-topics/cervical-cancer
I. Guyon and A. Elisseeff, “An introduction to variable and feature selection,” Journal of Machine Learning Research, vol. 3, pp. 1157–1182, 2003.
X. Zhu and X. Wu, “Class noise vs. attribute noise: A quantitative study of their impacts,” Artificial Intelligence Review, vol. 22, no. 3, pp. 177–210, 2005.
J. Zhang, Y. Xiong, and S. Min, “A new hybrid filter/wrapper algorithm for feature selection in classification,” Analytica Chimica Acta, vol. 1080, pp. 43–54, 2019.
P. N. Njume, A. Lyndal, M. Robert, and M. Frank, “Endometrial polyps: Pathogenesis, sequelae and treatment,” SAGE Open Medicine, vol. 7, 2019. https://doi.org/10.1177/2050312119848247
Y. D. Rocky, C. M. Gloria, and L. L. Jans, “Early warning system for cervical cancer diagnosis using Ridge Polynomial neural network and chaos optimization,” Journal of Theoretical and Applied Information Technology, vol. 96, no. 7, 2005.
P. M. M. Kumar, S. M. M. Das, and K. P. Sankar, “Staging of cervical cancer with soft computing,” IEEE Transactions on Biomedical Engineering, vol. 47, no. 7, pp. 934–940, 2000.
M. Noshad, J. W. Choi, Y. Sun, A. Hero, and D. Ivo, “A data value metric for quantifying information,” Journal of Big Data, 2021. https://doi.org/10.1186/s40537-021-00504-2
O. Bogdan, K. Maciej, S. Andrzej, O. Marzanna, and K. Jacek, “Prediction of 10 years overall survival in patients with operable cervical cancer using a probabilistic neural network,” Journal of Cancer, vol. 10, no. 18, pp. 4189–4195, 2019. https://doi.org/10.7150/jca.33945
K. Hemalatha and K. Usha Rani, “Improvement of multi-layer perceptron classification on cervical Pap smear data with feature extraction,” International Journal of Innovative Research in Science, Engineering and Technology, vol. 5, no. 12, 2016.
S. M. K. Soumya, K. Sneha, and C. Arunvinodh, “Cervical cancer detection and classification using texture analysis,” Biomedical and Pharmacology Journal, vol. 9, no. 2, 2016. http://biomedpharmajournal.org/?p=7750
M. I. Abdullah and F. Chastine, “A quantum hybrid PSO combined with fuzzy K-NN approach to feature selection and cell classification in cervical cancer detection,” 2017.
M. F. U. Mohammed, “Determining cervical cancer possibility by using machine learning methods,” International Journal of Latest Research in Engineering and Technology, vol. 3, no. 12, pp. 65–71, 2017.
Z. C. and E. P., “Comparison of multi-label classification methods for pre-diagnosis of cervical cancer,” International Journal of Intelligent Systems and Applications in Engineering, vol. 5, no. 4, pp. 232–236, 2017.
A. O. Akinrotimi and M. A. Olugbebi, “Modeling and diagnosis of cervical cancer using adaptive neuro-fuzzy inference system,” World Journal of Research and Review, vol. 6, no. 5, pp. 1–3, 2018.
B. Bargana and A. Nagarajan, “An innovation system for classifying cervical cancer using feature base ANFIS classifier,” 2018.
B. Bargana and A. Nagarajan, “An expert system for predicting the cervical cancer using data mining techniques,” International Journal of Pure and Applied Mathematics, 2018.
K. F. et al., “Supervised deep learning embeddings for the prediction of cervical cancer diagnosis,” PeerJ Computer Science, vol. 4, Art. no. e154, 2018. https://doi.org/10.7717/peerj-cs.154
O. Bogdan et al., “Prediction of 10 years overall survival in patients with operable cervical cancer using a probabilistic neural network,” Journal of Cancer, vol. 10, no. 18, pp. 4189–4195, 2019. (Duplicate of [13])
M. N. Asiedu et al., “Algorithms for automated detection of cervical pre-cancers with a low-cost, point-of-care, pocket colposcope,” bioRxiv, 2018. https://doi.org/10.1101/325419
K. Sowjanya et al., “Machine aided identification of risk factors of cervical cancer,” 2019.
O. Bogdan et al., “Prediction of 5-year overall survival in cervical cancer patients treated with radical hysterectomy using computational intelligence method,” 2019.
D. I. Acheme, A. S. Makinde, U. Osemengbe, and W. Nwankwo, “An intelligent agent-based stock market decision support system using fuzzy logic,” The IUP Journal of Information Technology, vol. 16, no. 4, pp. 1–20, 2020.
P. U. Chinedu, W. Nwankwo, F. U. Masajuwa, and S. Imoisi, “Cybercrime detection and prevention efforts in the last decade: An overview of the possibilities of machine learning models,” RIGEO, vol. 11, no. 7, pp. 956–974, 2021. https://doi.org/10.48047/rigeo.11.07.92
W. Nwankwo et al., “Application of support vector machine to Lassa fever diagnosis,” in Advances in Artificial Systems for Logistics Engineering III, Lecture Notes on Data Engineering and Communications Technologies, vol. 180, Springer, 2023, pp. 1–??.
T. C. Olayinka, A. S. Olayinka, and W. Nwankwo, “Evolving feed-forward artificial neural networks using binary and denary dataset,” SAU Science-Tech Journal, vol. 6, no. 1, pp. 96–108, 2021.
A. S. Olayinka et al., “A study on the application of Bayesian learning and decision trees IoT-enabled system in postharvest storage,” in Artificial Intelligence-based Internet of Things Systems, Springer, 2022, pp. 1–??. https://doi.org/10.1007/978-3-030-87059-1_18
Refbacks
- There are currently no refbacks.