Comparison of Cox Proportional Hazards Regression and Self-Supervised Learning Algorithm in Estimating Lung Cancer Risk

Enas  Abid Alhafidh Mohamed

Enas Abid Alhafidh Mohamed Department of Statistics, Administration and Economics College, Kerbala University, Iraq

Keywords: Risk function, Surviaval function, Cox proportional regression, self-learning, algorithm, estimation, proportional hazard, hazard capacity

Abstract

This research compares three models for analysing survival data for lung cancer patients: the Cox proportional hazard model, the supervised self-learning algorithm, and a hybrid model that combines the best parts of the two models. Using MatLab, the comparison was made using multiple performance assessment criteria, such as the mean absolute error (MAE), the mean square error (MSE), the accuracy index (C-index), and Akaike's criterion (AIC). The hybrid model was more accurate than the baseline models, with an accuracy of 0.94 and reduced comparison criteria. The Cox model, on the other hand, only had an accuracy of 0.82. The risk data from the sample also indicated that advanced disease stage, smoking, age, and being male were the factors that most elevated the risk of lung cancer. On the other hand, immunotherapy and radiation lowered the risk of lung cancer. So, the hybrid model is a good way to figure out how likely someone is to die.

References

D. R. Cox, "Regression models and life-tables," J. Royal Stat. Soc.: Series B (Methodological), vol. 34, no. 2, pp. 187–220, 1972, doi: 10.1111/j.2517-6161.1972.tb00899.x.

I. Kuitunen, V. Ponkilainen, M. Uimonen, A. Reito, and J. Mäkelä, "Testing the proportional hazards assumption in Cox regression and dealing with possible non-proportionality in total joint arthroplasty research: methodological perspectives and review," BMC Musculoskelet. Disord., vol. 22, no. 1, Article 435, 2021, doi: 10.1186/s12891-021-04379-2.

A. S. Singh and S. Dlamini, Analytical Models of Survival Analysis: Concepts and Their Applications, Aug. 2021. [Online].Available: https://www.researchgate.net/publication/353726189_Analytical_Models_of_Survival_Analysis_Concepts_and_Their_Applications [Accessed: Jun. 22, 2025].

D. Collett, Modelling Survival Data in Medical Research, 2nd ed., Boca Raton: CRC, 2003, ISBN 978-1584883258.

"Proportional Hazards Model — an overview," ScienceDirect Topics. [Online]. Available: https://www.sciencedirect.com/topics/medicine-and-dentistry/proportional-hazards-model [Accessed: Jun. 22, 2025].

T. Chen, S. Kornblith, M. Norouzi, and G. Hinton, "A simple framework for contrastive learning of visual representations," arXiv preprint, arXiv:2002.05709, 2020, doi: 10.48550/arXiv.2002.05709.

N. Giakoumoglou and T. Stathaki, "A review on discriminative self-supervised learning methods," arXiv, May 2024.

L. Zheng, B. Jing, Z. Li, H. Tong, and J. He, "Heterogeneous contrastive learning for foundation models and beyond," arXiv, Apr. 2024.

N. Giakoumoglou and T. Stathaki, "A review on discriminative self-supervised learning methods," arXiv, May 2024.

M. R. Mohd Rosli, M. I. Ramli, X. Gao, et al., "Revisiting self-supervised contrastive learning for imbalanced classification," Int. J. Electr. Comput. Eng., vol. 15, no. 2, pp. 1949–1960, 2025.

I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning, The MIT Press, 2016.

A. Ramadan, K. Omar, and M. F. Mohammad, "A novel method to detect segmentation points of Arabic words using peaks and neural network," Int. J. Adv. Sci. Eng. Inf. Technol., vol. 7, no. 2, pp. 625–631, Apr. 2017, doi: 10.18517/ijaseit.7.2.1824.

B. Suvarnam and V. S. Ch, "Combination of CNN-GRU model to recognize characters of a license plate number without segmentation," in 2019 5th Int. Conf. Adv. Comput. Commun. Syst. (ICACCS), pp. 317–322, 2019, IEEE.

M. S. Gondere, L. Schmidt-Thieme, A. S. Boltena, and H. S. Jomaa, "Handwritten Amharic character recognition using a convolutional neural network," Arch. Data Sci., 2020.

A. Lamsaf, M. A. Kerroum, S. Boulaknadel, and Y. Fakhri, "Recognition of Arabic handwritten words using convolutional neural network," Indonesian J. Electr. Eng. Comput. Sci., vol. 25, no. 2, pp. 939–944, May 2022.