Deep learning-based survival analysis models: principles, implementation, and applications_Chinese Journal of Evidence-Based Medicine

Authors：

XIAO Xiaochun ¹ , LI Dongdong ¹ , KUANG Mengling ¹ , QIN Yingyi ¹ , GUO Yibin ¹ , LI Xian ¹ ,  WANG Zhiyong ² ,  WU Cheng ¹

1. Department of Military Health Statistics, Naval Medical University, Shanghai 200433, P. R. China;
2. Department of Information Management, Shanghai Fourth People's Hospital, Shanghai 200434, P. R. China;

Corresponding?author：

WANG Zhiyong, Email: pbkiller@126.com; WU Cheng, Email: wucheng_wu@126.com

Keywords：

Survival analysis; Deep learning; Censoring unbiased loss

DOI：

10.7507/1672-2531.202510114

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Objective To introduce and validate the censoring unbiased deep learning (CUDL) algorithm suitable for survival data and to explore its utility in survival analysis. Methods The principles of the CUDL algorithm were detailed, followed by an introduction to its software implementation steps and parameter settings. Using the construction of a long-term survival prognosis model for patients with myocardial infarction (MI) as a case study, the algorithms were compared with the conventional Cox proportional hazards model and random survival forest model. Results In the task of predicting the 5-year survival probability of patients, the mean Brier score (BS) of both the Buckley-James censoring unbiased deep learning (BJDL) model and the regularized BJDL were superior to that of the Cox proportional hazards model (mean BS: 0.1722 and 0.1715 vs. 0.1765, P<0.001). The regularized BJDL also outperformed the random survival forest model (RSF500) (mean BS: 0.1715 vs. 0.1750, P<0.001). Conclusion The BJDL can effectively leverage censored information within survival data to achieve accurate predictions without the need to satisfy the proportional hazards assumption. It provides a novel methodological alternative for the analysis of medical survival data.

Citation： XIAO Xiaochun, LI Dongdong, KUANG Mengling, QIN Yingyi, GUO Yibin, LI Xian, WANG Zhiyong, WU Cheng. Deep learning-based survival analysis models: principles, implementation, and applications. Chinese Journal of Evidence-Based Medicine, 2026, 26(6): 717-722. doi: 10.7507/1672-2531.202510114 Copy

1.
2.
3.	焦志剛. 基于機器學習的生存分析方法探討. 南京: 東南大學, 2022.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.	Goldberg Y. A primer on neural network models for natural language processing. arXiv, 2015, arXiv: 1510.00726.
27.	He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. arXiv, 2015, arXiv. 1512.03385.
28.	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. arXiv, 2017, arXiv. 1706.03762.

1.
2.
3. 焦志剛. 基于機器學習的生存分析方法探討. 南京: 東南大學, 2022.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26. Goldberg Y. A primer on neural network models for natural language processing. arXiv, 2015, arXiv: 1510.00726.
27. He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. arXiv, 2015, arXiv. 1512.03385.
28. Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. arXiv, 2017, arXiv. 1706.03762.

Chinese Journal of Evidence-Based Medicine

Deep learning-based survival analysis models: principles, implementation, and applications

Abstract Full text Figures/Tables Video References Cited by

Previous Article

Next Article

Format

Content