Paper proposes two step approach to model survival data using deep learning. In the first step, we transform each subject's survival time into a series of jackknife pseudo conditional survival probabilities and then use these pseudo probabilities as a quantitative response variable in the deep neural network model. By using the pseudo values, we reduce a complex survival analysis to a standard regression problem, which greatly simplifies the neural network construction. Our two-step approach is simple, yet very flexible in making risk predictions for survival data, which is very appealing from the practice point of view.
Lili Zhao (University of Michigan), Dai Feng (Merck Research Laboratories).
Overview
Paper proposes two step approach to model survival data using deep learning. In the first step, we transform each subject's survival time into a series of jackknife pseudo conditional survival probabilities and then use these pseudo probabilities as a quantitative response variable in the deep neural network model. By using the pseudo values, we reduce a complex survival analysis to a standard regression problem, which greatly simplifies the neural network construction. Our two-step approach is simple, yet very flexible in making risk predictions for survival data, which is very appealing from the practice point of view.
Contributions and Distinctions from Previous Works
Methods
deep learning. survival analysis. competing risk events.
Results
Cite
L. Zhao and D. Feng, "Deep Neural Networks for Survival Analysis Using Pseudo Values," in IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 11, pp. 3308-3314, Nov. 2020, doi: 10.1109/JBHI.2020.2980204.
TL;DR
Paper proposes two step approach to model survival data using deep learning. In the first step, we transform each subject's survival time into a series of jackknife pseudo conditional survival probabilities and then use these pseudo probabilities as a quantitative response variable in the deep neural network model. By using the pseudo values, we reduce a complex survival analysis to a standard regression problem, which greatly simplifies the neural network construction. Our two-step approach is simple, yet very flexible in making risk predictions for survival data, which is very appealing from the practice point of view.
Paper Link
https://ieeexplore.ieee.org/document/9034100
Author/Institution
Lili Zhao (University of Michigan), Dai Feng (Merck Research Laboratories).
Overview
Paper proposes two step approach to model survival data using deep learning. In the first step, we transform each subject's survival time into a series of jackknife pseudo conditional survival probabilities and then use these pseudo probabilities as a quantitative response variable in the deep neural network model. By using the pseudo values, we reduce a complex survival analysis to a standard regression problem, which greatly simplifies the neural network construction. Our two-step approach is simple, yet very flexible in making risk predictions for survival data, which is very appealing from the practice point of view.
Contributions and Distinctions from Previous Works
Methods
deep learning. survival analysis. competing risk events.
Results
Cite
L. Zhao and D. Feng, "Deep Neural Networks for Survival Analysis Using Pseudo Values," in IEEE Journal of Biomedical and Health Informatics, vol. 24, no. 11, pp. 3308-3314, Nov. 2020, doi: 10.1109/JBHI.2020.2980204.
Comments
Source code is freely available at: http://github.com/lilizhaoUM/DNNSurv