A novel approach for personalized response model: deep learning with individual dropout feature ranking
- PDF / 3,631,758 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 40 Downloads / 204 Views
ORIGINAL PAPER
A novel approach for personalized response model: deep learning with individual dropout feature ranking Ruihao Huang1 • Qi Liu2 • Ge Feng1 • Yaning Wang2 • Chao Liu2 • Mathangi Gopalakrishnan3 Xiangyu Liu2 • Yutao Gong4 • Hao Zhu2,5
•
Received: 23 January 2020 / Accepted: 8 October 2020 Ó This is a U.S. Government work and not under copyright protection in the US; foreign copyright protection may apply 2020
Abstract Deep learning is the fastest growing field in artificial intelligence and has led to many transformative innovations in various domains. However, lack of interpretability sometimes hinders its application in hypothesis-driven domains such as biology and healthcare. In this paper, we propose a novel deep learning model with individual feature ranking. Several simulated datasets with the scenarios that contributing features are correlated and buried among non-contributing features were used to characterize the novel analysis approach. A publicly available clinical dataset was also applied. The performance of the individual level dropout feature ranking model was compared with commonly used artificial neural network model, random forest model, and population level dropout feature ranking model. The individual level dropout feature ranking model provides a reasonable prediction of the outcomes. Unlike the random forest model and population level dropout feature ranking model, which can only identify global-wise contributing features (i.e., at population level), the individual level dropout feature ranking model allows further identification of impactful features on response at individual level. Therefore, it provides a basis for clustering patients into subgroups. This may provide a new tool for enriching patients in clinical drug development and developing personalized or individualized medicine. Keywords Variational lower bound Individual dropout feature ranking Deep learning Machine learning Artificial intelligence
Introduction Disclaimer: The views expressed in this paper are those of the authors and do not necessarily represent the views of the FDA.
Electronic supplementary material The online version of this article (https://doi.org/10.1007/s10928-020-09724-x) contains supplementary material, which is available to authorized users.
Rational and individualized treatment decision relies on adequate understanding of the heterogeneity of patients. To realize the full potential of precision medicine, it will be critical to identify factors that will impact individual’s responses, in terms of desired (i.e., efficacy) and undesired (i.e., adverse events) response to each treatment, and to 4
Oncology Center of Excellence, Office of Hematology and Oncology Products, US Food and Drug Administration, Silver Spring, MD, USA
5
Division of Pharmacometrics, Office of Clinical Pharmacology, Office of Translational Sciences/CDER, US Food and Drug Administration, 10903 New Hampshire Avenue, Silver Spring, MD 20993, USA
& Hao Zhu [email protected] 1
Department of Mathematical Science
Data Loading...
