The Community for Technology Leaders
2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (2017)
Kansas City, MO, USA
Nov. 13, 2017 to Nov. 16, 2017
ISBN: 978-1-5090-3051-4
pp: 811-816
Qiuling Suo , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
Fenglong Ma , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
Ye Yuan , College of Information and Communication Engineering Beijing University of Technology, Beijing, China
Mengdi Huai , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
Weida Zhong , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
Aidong Zhang , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
Jing Gao , Department of Computer Science and Engineering State University of New York at Buffalo, NY, USA
ABSTRACT
Predicting patients' risk of developing certain diseases is an important research topic in healthcare. Personalized predictive modeling, which focuses on building specific models for individual patients, has shown its advantages on utilizing heterogeneous health data compared to global models trained on the entire population. Personalized predictive models use information from similar patient cohorts, in order to capture the specific characteristics. Accurately identifying and ranking the similarity among patients based on their historical records is a key step in personalized modeling. The electric health records (EHRs), which are irregular sampled and have varied patient visit lengths, cannot be directly used to measure patient similarity due to lack of an appropriate vector representation. In this paper, we build a novel time fusion CNN framework to simultaneously learn patient representations and measure pairwise similarity. Compared to a traditional CNN, our time fusion CNN can learn not only the local temporal relationships but also the contributions from each time interval. Along with the similarity learning process, the output information which is the probability distribution is used to rank similar patients. Utilizing the similarity scores, we perform personalized disease predictions, and compare the effect of different vector representations and similarity learning metrics.
INDEX TERMS
Predictive models, Diseases, Convolution, Data models, Training, Buildings
CITATION

Q. Suo et al., "Personalized disease prediction using a CNN-based similarity learning method," 2017 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Kansas City, MO, USA, 2017, pp. 811-816.
doi:10.1109/BIBM.2017.8217759
464 ms
(Ver 3.3 (11022016))