ShortFuse: Biomedical Time Series Representations in the Presence of Structured Information
Abstract
In healthcare applications, temporal variables that encode movement, health status, and longitudinal patient evolution are often accompanied by rich structured information such as demographics, diagnostics and medical exam data. However, current methods do not jointly optimize over structured covariates and time series in the feature extraction process. We present ShortFuse, a method that boosts the accuracy of deep learning models for time series by explicitly modeling temporal interactions and dependencies with structured covariates. ShortFuse introduces hybrid convolutional and LSTM cells that incorporate the covariates via weights that are shared across the temporal domain. ShortFuse outperforms competing models by 3% on two biomedical applications, forecasting osteoarthritis-related cartilage degeneration and predicting surgical outcomes for cerebral palsy patients.
Cite
Text
Fiterau et al. "ShortFuse: Biomedical Time Series Representations in the Presence of Structured Information." Proceedings of the 2nd Machine Learning for Healthcare Conference, 2017.Markdown
[Fiterau et al. "ShortFuse: Biomedical Time Series Representations in the Presence of Structured Information." Proceedings of the 2nd Machine Learning for Healthcare Conference, 2017.](https://mlanthology.org/mlhc/2017/fiterau2017mlhc-shortfuse/)BibTeX
@inproceedings{fiterau2017mlhc-shortfuse,
title = {{ShortFuse: Biomedical Time Series Representations in the Presence of Structured Information}},
author = {Fiterau, Madalina and Bhooshan, Suvrat and Fries, Jason and Bournhonesque, Charles and Hicks, Jennifer and Halilaj, Eni and Re, Christopher and Delp, Scott},
booktitle = {Proceedings of the 2nd Machine Learning for Healthcare Conference},
year = {2017},
pages = {59-74},
volume = {68},
url = {https://mlanthology.org/mlhc/2017/fiterau2017mlhc-shortfuse/}
}