ShortFuse: Biomedical Time Series Representations in the Presence of Structured Information
arXiv:1705.04790
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, matching or exceeding the accuracy of models that use features engineered by domain experts.
arXiv commentManuscript under review for the Machine Learning in Healthcare Conference, 2017 (www.mucmd.org). 15 pages, 4 figures, 3 tables
References in corpus (5)
- Multi-Scale Convolutional Neural Networks for Time Series Classification
- Conditional Time Series Forecasting with Convolutional Neural Networks
- LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection
- Temporal Convolutional Neural Networks for Diagnosis from Lab Tests
- Earliness-Aware Deep Convolutional Networks for Early Time Series Classification