论文信息 - Medical Time-series Prediction With LSTM-MDN-ATTN

Medical Time-series Prediction With LSTM-MDN-ATTN

We introduce the LSTM-MDN-ATTN model for predicting the medical time-series data. The LSTM-MDN-ATTN model predicts the future value of medical data by approximating the distribution of target data. Since medical data is multivariate data with various test items, attention mechanism is used to model the distribution suitable for target data. The attention layer used in this study predicts target data by focusing on the distribution that is related to the target data. The proposed LSTM-MDN-ATTN model shows better results compared to baseline models using lab test data from Asan Medical Center in Seoul.

Hwin Dol Park | Youngwoong Han | Jae Hun Choi | Youngwoong Han

[1] Ruqiang Yan,et al. Learning to Monitor Machine Health with Convolutional Bi-Directional LSTM Networks , 2017, Sensors.

[2] Jürgen Schmidhuber,et al. Recurrent World Models Facilitate Policy Evolution , 2018, NeurIPS.

[3] Rouhollah Rahmatizadeh,et al. From Virtual Demonstration to Real-World Manipulation Using LSTM and MDN , 2016, AAAI.

[4] Fei Wang,et al. Patient Subtyping via Time-Aware LSTM Networks , 2017, KDD.

[5] Yoshua Bengio,et al. Show, Attend and Tell: Neural Image Caption Generation with Visual Attention , 2015, ICML.

[6] Ruqiang Yan,et al. Machine health monitoring with LSTM networks , 2016, 2016 10th International Conference on Sensing Technology (ICST).

[7] Jürgen Schmidhuber,et al. Long Short-Term Memory , 1997, Neural Computation.