Using a Multi-Task Recurrent Neural Network With Attention Mechanisms to Predict Hospital Mortality of Patients.
Ruoxi YuYali ZhengRuikai ZhangYuqi JiangCarmen C. Y. PoonPublished in: IEEE J. Biomed. Health Informatics (2020)
Keyphrases
- recurrent neural networks
- multi task
- intensive care unit
- patient outcomes
- intensive care
- multi task learning
- prognostic models
- high risk
- vital signs
- learning tasks
- medical data
- neural network
- multiple tasks
- multi class
- complex valued
- recurrent networks
- clinical data
- learning problems
- feature selection
- gaussian processes
- patient groups
- risk factors
- early diagnosis
- feed forward
- artificial neural networks
- patient data
- sparse learning
- logistic regression
- health care
- transfer learning
- patient care
- information gain
- learning algorithm
- medical practice
- machine learning
- heart rate
- support vector
- fuzzy logic
- supervised learning
- mutual information