Journal
KDD'15: PROCEEDINGS OF THE 21ST ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING
Volume -, Issue -, Pages 507-516Publisher
ASSOC COMPUTING MACHINERY
DOI: 10.1145/2783258.2783365
Keywords
Medical informatics; Phenotyping; Deep learning; Multivariate time series; Healthcare; Multi-label classification
Categories
Funding
- Alfred E. Mann Innovation in Engineering Doctoral Fellowship
- Laura P. and Leland K. Whitter Foundation
- NSF [IIS-1254206, IIS-1134990]
- Direct For Computer & Info Scie & Enginr
- Div Of Information & Intelligent Systems [1254206] Funding Source: National Science Foundation
Ask authors/readers for more resources
We apply deep learning to the problem of discovery and detection of characteristic patterns of physiology in clinical time series data. We propose two novel modifications to standard neural net training that address challenges and exploit properties that are peculiar, if not exclusive, to medical data. First, we examine a general framework for using prior knowledge to regularize parameters in the topmost layers. This framework can leverage priors of any form, ranging from formal ontologies (e.g., ICD9 codes) to data-derived similarity. Second, we describe a scalable procedure for training a collection of neural networks of different sizes but with partially shared architectures. Both of these innovations are well-suited to medical applications, where available data are not yet Internet scale and have many sparse outputs (e.g., rare diagnoses) but which have exploitable structure (e.g., temporal order and relationships between labels). However, both techniques are sufficiently general to be applied to other problems and domains. We demonstrate the empirical efficacy of both techniques on two real-world hospital data sets and show that the resulting neural nets learn interpretable and clinically relevant features.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available