File(s) under permanent embargo
Deepr: a convolutional net for medical records
journal contribution
posted on 2017-01-01, 00:00 authored by Phuoc NguyenPhuoc Nguyen, Truyen TranTruyen Tran, Nilmini Wickramasinghe, Svetha VenkateshSvetha VenkateshFeature engineering remains a major bottleneck when creating predictive systems from electronic medical records. At present, an important missing element is detecting predictive regular clinical motifs from irregular episodic records. We present Deepr (short for Deep record), a new end-to-end deep learning system that learns to extract features from medical records and predicts future risk automatically. Deepr transforms a record into a sequence of discrete elements separated by coded time gaps and hospital transfers. On top of the sequence is a convolutional neural net that detects and combines predictive local clinical motifs to stratify the risk. Deepr permits transparent inspection and visualization of its inner working. We validate Deepr on hospital data to predict unplanned readmission after discharge. Deepr achieves superior accuracy compared to traditional techniques, detects meaningful clinical motifs, and uncovers the underlying structure of the disease and intervention space.
History
Journal
IEEE journal of biomedical and health informaticsVolume
21Issue
1Pagination
22 - 30Publisher
IEEELocation
Piscataway, N.J.Publisher DOI
ISSN
2168-2194Language
engPublication classification
C Journal article; C1 Refereed article in a scholarly journalCopyright notice
2016, IEEEUsage metrics
Categories
No categories selectedKeywords
convolutional neural networksdeep learningmedical recordsScience & TechnologyTechnologyLife Sciences & BiomedicineComputer Science, Information SystemsComputer Science, Interdisciplinary ApplicationsMathematical & Computational BiologyMedical InformaticsComputer ScienceNEURAL-NETWORKSHEALTHREPRESENTATIONPREDICTMODELS
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC