Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM)
Electronic medical record (EMR) offers promises for novel analytics. However, manual feature engineering from EMR is labor intensive because EMR is complex – it contains temporal, mixed-type and multimodal data packed in irregular episodes. We present a computational framework to harness EMR with mi...
| Main Authors: | , , , |
|---|---|
| Format: | Journal Article |
| Published: |
Academic Press Inc.
2015
|
| Online Access: | http://hdl.handle.net/20.500.11937/23215 |
| _version_ | 1848751088499425280 |
|---|---|
| author | Tran, The Truyen Nguyen, T. Phung, D. Venkatesh, S. |
| author_facet | Tran, The Truyen Nguyen, T. Phung, D. Venkatesh, S. |
| author_sort | Tran, The Truyen |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | Electronic medical record (EMR) offers promises for novel analytics. However, manual feature engineering from EMR is labor intensive because EMR is complex – it contains temporal, mixed-type and multimodal data packed in irregular episodes. We present a computational framework to harness EMR with minimal human supervision via restricted Boltzmann machine (RBM). The framework derives a new representation of medical objects by embedding them in a low-dimensional vector space. This new representation facilitates algebraic and statistical manipulations such as projection onto 2D plane (thereby offering intuitive visualization), object grouping (hence enabling automated phenotyping), and risk stratification. To enhance model interpretability, we introduced two constraints into model parameters: (a) nonnegative coefficients, and (b) structural smoothness. These result in a novel model called eNRBM (EMR-driven nonnegative RBM). We demonstrate the capability of the eNRBM on a cohort of 7578 mental health patients under suicide risk assessment. The derived representation not only shows clinically meaningful feature grouping but also facilitates short-term risk stratification. The F-scores, 0.21 for moderate-risk and 0.36 for high-risk, are significantly higher than those obtained by clinicians and competitive with the results obtained by support vector machines. |
| first_indexed | 2025-11-14T07:47:10Z |
| format | Journal Article |
| id | curtin-20.500.11937-23215 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T07:47:10Z |
| publishDate | 2015 |
| publisher | Academic Press Inc. |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-232152017-09-13T13:58:22Z Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) Tran, The Truyen Nguyen, T. Phung, D. Venkatesh, S. Electronic medical record (EMR) offers promises for novel analytics. However, manual feature engineering from EMR is labor intensive because EMR is complex – it contains temporal, mixed-type and multimodal data packed in irregular episodes. We present a computational framework to harness EMR with minimal human supervision via restricted Boltzmann machine (RBM). The framework derives a new representation of medical objects by embedding them in a low-dimensional vector space. This new representation facilitates algebraic and statistical manipulations such as projection onto 2D plane (thereby offering intuitive visualization), object grouping (hence enabling automated phenotyping), and risk stratification. To enhance model interpretability, we introduced two constraints into model parameters: (a) nonnegative coefficients, and (b) structural smoothness. These result in a novel model called eNRBM (EMR-driven nonnegative RBM). We demonstrate the capability of the eNRBM on a cohort of 7578 mental health patients under suicide risk assessment. The derived representation not only shows clinically meaningful feature grouping but also facilitates short-term risk stratification. The F-scores, 0.21 for moderate-risk and 0.36 for high-risk, are significantly higher than those obtained by clinicians and competitive with the results obtained by support vector machines. 2015 Journal Article http://hdl.handle.net/20.500.11937/23215 10.1016/j.jbi.2015.01.012 Academic Press Inc. fulltext |
| spellingShingle | Tran, The Truyen Nguyen, T. Phung, D. Venkatesh, S. Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title | Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title_full | Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title_fullStr | Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title_full_unstemmed | Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title_short | Learning vector representation of medical objects via EMR-driven nonnegative restricted Boltzmann machines (eNRBM) |
| title_sort | learning vector representation of medical objects via emr-driven nonnegative restricted boltzmann machines (enrbm) |
| url | http://hdl.handle.net/20.500.11937/23215 |