Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration
Predictive models built using temporal data in electronic health records (EHRs) can potentially play a major role in improving management of chronic diseases. However, these data present a multitude of technical challenges, including irregular sampling of data and varying length of available patient...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Elsevier
2016
|
| Online Access: | http://hdl.handle.net/1721.1/101133 https://orcid.org/0000-0003-0992-0906 |
| _version_ | 1811080344861933568 |
|---|---|
| author | Singh, Anima Nadkarni, Girish Gottesman, Omri Ellis, Stephen B. Bottinger, Erwin P. Guttag, John V. |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Singh, Anima Nadkarni, Girish Gottesman, Omri Ellis, Stephen B. Bottinger, Erwin P. Guttag, John V. |
| author_sort | Singh, Anima |
| collection | MIT |
| description | Predictive models built using temporal data in electronic health records (EHRs) can potentially play a major role in improving management of chronic diseases. However, these data present a multitude of technical challenges, including irregular sampling of data and varying length of available patient history. In this paper, we describe and evaluate three different approaches that use machine learning to build predictive models using temporal EHR data of a patient.
The first approach is a commonly used non-temporal approach that aggregates values of the predictors in the patient’s medical history. The other two approaches exploit the temporal dynamics of the data. The two temporal approaches vary in how they model temporal information and handle missing data. Using data from the EHR of Mount Sinai Medical Center, we learned and evaluated the models in the context of predicting loss of estimated glomerular filtration rate (eGFR), the most common assessment of kidney function.
Our results show that incorporating temporal information in patient’s medical history can lead to better prediction of loss of kidney function. They also demonstrate that exactly how this information is incorporated is important. In particular, our results demonstrate that the relative importance of different predictors varies over time, and that using multi-task learning to account for this is an appropriate way to robustly capture the temporal dynamics in EHR data. Using a case study, we also demonstrate how the multi-task learning based model can yield predictive models with better performance for identifying patients at high risk of short-term loss of kidney function. |
| first_indexed | 2024-09-23T11:29:51Z |
| format | Article |
| id | mit-1721.1/101133 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T11:29:51Z |
| publishDate | 2016 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | mit-1721.1/1011332022-10-01T04:02:58Z Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration Singh, Anima Nadkarni, Girish Gottesman, Omri Ellis, Stephen B. Bottinger, Erwin P. Guttag, John V. Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Singh, Anima Guttag, John V. Predictive models built using temporal data in electronic health records (EHRs) can potentially play a major role in improving management of chronic diseases. However, these data present a multitude of technical challenges, including irregular sampling of data and varying length of available patient history. In this paper, we describe and evaluate three different approaches that use machine learning to build predictive models using temporal EHR data of a patient. The first approach is a commonly used non-temporal approach that aggregates values of the predictors in the patient’s medical history. The other two approaches exploit the temporal dynamics of the data. The two temporal approaches vary in how they model temporal information and handle missing data. Using data from the EHR of Mount Sinai Medical Center, we learned and evaluated the models in the context of predicting loss of estimated glomerular filtration rate (eGFR), the most common assessment of kidney function. Our results show that incorporating temporal information in patient’s medical history can lead to better prediction of loss of kidney function. They also demonstrate that exactly how this information is incorporated is important. In particular, our results demonstrate that the relative importance of different predictors varies over time, and that using multi-task learning to account for this is an appropriate way to robustly capture the temporal dynamics in EHR data. Using a case study, we also demonstrate how the multi-task learning based model can yield predictive models with better performance for identifying patients at high risk of short-term loss of kidney function. 2016-02-09T16:11:32Z 2016-02-09T16:11:32Z 2014-11 2014-06 Article http://purl.org/eprint/type/JournalArticle 15320464 1532-0480 http://hdl.handle.net/1721.1/101133 Singh, Anima, Girish Nadkarni, Omri Gottesman, Stephen B. Ellis, Erwin P. Bottinger, and John V. Guttag. “Incorporating Temporal EHR Data in Predictive Models for Risk Stratification of Renal Function Deterioration.” Journal of Biomedical Informatics 53 (February 2015): 220–228. https://orcid.org/0000-0003-0992-0906 en_US http://dx.doi.org/10.1016/j.jbi.2014.11.005 Journal of Biomedical Informatics Creative Commons Attribution-NonCommercial-NoDerivs License http://creativecommons.org/licenses/by-nc-nd/4.0/ application/pdf Elsevier PMC |
| spellingShingle | Singh, Anima Nadkarni, Girish Gottesman, Omri Ellis, Stephen B. Bottinger, Erwin P. Guttag, John V. Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title | Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title_full | Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title_fullStr | Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title_full_unstemmed | Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title_short | Incorporating temporal EHR data in predictive models for risk stratification of renal function deterioration |
| title_sort | incorporating temporal ehr data in predictive models for risk stratification of renal function deterioration |
| url | http://hdl.handle.net/1721.1/101133 https://orcid.org/0000-0003-0992-0906 |
| work_keys_str_mv | AT singhanima incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration AT nadkarnigirish incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration AT gottesmanomri incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration AT ellisstephenb incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration AT bottingererwinp incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration AT guttagjohnv incorporatingtemporalehrdatainpredictivemodelsforriskstratificationofrenalfunctiondeterioration |