Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients
Background: We re-analyzed data from the Systolic Blood Pressure Intervention Trial (SPRINT) trial to identify features of systolic blood pressure (SBP) variability that portend poor cardiovascular outcomes using a nonlinear machine-learning algorithm. Methods: We included all patients who completed...
| Main Authors: | , , , , , |
|---|---|
| מחברים אחרים: | |
| פורמט: | Article |
| שפה: | English |
| יצא לאור: |
Oxford University Press
2019
|
| גישה מקוונת: | https://hdl.handle.net/1721.1/122821 |
| _version_ | 1826192789359034368 |
|---|---|
| author | Lacson, Ronilda Baker, Bowen Suresh, Harini Andriole, Katherine Szolovits, Peter Lacson, Eduardo |
| author2 | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science |
| author_facet | Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Lacson, Ronilda Baker, Bowen Suresh, Harini Andriole, Katherine Szolovits, Peter Lacson, Eduardo |
| author_sort | Lacson, Ronilda |
| collection | MIT |
| description | Background: We re-analyzed data from the Systolic Blood Pressure Intervention Trial (SPRINT) trial to identify features of systolic blood pressure (SBP) variability that portend poor cardiovascular outcomes using a nonlinear machine-learning algorithm. Methods: We included all patients who completed 1 year of the study without reaching any primary endpoint during the first year, specifically: myocardial infarction, other acute coronary syndromes, stroke, heart failure or death from a cardiovascular event (n = 8799; 94%). In addition to clinical variables, features representing longitudinal SBP trends and variability were determined and combined in a random forest algorithm, optimized using cross-validation, using 70% of patients in the training set. Area under the curve (AUC) was measured using a 30% testing set. Finally, feature importance was determined by minimizing node impurity averaging over all trees in the forest for a specific feature. Results: A total of 365 patients (4.1%) reached the combined primary outcome over 37 months of follow-up. The random forest classifier had an AUC of 0.71 on the testing set. The 10 most significant features selected in order of importance by the automated algorithm included the urine albumin/creatinine (CR) ratio, estimated glomerular filtration rate, age, serum CR, history of subclinical cardiovascular disease (CVD), cholesterol, a variable representing SBP signals using wavelet transformation, high-density lipoprotein, the 90th percentile of SBP and triglyceride level. Conclusions: We successfully demonstrated use of random forest algorithm to define best prognostic longitudinal SBP representations. In addition to known risk factors for CVD, transformed variables for time series SBP measurements were found to be important in predicting poor cardiovascular outcomes and require further evaluation. Keywords: blood pressure; cardiovascular diseases; heart disease; hypertension; machine learning |
| first_indexed | 2024-09-23T09:28:54Z |
| format | Article |
| id | mit-1721.1/122821 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T09:28:54Z |
| publishDate | 2019 |
| publisher | Oxford University Press |
| record_format | dspace |
| spelling | mit-1721.1/1228212022-09-30T14:43:18Z Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients Lacson, Ronilda Baker, Bowen Suresh, Harini Andriole, Katherine Szolovits, Peter Lacson, Eduardo Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Background: We re-analyzed data from the Systolic Blood Pressure Intervention Trial (SPRINT) trial to identify features of systolic blood pressure (SBP) variability that portend poor cardiovascular outcomes using a nonlinear machine-learning algorithm. Methods: We included all patients who completed 1 year of the study without reaching any primary endpoint during the first year, specifically: myocardial infarction, other acute coronary syndromes, stroke, heart failure or death from a cardiovascular event (n = 8799; 94%). In addition to clinical variables, features representing longitudinal SBP trends and variability were determined and combined in a random forest algorithm, optimized using cross-validation, using 70% of patients in the training set. Area under the curve (AUC) was measured using a 30% testing set. Finally, feature importance was determined by minimizing node impurity averaging over all trees in the forest for a specific feature. Results: A total of 365 patients (4.1%) reached the combined primary outcome over 37 months of follow-up. The random forest classifier had an AUC of 0.71 on the testing set. The 10 most significant features selected in order of importance by the automated algorithm included the urine albumin/creatinine (CR) ratio, estimated glomerular filtration rate, age, serum CR, history of subclinical cardiovascular disease (CVD), cholesterol, a variable representing SBP signals using wavelet transformation, high-density lipoprotein, the 90th percentile of SBP and triglyceride level. Conclusions: We successfully demonstrated use of random forest algorithm to define best prognostic longitudinal SBP representations. In addition to known risk factors for CVD, transformed variables for time series SBP measurements were found to be important in predicting poor cardiovascular outcomes and require further evaluation. Keywords: blood pressure; cardiovascular diseases; heart disease; hypertension; machine learning 2019-11-11T20:04:04Z 2019-11-11T20:04:04Z 2018-07-03 2018-03 2019-07-11T12:26:02Z Article http://purl.org/eprint/type/JournalArticle 2048-8513 https://hdl.handle.net/1721.1/122821 Lacson, Ronilda C. et al. "Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients." Clinical Kidney Journal, 12, 2 (April 2019): 206–212 © 2018 the Authors en https://doi.org/10.1093/ckj/sfy049 Clinical Kidney Journal Creative Commons Attribution NonCommercial License 4.0 https://creativecommons.org/licenses/by-nc/4.0/ application/pdf Oxford University Press Oxford University Press |
| spellingShingle | Lacson, Ronilda Baker, Bowen Suresh, Harini Andriole, Katherine Szolovits, Peter Lacson, Eduardo Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title | Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title_full | Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title_fullStr | Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title_full_unstemmed | Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title_short | Use of machine-learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| title_sort | use of machine learning algorithms to determine features of systolic blood pressure variability that predict poor outcomes in hypertensive patients |
| url | https://hdl.handle.net/1721.1/122821 |
| work_keys_str_mv | AT lacsonronilda useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients AT bakerbowen useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients AT sureshharini useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients AT andriolekatherine useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients AT szolovitspeter useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients AT lacsoneduardo useofmachinelearningalgorithmstodeterminefeaturesofsystolicbloodpressurevariabilitythatpredictpooroutcomesinhypertensivepatients |