Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death.
<h4>Background</h4>The prediction of readmission or death after a hospital discharge for heart failure (HF) remains a major challenge. Modern healthcare systems, electronic health records, and machine learning (ML) techniques allow us to mine data to select the most significant variables...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2019-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0218760 |
| _version_ | 1819140231293566976 |
|---|---|
| author | Saqib E Awan Mohammed Bennamoun Ferdous Sohel Frank M Sanfilippo Benjamin J Chow Girish Dwivedi |
| author_facet | Saqib E Awan Mohammed Bennamoun Ferdous Sohel Frank M Sanfilippo Benjamin J Chow Girish Dwivedi |
| author_sort | Saqib E Awan |
| collection | DOAJ |
| description | <h4>Background</h4>The prediction of readmission or death after a hospital discharge for heart failure (HF) remains a major challenge. Modern healthcare systems, electronic health records, and machine learning (ML) techniques allow us to mine data to select the most significant variables (allowing for reduction in the number of variables) without compromising the performance of models used for prediction of readmission and death. Moreover, ML methods based on transformation of variables may potentially further improve the performance.<h4>Objective</h4>To use ML techniques to determine the most relevant and also transform variables for the prediction of 30-day readmission or death in HF patients.<h4>Methods</h4>We identified all Western Australian patients aged 65 years and above admitted for HF between 2003-2008 in linked administrative data. We evaluated variables associated with HF readmission or death using standard statistical and ML based selection techniques. We also tested the new variables produced by transformation of the original variables. We developed multi-layer perceptron prediction models and compared their predictive performance using metrics such as Area Under the receiver operating characteristic Curve (AUC), sensitivity and specificity.<h4>Results</h4>Following hospital discharge, the proportion of 30-day readmissions or death was 23.7% in our cohort of 10,757 HF patients. The prediction model developed by us using a smaller set of variables (n = 8) had comparable performance (AUC 0.62) to the traditional model (n = 47, AUC 0.62). Transformation of the original 47 variables further improved (p<0.001) the performance of the predictive model (AUC 0.66).<h4>Conclusions</h4>A small set of variables selected using ML matched the performance of the model that used the full set of 47 variables for predicting 30-day readmission or death in HF patients. Model performance can be further significantly improved by transforming the original variables using ML methods. |
| first_indexed | 2024-12-22T11:35:16Z |
| format | Article |
| id | doaj.art-352887a76bcb493ba44ef919b9c3d2e2 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-12-22T11:35:16Z |
| publishDate | 2019-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-352887a76bcb493ba44ef919b9c3d2e22022-12-21T18:27:27ZengPublic Library of Science (PLoS)PLoS ONE1932-62032019-01-01146e021876010.1371/journal.pone.0218760Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death.Saqib E AwanMohammed BennamounFerdous SohelFrank M SanfilippoBenjamin J ChowGirish Dwivedi<h4>Background</h4>The prediction of readmission or death after a hospital discharge for heart failure (HF) remains a major challenge. Modern healthcare systems, electronic health records, and machine learning (ML) techniques allow us to mine data to select the most significant variables (allowing for reduction in the number of variables) without compromising the performance of models used for prediction of readmission and death. Moreover, ML methods based on transformation of variables may potentially further improve the performance.<h4>Objective</h4>To use ML techniques to determine the most relevant and also transform variables for the prediction of 30-day readmission or death in HF patients.<h4>Methods</h4>We identified all Western Australian patients aged 65 years and above admitted for HF between 2003-2008 in linked administrative data. We evaluated variables associated with HF readmission or death using standard statistical and ML based selection techniques. We also tested the new variables produced by transformation of the original variables. We developed multi-layer perceptron prediction models and compared their predictive performance using metrics such as Area Under the receiver operating characteristic Curve (AUC), sensitivity and specificity.<h4>Results</h4>Following hospital discharge, the proportion of 30-day readmissions or death was 23.7% in our cohort of 10,757 HF patients. The prediction model developed by us using a smaller set of variables (n = 8) had comparable performance (AUC 0.62) to the traditional model (n = 47, AUC 0.62). Transformation of the original 47 variables further improved (p<0.001) the performance of the predictive model (AUC 0.66).<h4>Conclusions</h4>A small set of variables selected using ML matched the performance of the model that used the full set of 47 variables for predicting 30-day readmission or death in HF patients. Model performance can be further significantly improved by transforming the original variables using ML methods.https://doi.org/10.1371/journal.pone.0218760 |
| spellingShingle | Saqib E Awan Mohammed Bennamoun Ferdous Sohel Frank M Sanfilippo Benjamin J Chow Girish Dwivedi Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. PLoS ONE |
| title | Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. |
| title_full | Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. |
| title_fullStr | Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. |
| title_full_unstemmed | Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. |
| title_short | Feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death. |
| title_sort | feature selection and transformation by machine learning reduce variable numbers and improve prediction for heart failure readmission or death |
| url | https://doi.org/10.1371/journal.pone.0218760 |
| work_keys_str_mv | AT saqibeawan featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath AT mohammedbennamoun featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath AT ferdoussohel featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath AT frankmsanfilippo featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath AT benjaminjchow featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath AT girishdwivedi featureselectionandtransformationbymachinelearningreducevariablenumbersandimprovepredictionforheartfailurereadmissionordeath |