Next Article in Journal
No-Reflow after PPCI—A Predictor of Short-Term Outcomes in STEMI Patients
Next Article in Special Issue
Biopsy-Controlled Non-Invasive Quantification of Collagen Type VI in Kidney Transplant Recipients: A Post-Hoc Analysis of the MECANO Trial
Previous Article in Journal
Blood Pressure and Body Weight Have Different Effects on Pulse Wave Velocity and Cardiac Mass in Children
Previous Article in Special Issue
The Macrophage Migration Inhibitory Factor (MIF) Promoter Polymorphisms (rs3063368, rs755622) Predict Acute Kidney Injury and Death after Cardiac Surgery
Article

Optimized Identification of Advanced Chronic Kidney Disease and Absence of Kidney Disease by Combining Different Electronic Health Data Resources and by Applying Machine Learning Strategies

1
Department of Clinical Chemistry and Laboratory Diagnostics and Integrated Biobank Jena (IBBJ), Jena University Hospital, 07747 Jena, Germany
2
Jena University Language & Information Engineering (JULIE) Lab, Friedrich Schiller University Jena, 07743 Jena, Germany
3
Data Integration Center, Jena University Hospital, 07743 Jena, Germany
*
Authors to whom correspondence should be addressed.
Christoph Weber and Lena Röschke contributed equally.
Boris Betz and Michael Kiehntopf contributed equally.
J. Clin. Med. 2020, 9(9), 2955; https://doi.org/10.3390/jcm9092955
Received: 25 June 2020 / Revised: 26 August 2020 / Accepted: 28 August 2020 / Published: 12 September 2020
Automated identification of advanced chronic kidney disease (CKD ≥ III) and of no known kidney disease (NKD) can support both clinicians and researchers. We hypothesized that identification of CKD and NKD can be improved, by combining information from different electronic health record (EHR) resources, comprising laboratory values, discharge summaries and ICD-10 billing codes, compared to using each component alone. We included EHRs from 785 elderly multimorbid patients, hospitalized between 2010 and 2015, that were divided into a training and a test (n = 156) dataset. We used both the area under the receiver operating characteristic (AUROC) and under the precision-recall curve (AUCPR) with a 95% confidence interval for evaluation of different classification models. In the test dataset, the combination of EHR components as a simple classifier identified CKD ≥ III (AUROC 0.96[0.93–0.98]) and NKD (AUROC 0.94[0.91–0.97]) better than laboratory values (AUROC CKD 0.85[0.79–0.90], NKD 0.91[0.87–0.94]), discharge summaries (AUROC CKD 0.87[0.82–0.92], NKD 0.84[0.79–0.89]) or ICD-10 billing codes (AUROC CKD 0.85[0.80–0.91], NKD 0.77[0.72–0.83]) alone. Logistic regression and machine learning models improved recognition of CKD ≥ III compared to the simple classifier if only laboratory values were used (AUROC 0.96[0.92–0.99] vs. 0.86[0.81–0.91], p < 0.05) and improved recognition of NKD if information from previous hospital stays was used (AUROC 0.99[0.98–1.00] vs. 0.95[0.92–0.97]], p < 0.05). Depending on the availability of data, correct automated identification of CKD ≥ III and NKD from EHRs can be improved by generating classification models based on the combination of different EHR components. View Full-Text
Keywords: chronic kidney disease (CKD); no known kidney disease (NKD); ICD-10 billing codes; phenotyping; electronic health record (EHR); estimated glomerular filtration rate (eGFR); machine learning (ML); generalized linear model network (GLMnet); random forest (RF); artificial neural network (ANN), clinical natural language processing (clinical NLP); discharge summaries; laboratory values; area under the receiver operating characteristic (AUROC); area under the precision-recall curve (AUCPR) chronic kidney disease (CKD); no known kidney disease (NKD); ICD-10 billing codes; phenotyping; electronic health record (EHR); estimated glomerular filtration rate (eGFR); machine learning (ML); generalized linear model network (GLMnet); random forest (RF); artificial neural network (ANN), clinical natural language processing (clinical NLP); discharge summaries; laboratory values; area under the receiver operating characteristic (AUROC); area under the precision-recall curve (AUCPR)
Show Figures

Figure 1

MDPI and ACS Style

Weber, C.; Röschke, L.; Modersohn, L.; Lohr, C.; Kolditz, T.; Hahn, U.; Ammon, D.; Betz, B.; Kiehntopf, M. Optimized Identification of Advanced Chronic Kidney Disease and Absence of Kidney Disease by Combining Different Electronic Health Data Resources and by Applying Machine Learning Strategies. J. Clin. Med. 2020, 9, 2955. https://doi.org/10.3390/jcm9092955

AMA Style

Weber C, Röschke L, Modersohn L, Lohr C, Kolditz T, Hahn U, Ammon D, Betz B, Kiehntopf M. Optimized Identification of Advanced Chronic Kidney Disease and Absence of Kidney Disease by Combining Different Electronic Health Data Resources and by Applying Machine Learning Strategies. Journal of Clinical Medicine. 2020; 9(9):2955. https://doi.org/10.3390/jcm9092955

Chicago/Turabian Style

Weber, Christoph, Lena Röschke, Luise Modersohn, Christina Lohr, Tobias Kolditz, Udo Hahn, Danny Ammon, Boris Betz, and Michael Kiehntopf. 2020. "Optimized Identification of Advanced Chronic Kidney Disease and Absence of Kidney Disease by Combining Different Electronic Health Data Resources and by Applying Machine Learning Strategies" Journal of Clinical Medicine 9, no. 9: 2955. https://doi.org/10.3390/jcm9092955

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop