Abstract
Background: Acute Kidney Injury (AKI) occurs in at least 5 % of hospitalized patients and can result in 40-70 % morbidity and mortality. Even following recovery, many subjects may experience progressive deterioration of renal function. The heterogeneous etiology and pathophysiology of AKI complicates its diagnosis and medical management and can add to poor patient outcomes and incur substantial hospital costs. AKI is predictable and may be avoidable if early risk factors are identified and utilized in the clinical setting. Timely detection of undiagnosed AKI in hospitalized patients can also lead to better disease management. Methods: Data from 25,521 hospital stays in one calendar year of patients 60 years and older was collected from a large health care system. Four machine learning models (logistic regression, support vector machines, decision trees and native Bayes) along with their ensemble were tested for AKI prediction and detection tasks. Patient demographics, laboratory tests, medications and comorbid conditions were used as the predictor variables. The models were compared using the area under ROC curve (AUC) evaluation metric. Results: Logistic regression performed the best for AKI detection (AUC 0.743) and was a close second to the ensemble for AKI prediction (AUC ensemble: 0.664, AUC logistic regression: 0.660). History of prior AKI, use of combination drugs such as ACE inhibitors, NSAIDS and diuretics, and presence of comorbid conditions such as respiratory failure were found significant for both AKI detection and risk prediction. Conclusions: The machine learning models performed fairly well on both predicting AKI and detecting undiagnosed AKI. To the best of our knowledge, this is the first study examining the difference between prediction and detection of AKI. The distinction has clinical relevance, and can help providers either identify at risk subjects and implement preventative strategies or manage their treatment depending on whether AKI is predicted or detected.
| Original language | English (US) |
|---|---|
| Article number | 39 |
| Journal | BMC Medical Informatics and Decision Making |
| Volume | 16 |
| Issue number | 1 |
| DOIs | |
| State | Published - 2016 |
Fingerprint
Keywords
- Acute kidney injury (AKI)
- Detection
- Elderly
- Machine learning
- Modeling
- Prediction
ASJC Scopus subject areas
- Health Policy
- Health Informatics
Cite this
Prediction and detection models for acute kidney injury in hospitalized older adults. / Kate, Rohit J.; Perez, Ruth M.; Mazumdar, Debesh; Pasupathy, Kalyan S; Nilakantan, Vani.
In: BMC Medical Informatics and Decision Making, Vol. 16, No. 1, 39, 2016.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Prediction and detection models for acute kidney injury in hospitalized older adults
AU - Kate, Rohit J.
AU - Perez, Ruth M.
AU - Mazumdar, Debesh
AU - Pasupathy, Kalyan S
AU - Nilakantan, Vani
PY - 2016
Y1 - 2016
N2 - Background: Acute Kidney Injury (AKI) occurs in at least 5 % of hospitalized patients and can result in 40-70 % morbidity and mortality. Even following recovery, many subjects may experience progressive deterioration of renal function. The heterogeneous etiology and pathophysiology of AKI complicates its diagnosis and medical management and can add to poor patient outcomes and incur substantial hospital costs. AKI is predictable and may be avoidable if early risk factors are identified and utilized in the clinical setting. Timely detection of undiagnosed AKI in hospitalized patients can also lead to better disease management. Methods: Data from 25,521 hospital stays in one calendar year of patients 60 years and older was collected from a large health care system. Four machine learning models (logistic regression, support vector machines, decision trees and native Bayes) along with their ensemble were tested for AKI prediction and detection tasks. Patient demographics, laboratory tests, medications and comorbid conditions were used as the predictor variables. The models were compared using the area under ROC curve (AUC) evaluation metric. Results: Logistic regression performed the best for AKI detection (AUC 0.743) and was a close second to the ensemble for AKI prediction (AUC ensemble: 0.664, AUC logistic regression: 0.660). History of prior AKI, use of combination drugs such as ACE inhibitors, NSAIDS and diuretics, and presence of comorbid conditions such as respiratory failure were found significant for both AKI detection and risk prediction. Conclusions: The machine learning models performed fairly well on both predicting AKI and detecting undiagnosed AKI. To the best of our knowledge, this is the first study examining the difference between prediction and detection of AKI. The distinction has clinical relevance, and can help providers either identify at risk subjects and implement preventative strategies or manage their treatment depending on whether AKI is predicted or detected.
AB - Background: Acute Kidney Injury (AKI) occurs in at least 5 % of hospitalized patients and can result in 40-70 % morbidity and mortality. Even following recovery, many subjects may experience progressive deterioration of renal function. The heterogeneous etiology and pathophysiology of AKI complicates its diagnosis and medical management and can add to poor patient outcomes and incur substantial hospital costs. AKI is predictable and may be avoidable if early risk factors are identified and utilized in the clinical setting. Timely detection of undiagnosed AKI in hospitalized patients can also lead to better disease management. Methods: Data from 25,521 hospital stays in one calendar year of patients 60 years and older was collected from a large health care system. Four machine learning models (logistic regression, support vector machines, decision trees and native Bayes) along with their ensemble were tested for AKI prediction and detection tasks. Patient demographics, laboratory tests, medications and comorbid conditions were used as the predictor variables. The models were compared using the area under ROC curve (AUC) evaluation metric. Results: Logistic regression performed the best for AKI detection (AUC 0.743) and was a close second to the ensemble for AKI prediction (AUC ensemble: 0.664, AUC logistic regression: 0.660). History of prior AKI, use of combination drugs such as ACE inhibitors, NSAIDS and diuretics, and presence of comorbid conditions such as respiratory failure were found significant for both AKI detection and risk prediction. Conclusions: The machine learning models performed fairly well on both predicting AKI and detecting undiagnosed AKI. To the best of our knowledge, this is the first study examining the difference between prediction and detection of AKI. The distinction has clinical relevance, and can help providers either identify at risk subjects and implement preventative strategies or manage their treatment depending on whether AKI is predicted or detected.
KW - Acute kidney injury (AKI)
KW - Detection
KW - Elderly
KW - Machine learning
KW - Modeling
KW - Prediction
UR - http://www.scopus.com/inward/record.url?scp=85007609237&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85007609237&partnerID=8YFLogxK
U2 - 10.1186/s12911-016-0277-4
DO - 10.1186/s12911-016-0277-4
M3 - Article
C2 - 27025458
AN - SCOPUS:85007609237
VL - 16
JO - BMC Medical Informatics and Decision Making
JF - BMC Medical Informatics and Decision Making
SN - 1472-6947
IS - 1
M1 - 39
ER -