scholarly journals Detection of Drug–Drug Interactions Inducing Acute Kidney Injury by Electronic Health Records Mining

Drug Safety ◽  
2015 ◽  
Vol 38 (9) ◽  
pp. 799-809 ◽  
Author(s):  
Yannick Girardeau ◽  
Claire Trivin ◽  
Pierre Durieux ◽  
Christine Le Beller ◽  
Lillo-Le Louet Agnes ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Drug Interactions ◽  
Kidney Injury ◽  
Health Records ◽  
Electronic Health

scholarly journals Utilizing Electronic Health Records to Predict Acute Kidney Injury Risk and Outcomes: Workgroup Statements from the 15thADQI Consensus Conference

2016 ◽  
Vol 3 ◽  
pp. 99 ◽  
Author(s):  
Scott M. Sutherland ◽  
Lakhmir S. Chawla ◽  
Sandra L. Kane-Gill ◽  
Raymond K. Hsu ◽  
Andrew A. Kramer ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Injury Risk ◽  
Kidney Injury ◽  
Consensus Conference ◽  
Health Records ◽  
Electronic Health

scholarly journals Leveraging Big Data and Electronic Health Records to Enhance Novel Approaches to Acute Kidney Injury Research and Care

2017 ◽  
Vol 44 (1) ◽  
pp. 68-76 ◽  
Author(s):  
Scott M. Sutherland ◽  
Stuart L. Goldstein ◽  
Sean M. Bagshaw
Keyword(s):  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Kidney Injury ◽  
Diagnostic Strategy ◽  
Consensus Definition ◽  
Health Records ◽  
Electronic Health ◽  
Predictive Approaches ◽  
Care Improvement ◽  
Injury Research

While acute kidney injury (AKI) has been poorly defined historically, a decade of effort has culminated in a standardized, consensus definition. In parallel, electronic health records (EHRs) have been adopted with greater regularity, clinical informatics approaches have been refined, and the field of EHR-enabled care improvement and research has burgeoned. Although both fields have matured in isolation, uniting the 2 has the capacity to redefine AKI-related care and research. This article describes how the application of a consistent AKI definition to the EHR dataset can accurately and rapidly diagnose and identify AKI events. Furthermore, this electronic, automated diagnostic strategy creates the opportunity to develop predictive approaches, optimize AKI alerts, and trace AKI events across institutions, care platforms, and administrative datasets.

scholarly journals Health Care Analytics With Time-Invariant and Time-Variant Feature Importance to Predict Hospital-Acquired Acute Kidney Injury: Observational Longitudinal Study (Preprint)

2021 ◽  
Author(s):  
Horng-Ruey Chua ◽  
Kaiping Zheng ◽  
Anantharaman Vathsala ◽  
Kee-Yuan Ngiam ◽  
Hui-Kim Yap ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Lead Time ◽  
Kidney Injury ◽  
False Positives ◽  
Health Records ◽  
Electronic Health ◽  
Hospital Acquired ◽  
Time Invariant

BACKGROUND Acute kidney injury (AKI) develops in 4% of hospitalized patients and is a marker of clinical deterioration and nephrotoxicity. AKI onset is highly variable in hospitals, which makes it difficult to time biomarker assessment in all patients for preemptive care. OBJECTIVE The study sought to apply machine learning techniques to electronic health records and predict hospital-acquired AKI by a 48-hour lead time, with the aim to create an AKI surveillance algorithm that is deployable in real time. METHODS The data were sourced from 20,732 case admissions in 16,288 patients over 1 year in our institution. We enhanced the bidirectional recurrent neural network model with a novel time-invariant and time-variant aggregated module to capture important clinical features temporal to AKI in every patient. Time-series features included laboratory parameters that preceded a 48-hour prediction window before AKI onset; the latter’s corresponding reference was the final in-hospital serum creatinine performed in case admissions without AKI episodes. RESULTS The cohort was of mean age 53 (SD 25) years, of whom 29%, 12%, 12%, and 53% had diabetes, ischemic heart disease, cancers, and baseline eGFR &lt;90 mL/min/1.73 m<sup>2</sup>, respectively. There were 911 AKI episodes in 869 patients. We derived and validated an algorithm in the testing dataset with an AUROC of 0.81 (0.78-0.85) for predicting AKI. At a 15% prediction threshold, our model generated 699 AKI alerts with 2 false positives for every true AKI and predicted 26% of AKIs. A lowered 5% prediction threshold improved the recall to 60% but generated 3746 AKI alerts with 6 false positives for every true AKI. Representative interpretation results produced by our model alluded to the top-ranked features that predicted AKI that could be categorized in association with sepsis, acute coronary syndrome, nephrotoxicity, or multiorgan injury, specific to every case at risk. CONCLUSIONS We generated an accurate algorithm from electronic health records through machine learning that predicted AKI by a lead time of at least 48 hours. The prediction threshold could be adjusted during deployment to optimize recall and minimize alert fatigue, while its precision could potentially be augmented by targeted AKI biomarker assessment in the high-risk cohort identified.

scholarly journals Predicting Acute Kidney Injury: A Machine Learning Approach Using Electronic Health Records

Information ◽  
2020 ◽  
Vol 11 (8) ◽  
pp. 386
Author(s):  
Sheikh S. Abdullah ◽  
Neda Rostamzadeh ◽  
Kamran Sedig ◽  
Amit X. Garg ◽  
Eric McArthur
Keyword(s):  
Machine Learning ◽  
Emergency Department ◽  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Prediction Models ◽  
Kidney Injury ◽  
Machine Learning Techniques ◽  
Health Records ◽  
Mortality And Morbidity ◽  
Electronic Health

Acute kidney injury (AKI) is a common complication in hospitalized patients and can result in increased hospital stay, health-related costs, mortality and morbidity. A number of recent studies have shown that AKI is predictable and avoidable if early risk factors can be identified by analyzing Electronic Health Records (EHRs). In this study, we employ machine learning techniques to identify older patients who have a risk of readmission with AKI to the hospital or emergency department within 90 days after discharge. One million patients’ records are included in this study who visited the hospital or emergency department in Ontario between 2014 and 2016. The predictor variables include patient demographics, comorbid conditions, medications and diagnosis codes. We developed 31 prediction models based on different combinations of two sampling techniques, three ensemble methods, and eight classifiers. These models were evaluated through 10-fold cross-validation and compared based on the AUROC metric. The performances of these models were consistent, and the AUROC ranged between 0.61 and 0.88 for predicting AKI among 31 prediction models. In general, the performances of ensemble-based methods were higher than the cost-sensitive logistic regression. We also validated features that are most relevant in predicting AKI with a healthcare expert to improve the performance and reliability of the models. This study predicts the risk of AKI for a patient after being discharged, which provides healthcare providers enough time to intervene before the onset of AKI.

scholarly journals Causal risk factor discovery for severe acute kidney injury using electronic health records

2018 ◽  
Vol 18 (S1) ◽  
Author(s):  
Weiqi Chen ◽  
Yong Hu ◽  
Xiangzhou Zhang ◽  
Lijuan Wu ◽  
Kang Liu ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Risk Factor ◽  
Electronic Health Records ◽  
Kidney Injury ◽  
Health Records ◽  
Causal Risk Factor ◽  
Electronic Health

scholarly journals Health Care Analytics With Time-Invariant and Time-Variant Feature Importance to Predict Hospital-Acquired Acute Kidney Injury: Observational Longitudinal Study

10.2196/30805 ◽  
2021 ◽  
Vol 23 (12) ◽  
pp. e30805
Author(s):  
Horng-Ruey Chua ◽  
Kaiping Zheng ◽  
Anantharaman Vathsala ◽  
Kee-Yuan Ngiam ◽  
Hui-Kim Yap ◽  
...  
Keyword(s):  
Machine Learning ◽  
Acute Kidney Injury ◽  
Electronic Health Records ◽  
Lead Time ◽  
Kidney Injury ◽  
False Positives ◽  
Health Records ◽  
Electronic Health ◽  
Hospital Acquired ◽  
Time Invariant

Background Acute kidney injury (AKI) develops in 4% of hospitalized patients and is a marker of clinical deterioration and nephrotoxicity. AKI onset is highly variable in hospitals, which makes it difficult to time biomarker assessment in all patients for preemptive care. Objective The study sought to apply machine learning techniques to electronic health records and predict hospital-acquired AKI by a 48-hour lead time, with the aim to create an AKI surveillance algorithm that is deployable in real time. Methods The data were sourced from 20,732 case admissions in 16,288 patients over 1 year in our institution. We enhanced the bidirectional recurrent neural network model with a novel time-invariant and time-variant aggregated module to capture important clinical features temporal to AKI in every patient. Time-series features included laboratory parameters that preceded a 48-hour prediction window before AKI onset; the latter’s corresponding reference was the final in-hospital serum creatinine performed in case admissions without AKI episodes. Results The cohort was of mean age 53 (SD 25) years, of whom 29%, 12%, 12%, and 53% had diabetes, ischemic heart disease, cancers, and baseline eGFR <90 mL/min/1.73 m2, respectively. There were 911 AKI episodes in 869 patients. We derived and validated an algorithm in the testing dataset with an AUROC of 0.81 (0.78-0.85) for predicting AKI. At a 15% prediction threshold, our model generated 699 AKI alerts with 2 false positives for every true AKI and predicted 26% of AKIs. A lowered 5% prediction threshold improved the recall to 60% but generated 3746 AKI alerts with 6 false positives for every true AKI. Representative interpretation results produced by our model alluded to the top-ranked features that predicted AKI that could be categorized in association with sepsis, acute coronary syndrome, nephrotoxicity, or multiorgan injury, specific to every case at risk. Conclusions We generated an accurate algorithm from electronic health records through machine learning that predicted AKI by a lead time of at least 48 hours. The prediction threshold could be adjusted during deployment to optimize recall and minimize alert fatigue, while its precision could potentially be augmented by targeted AKI biomarker assessment in the high-risk cohort identified.

scholarly journals Electronic health records accurately predict renal replacement therapy in acute kidney injury

2019 ◽  
Vol 20 (1) ◽  
Author(s):  
Sanmay Low ◽  
Anantharaman Vathsala ◽  
Tanusya Murali Murali ◽  
Long Pang ◽  
Graeme MacLaren ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Renal Replacement Therapy ◽  
Electronic Health Records ◽  
Replacement Therapy ◽  
Kidney Injury ◽  
Health Records ◽  
Electronic Health

scholarly journals Refining Safe Contrast Limits for Preventing Acute Kidney Injury After Percutaneous Coronary Intervention

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Neal Yuan ◽  
Khalid Latif ◽  
Patrick G. Botting ◽  
Yaron Elad ◽  
Steven M. Bradley ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Percutaneous Coronary Intervention ◽  
Electronic Health Records ◽  
Kidney Injury ◽  
Coronary Intervention ◽  
Electronic Health Record Data ◽  
Data Set ◽  
Health Records ◽  
Percutaneous Coronary ◽  
Electronic Health

Background Contrast‐associated acute kidney injury (CA‐AKI) is associated with substantial morbidity and may be prevented by using less contrast during percutaneous coronary intervention (PCI). However, tools for determining safe contrast volumes are limited. We developed risk models to tailor safe contrast volume limits during PCI. Methods and Results Using data from all PCIs performed at 18 hospitals from January 2015 to March 2018, we developed logistic regression models for predicting CA‐AKI, including simpler models (“pragmatic full,” “pragmatic minimum”) using only predictors easily derivable from electronic health records. We prospectively validated these models using PCI data from April 2018 to December 2018 and compared them to preexisting safe contrast models using the area under the receiver operating characteristic curve (AUC). The model derivation data set included 20 579 PCIs with 2102 CA‐AKI cases. When applying models to the separate validation data set (5423 PCIs, 488 CA‐AKI cases), prior safe contrast limits (5*Weight/Creatinine, 2*CreatinineClearance) were poor measures of safety with accuracies of 53.7% and 56.6% in predicting CA‐AKI, respectively. The full, pragmatic full, and pragmatic minimum models performed significantly better (accuracy, 73.1%, 69.3%, 66.6%; AUC, 0.80, 0.76, 0.72 versus 0.59 for 5 * Weight/Creatinine, 0.61 for 2*CreatinineClearance). We found that applying safe contrast limits could meaningfully reduce CA‐AKI risk in one‐quarter of patients. Conclusions Compared with preexisting equations, new multivariate models for safe contrast limits were substantially more accurate in predicting CA‐AKI and could help determine which patients benefit most from limiting contrast during PCI. Using readily available electronic health record data, these models could be implemented into electronic health records to provide actionable information for improving PCI safety.

Sign in / Sign up

Export Citation Format

Share Document