Machine learning model for predicting acute kidney injury progression in critically ill patients

Artificial intelligence and machine learning's role in sepsis-associated acute kidney injury

Sepsis-associated acute kidney injury (SA-AKI) is a serious complication in critically ill patients, resulting in higher mortality, morbidity, and cost. The intricate pathophysiology of SA-AKI requires vigilant clinical monitoring and appropriate, prompt intervention. While traditional statistical analyses have identified severe risk factors for SA-AKI, the results have been inconsistent across studies. This has led to growing interest in leveraging artificial intelligence (AI) and machine learning (ML) to predict SA-AKI better. ML can uncover complex patterns beyond human discernment by analyzing vast datasets. Supervised learning models like XGBoost and RNN-LSTM have proven remarkably accurate at predicting SA-AKI onset and subsequent mortality, often surpassing traditional risk scores. Meanwhile, unsupervised learning reveals clinically relevant sub-phenotypes among diverse SA-AKI patients, enabling more tailored care. In addition, it potentially optimizes sepsis treatment to prevent SA-AKI through continual refinement based on patient outcomes. However, utilizing AI/ML presents ethical and practical challenges regarding data privacy, algorithmic biases, and regulatory compliance. AI/ML allows early risk detection, personalized management, optimal treatment strategies, and collaborative learning for SA-AKI management. Future directions include real-time patient monitoring, simulated data generation, and predictive algorithms for timely interventions. However, a smooth transition to clinical practice demands continuous model enhancements and rigorous regulatory oversight. In this article, we outlined the conventional methods used to address SA-AKI and explore how AI and ML can be applied to diagnose and manage SA-AKI, highlighting their potential to revolutionize SA-AKI care.

Predictive approach for liberation from acute dialysis in ICU patients using interpretable machine learning

Renal recovery following dialysis-requiring acute kidney injury (AKI-D) is a vital clinical outcome in critical care, yet it remains an understudied area. This retrospective cohort study, conducted in a medical center in Taiwan from 2015 to 2020, enrolled patients with AKI-D during intensive care unit stays. We aimed to develop and temporally test models for predicting dialysis liberation before hospital discharge using machine learning algorithms and explore early predictors. The dataset comprised 90 routinely collected variables within the first three days of dialysis initiation. Out of 1,381 patients who received acute dialysis, 27.3% experienced renal recovery. The cohort was divided into the training group (N = 1135) and temporal testing group (N = 251). The models demonstrated good performance, with an area under the receiver operating characteristic curve of 0.85 (95% CI, 0.81-0.88) and an area under the precision-recall curve of 0.69 (95% CI, 0.62-0.76) for the XGBoost model. Key predictors included urine volume, Charlson comorbidity index, vital sign derivatives (trend of respiratory rate and SpO2), and lactate levels. We successfully developed early prediction models for renal recovery by integrating early changes in vital signs and inputs/outputs, which have the potential to aid clinical decision-making in the ICU.

Selective Partitioned Regression for Accurate Kidney Health Monitoring

The number of people diagnosed with advanced stages of kidney disease have been rising every year. Early detection and constant monitoring are the only minimally invasive means to prevent severe kidney damage or kidney failure. We propose a cost-effective machine learning-based testing system that can facilitate inexpensive yet accurate kidney health checks. Our proposed framework, which was developed into an iPhone application, uses a camera-based bio-sensor and state-of-the-art classical machine learning and deep learning techniques for predicting the concentration of creatinine in the sample, based on colorimetric change in the test strip. The predicted creatinine concentration is then used to classify the severity of the kidney disease as healthy, intermediate, or critical. In this article, we focus on the effectiveness of machine learning models to translate the colorimetric reaction to kidney health prediction. In this setting, we thoroughly evaluated the effectiveness of our novel proposed models against state-of-the-art classical machine learning and deep learning approaches. Additionally, we executed a number of ablation studies to measure the performance of our model when trained using different meta-parameter choices. Our evaluation results indicate that our selective partitioned regression (SPR) model, using histogram of colors-based features and a histogram gradient boosted trees underlying estimator, exhibits much better overall prediction performance compared to state-of-the-art methods. Our initial study indicates that SPR can be an effective tool for detecting the severity of kidney disease using inexpensive lateral flow assay test strips and a smart phone-based application. Additional work is needed to verify the performance of the model in various settings.

A machine learning radiomics model based on bpMRI to predict bone metastasis in newly diagnosed prostate cancer patients

OBJECTIVES

To develop and evaluate a machine learning radiomics model based on biparametric magnetic resonance imaging MRI (bpMRI) to predict bone metastasis (BM) status in newly diagnosed prostate cancer (PCa) patients.

METHODS

We retrospectively analyzed bpMRI scans of PCa patients from multiple centers between January 2016 and October 2021. 348 PCa patients were recruited from two institutions for this study. The first institution contributed 284 patients, stratified and randomly divided into training and internal validation cohorts at a 7:3 ratio. The remaining 64 patients were sourced from the second institution and comprised the external validation cohort. Radiomics features were extracted from axial T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) tumor regions. We developed the radiomics prediction model for BM in the training cohort and validated it in the internal and external validation cohorts. As a benchmark, we trained the logistic regression model with lasso feature reduction (LFR-LRM) in the training cohort and further compared it with Naive Bayes, eXtreme Gradient Boosting (XGboost), Random Forest (RF), GBDT, SVM, Adaboost, and KNN algorithms and validated in both the internal and external cohorts. The performance of several predictive models was assessed by receiver operating characteristic (ROC).

RESULTS

The LFR-LRM model achieved an area under the receiver operating characteristic curve (AUC) of 0.89 (95% CI: 0.822-0.974) and an accuracy of 0.828 (95% CI: 0.713-0.911). The AUC and accuracy in external validation were 0.866 (95% CI: 0.784-0.948) and 0.769 (95% CI: 0.648-0.864), respectively. The RF and XGBoost models outperformed the LFR-LRM, with AUCs of 0.907 (95% CI: 0.863-0.949) and 0.928 (95% CI: 0.882-0.974) and accuracies of 0.831 (95% CI: 0.727-0.907) and 0.884 (95% CI: 0.792-0.946). External validation for these models yielded AUCs and accuracies of 0.911 (95% CI: 0.861-0.966), 0.921 (95% CI: 0.889-0.953), and 0.846 (95% CI: 0.735-0.923) and 0.876 (95% CI: 0.771-0.945), respectively.

CONCLUSIONS

The XGboost machine learning model is more accurate than LFR-LRM for predicting BM in patients with newly confirmed PCa.

Machine-learning model for predicting oliguria in critically ill patients

This retrospective cohort study aimed to develop and evaluate a machine-learning algorithm for predicting oliguria, a sign of acute kidney injury (AKI). To this end, electronic health record data from consecutive patients admitted to the intensive care unit (ICU) between 2010 and 2019 were used and oliguria was defined as a urine output of less than 0.5 mL/kg/h. Furthermore, a light-gradient boosting machine was used for model development. Among the 9,241 patients who participated in the study, the proportions of patients with urine output < 0.5 mL/kg/h for 6 h and with AKI during the ICU stay were 27.4% and 30.2%, respectively. The area under the curve (AUC) values provided by the prediction algorithm for the onset of oliguria at 6 h and 72 h using 28 clinically relevant variables were 0.964 (a 95% confidence interval (CI) of 0.963-0.965) and 0.916 (a 95% CI of 0.914-0.918), respectively. The Shapley additive explanation analysis for predicting oliguria at 6 h identified urine values, severity scores, serum creatinine, oxygen partial pressure, fibrinogen/fibrin degradation products, interleukin-6, and peripheral temperature as important variables. Thus, this study demonstrates that a machine-learning algorithm can accurately predict oliguria onset in ICU patients, suggesting the importance of oliguria in the early diagnosis and optimal management of AKI.

A Machine Learning Algorithm Predicting Acute Kidney Injury in Intensive Care Unit Patients (NAVOY Acute Kidney Injury): Proof-of-Concept Study

BACKGROUND

Acute kidney injury (AKI) represents a significant global health challenge, leading to increased patient distress and financial health care burdens. The development of AKI in intensive care unit (ICU) settings is linked to prolonged ICU stays, a heightened risk of long-term renal dysfunction, and elevated short- and long-term mortality rates. The current diagnostic approach for AKI is based on late indicators, such as elevated serum creatinine and decreased urine output, which can only detect AKI after renal injury has transpired. There are no treatments to reverse or restore renal function once AKI has developed, other than supportive care. Early prediction of AKI enables proactive management and may improve patient outcomes.

OBJECTIVE

The primary aim was to develop a machine learning algorithm, NAVOY Acute Kidney Injury, capable of predicting the onset of AKI in ICU patients using data routinely collected in ICU electronic health records. The ultimate goal was to create a clinical decision support tool that empowers ICU clinicians to proactively manage AKI and, consequently, enhance patient outcomes.

METHODS

We developed the NAVOY Acute Kidney Injury algorithm using a hybrid ensemble model, which combines the strengths of both a Random Forest (Leo Breiman and Adele Cutler) and an XGBoost model (Tianqi Chen). To ensure the accuracy of predictions, the algorithm used 22 clinical variables for hourly predictions of AKI as defined by the Kidney Disease: Improving Global Outcomes guidelines. Data for algorithm development were sourced from the Massachusetts Institute of Technology Lab for Computational Physiology Medical Information Mart for Intensive Care IV clinical database, focusing on ICU patients aged 18 years or older.

RESULTS

The developed algorithm, NAVOY Acute Kidney Injury, uses 4 hours of input and can, with high accuracy, predict patients with a high risk of developing AKI 12 hours before onset. The prediction performance compares well with previously published prediction algorithms designed to predict AKI onset in accordance with Kidney Disease: Improving Global Outcomes diagnosis criteria, with an impressive area under the receiver operating characteristics curve (AUROC) of 0.91 and an area under the precision-recall curve (AUPRC) of 0.75. The algorithm's predictive performance was externally validated on an independent hold-out test data set, confirming its ability to predict AKI with exceptional accuracy.

CONCLUSIONS

NAVOY Acute Kidney Injury is an important development in the field of critical care medicine. It offers the ability to predict the onset of AKI with high accuracy using only 4 hours of data routinely collected in ICU electronic health records. This early detection capability has the potential to strengthen patient monitoring and management, ultimately leading to improved patient outcomes. Furthermore, NAVOY Acute Kidney Injury has been granted Conformite Europeenne (CE)-marking, marking a significant milestone as the first CE-marked AKI prediction algorithm for commercial use in European ICUs.

Machine-learning prediction models for any blood component transfusion in hospitalized dengue patients

BACKGROUND

Blood component transfusions are a common and often necessary medical practice during the epidemics of dengue. Transfusions are required for patients when they developed severe dengue fever or thrombocytopenia of 10×109/L or less. This study therefore investigated the risk factors, performance and effectiveness of eight different machine-learning algorithms to predict blood component transfusion requirements in confirmed dengue cases admitted to hospital. The objective was to study the risk factors that can help to predict blood component transfusion needs.

METHODS

Eight predictive models were developed based on retrospective data from a private group of hospitals in India. A python package SHAP (SHapley Additive exPlanations) was used to explain the output of the "XGBoost" model.

RESULTS

Sixteen vital variables were finally selected as having the most significant effects on blood component transfusion prediction. The XGBoost model presented significantly better predictive performance (area under the curve: 0.793; 95 % confidence interval: 0.699-0.795) than the other models.

CONCLUSION

Predictive modelling techniques can be utilized to streamline blood component preparation procedures and can help in the triage of high-risk patients and readiness of caregivers to provide blood component transfusions when required. This study demonstrates the potential of multilayer algorithms to reasonably predict any blood component transfusion needs which may help healthcare providers make more informed decisions regarding patient care.

Machine Learning Predicts Acute Kidney Injury in Hospitalized Patients with Sickle Cell Disease

INTRODUCTION

Acute kidney injury (AKI) is common among hospitalized patients with sickle cell disease (SCD) and contributes to increased morbidity and mortality. Early identification and management of AKI is essential to preventing poor outcomes. We aimed to predict AKI earlier in patients with SCD using a machine-learning model that utilized continuous minute-by-minute physiological data.

METHODS

A total of6,278 adult SCD patient encounters were admitted to inpatient units across five regional hospitals in Memphis, TN, over 3 years, from July 2017 to December 2020. From these, 1,178 patients were selected after filtering for data availability. AKI was identified in 82 (7%) patient encounters, using the 2012 Kidney Disease Improving Global Outcomes (KDIGO) criteria. The remaining 1,096 encounters served as controls. Features derived from five physiological data streams, heart rate, respiratory rate, and blood pressure (systolic, diastolic, and mean), captured every minute from bedside monitors were used. An XGBoost classifier was used for classification.

RESULTS

Our model accurately predicted AKI up to 12 h before onset with an area under the receiver operator curve (AUROC) of 0.91 (95% CI [0.89-0.93]) and up to 48 h before AKI with an AUROC of 0.82 (95% CI [0.80-0.83]). Patients with AKI were more likely to be female (64.6%) and have history of hypertension, pulmonary hypertension, chronic kidney disease, and pneumonia than the control group.

CONCLUSION

XGBoost accurately predicted AKI as early as 12 h before onset in hospitalized SCD patients and may enable the development of innovative prevention strategies.

Machine learning clinical prediction models for acute kidney injury: the impact of baseline creatinine on prediction efficacy

BACKGROUND

There are many Machine Learning (ML) models which predict acute kidney injury (AKI) for hospitalised patients. While a primary goal of these models is to support clinical decision-making, the adoption of inconsistent methods of estimating baseline serum creatinine (sCr) may result in a poor understanding of these models' effectiveness in clinical practice. Until now, the performance of such models with different baselines has not been compared on a single dataset. Additionally, AKI prediction models are known to have a high rate of false positive (FP) events regardless of baseline methods. This warrants further exploration of FP events to provide insight into potential underlying reasons.

OBJECTIVE

The first aim of this study was to assess the variance in performance of ML models using three methods of baseline sCr on a retrospective dataset. The second aim was to conduct an error analysis to gain insight into the underlying factors contributing to FP events.

MATERIALS AND METHODS

The Intensive Care Unit (ICU) patients of the Medical Information Mart for Intensive Care (MIMIC)-IV dataset was used with the KDIGO (Kidney Disease Improving Global Outcome) definition to identify AKI episodes. Three different methods of estimating baseline sCr were defined as (1) the minimum sCr, (2) the Modification of Diet in Renal Disease (MDRD) equation and the minimum sCr and (3) the MDRD equation and the mean of preadmission sCr. For the first aim of this study, a suite of ML models was developed for each baseline and the performance of the models was assessed. An analysis of variance was performed to assess the significant difference between eXtreme Gradient Boosting (XGB) models across all baselines. To address the second aim, Explainable AI (XAI) methods were used to analyse the XGB errors with Baseline 3.

RESULTS

Regarding the first aim, we observed variances in discriminative metrics and calibration errors of ML models when different baseline methods were adopted. Using Baseline 1 resulted in a 14% reduction in the f1 score for both Baseline 2 and Baseline 3. There was no significant difference observed in the results between Baseline 2 and Baseline 3. For the second aim, the FP cohort was analysed using the XAI methods which led to relabelling data with the mean of sCr in 180 to 0 days pre-ICU as the preferred sCr baseline method. The XGB model using this relabelled data achieved an AUC of 0.85, recall of 0.63, precision of 0.54 and f1 score of 0.58. The cohort size was 31,586 admissions, of which 5,473 (17.32%) had AKI.

CONCLUSION

In the absence of a widely accepted method of baseline sCr, AKI prediction studies need to consider the impact of different baseline methods on the effectiveness of ML models and their potential implications in real-world implementations. The utilisation of XAI methods can be effective in providing insight into the occurrence of prediction errors. This can potentially augment the success rate of ML implementation in routine care.

Comparing machine learning and logistic regression for acute kidney injury prediction in trauma patients: A retrospective observational study at a single tertiary medical center

Acute kidney injury (AKI) is common in patients with trauma and is associated with poor outcomes. Therefore, early prediction of AKI in patients with trauma is important for risk stratification and the provision of optimal intensive care unit treatment. This study aimed to compare 2 models, machine learning (ML) techniques and logistic regression, in predicting AKI in patients with trauma. We retrospectively reviewed the charts of 400 patients who sustained torso injuries between January 2016 and June 2020. Patients were included if they were aged > 15 years, admitted to the intensive care unit, survived for > 48 hours, had thoracic and/or abdominal injuries, had no end-stage renal disease, and had no missing data. AKI was defined in accordance with the Kidney Disease Improving Global Outcomes definition and staging system. The patients were divided into 2 groups: AKI (n = 78) and non-AKI (n = 322). We divided the original dataset into a training (80%) and a test set (20%), and the logistic regression with stepwise selection and ML (decision tree with hyperparameter optimization using grid search and cross-validation) was used to build a model for predicting AKI. The models established using the training dataset were evaluated using a confusion matrix receiver operating characteristic curve with the test dataset. We included 400 patients with torso injury, of whom 78 (19.5%) progressed to AKI. Age, intestinal injury, cumulative fluid balance within 24 hours, and the use of vasopressors were independent risk factors for AKI in the logistic regression model. In the ML model, vasopressors were the most important feature, followed by cumulative fluid balance within 24 hours and packed red blood cell transfusion within 4 hours. The accuracy score showed no differences between the 2 groups; however, the recall and F1 score were significantly higher in the ML model (.94 vs 56 and.75 vs 64, respectively). The ML model performed better than the logistic regression model in predicting AKI in patients with trauma. ML techniques can aid in risk stratification and the provision of optimal care.

Predicting outcomes of acute kidney injury in critically ill patients using machine learning

Acute Kidney Injury (AKI) is a sudden episode of kidney failure that is frequently seen in critically ill patients. AKI has been linked to chronic kidney disease (CKD) and mortality. We developed machine learning-based prediction models to predict outcomes following AKI stage 3 events in the intensive care unit. We conducted a prospective observational study that used the medical records of ICU patients diagnosed with AKI stage 3. A random forest algorithm was used to develop two models that can predict patients who will progress to CKD after three and six months of experiencing AKI stage 3. To predict mortality, two survival prediction models have been presented using random survival forests and survival XGBoost. We evaluated established CKD prediction models using AUCROC, and AUPR curves and compared them with the baseline logistic regression models. The mortality prediction models were evaluated with an external test set, and the C-indices were compared to baseline COXPH. We included 101 critically ill patients who experienced AKI stage 3. To increase the training set for the mortality prediction task, an unlabeled dataset has been added. The RF (AUPR: 0.895 and 0.848) and XGBoost (c-index: 0.8248) models have a better performance than the baseline models in predicting CKD and mortality, respectively Machine learning-based models can assist clinicians in making clinical decisions regarding critically ill patients with severe AKI who are likely to develop CKD following discharge. Additionally, we have shown better performance when unlabeled data are incorporated into the survival analysis task.

Prediction of blood pressure variability during thrombectomy using supervised machine learning and outcomes of patients with ischemic stroke from large vessel occlusion

Mechanical thrombectomy (MT) is the standard of care for patients with acute ischemic stroke from large vessel occlusion (AIS-LVO). The association of blood pressure variability (BPV) during MT and outcomes are unknown. We leveraged a supervised machine learning algorithm to predict patient characteristics that are associated with BPV indices. We performed a retrospective review of our comprehensive stroke center's registry of all adult patients undergoing MT between 01/01/2016 and 12/31/2019. The primary outcome was poor functional independence, defined as 90-day modified Rankin Scale (mRS) ≥ 3. We used probit analysis and multivariate logistic regressions to evaluate the association of patients' clinical factors and outcomes. We applied a machine learning algorithm (random forest, RF) to determine predictive factors for the different BPV indices during MT. Evaluation was performed with root-mean-square error (RMSE) and normalized-RMSE (nRMSE) metrics. We analyzed 375 patients with mean age (± standard deviation [SD]) of 65 (15) years. There were 234 (62%) patients with mRS ≥ 3. Univariate probit analysis demonstrated that BPV during MT was associated with poor functional independence. Multivariable logistic regression showed that age, admission National Institutes of Health Stroke Scale (NIHSS), mechanical ventilation, and thrombolysis in cerebral infarction (TICI) score (OR 0.42, 95% CI 0.17-0.98, P = 0.044) were significantly associated with outcome. RF analysis identified that the interval from last-known-well time-to-groin puncture, age, and mechanical ventilation were among important factors significantly associated with BPV. BPV during MT was associated with functional outcome in univariate probit analysis but not in multivariable regression analysis, however, NIHSS and TICI score were. RF algorithm identified risk factors influencing patients' BPV during MT. While awaiting further studies' results, clinicians should still monitor and avoid high BPV during thrombectomy while triaging AIS-LVO candidates quickly to MT.

Acute Kidney Injury and Renal Replacement Therapy: A Review and Update for the Perioperative Physician

Post-operative acute kidney injury is a devastating complication with significant morbidity and mortality associated with it. The perioperative anesthesiologist is in a unique position to potentially mitigate the risk of postoperative AKI, however, understanding the pathophysiology, risk factors and preventative strategies is paramount. There are also certain clinical scenarios, where renal replacement therapy may be indicated intraoperatively including severe electrolyte abnormalities, metabolic acidosis and massive volume overload. A multidisciplinary approach including the nephrologist, critical care physician, surgeon and anesthesiologist is necessary to determine the optimal management of these critically ill patients.

Machine learning for acute kidney injury: Changing the traditional disease prediction mode

Acute kidney injury (AKI) is a serious clinical comorbidity with clear short-term and long-term prognostic implications for inpatients. The diversity of risk factors for AKI has been recognized in previous studies, and a series of predictive models have been developed using traditional statistical methods in conjunction with its preventability, but they have failed to meet the expectations in limited clinical applications, the rapid spread of electronic health records and artificial intelligence machine learning technology has brought new hope for the construction of AKI prediction models. In this article, we systematically review the definition and classification of machine learning methods, modeling ideas and evaluation methods, and the characteristics and current status of modeling studies. According to the modeling objectives, we subdivided them into critical care medical setting models, all medical environment models, special surgery models, special disease models, and special nephrotoxin exposure models. As the first review article to comprehensively summarize and analyze machine learning prediction models for AKI, we aim to objectively describe the advantages and disadvantages of machine learning approaches to modeling, and help other researchers more quickly and intuitively understand the current status of modeling research, inspire ideas and learn from experience, so as to guide and stimulate more research and more in-depth exploration in the future, which will ultimately provide greater help to improve the overall status of AKI diagnosis and treatment.

Surgical or percutaneous coronary revascularization for heart failure: an in silico model using routinely collected health data to emulate a clinical trial

AIMS

The choice of revascularization with coronary artery bypass grafting (CABG) vs. percutaneous coronary intervention (PCI) in people with ischaemic left ventricular dysfunction is not guided by high-quality evidence.

METHODS AND RESULTS

A trial of CABG vs. PCI in people with heart failure (HF) was modelled in silico using routinely collected healthcare data. The in silico trial cohort was selected by matching the target trial cohort, identified from Hospital Episode Statistics in England, with individual patient data from the Surgical Treatment for Ischemic Heart Failure (STICH) trial. Allocation to CABG vs. complex PCI demonstrated random variation across administrative regions in England and was a valid statistical instrument. The primary outcome was 5-year all-cause mortality or cardiovascular hospitalization. Instrumental variable analysis (IVA) was used for the primary analysis. Results were expressed as average treatment effects (ATEs) with 95 confidence intervals (CIs). The target population included 13 519 HF patients undergoing CABG or complex PCI between April 2009 and March 2015. After matching, the emulated trial cohort included 2046 patients. The unadjusted primary outcome rate was 51.1 in the CABG group and 70.0 in the PCI group. IVA of the emulated cohort showed that CABG was associated with a lower risk of the primary outcome (ATE 16.2, 95 CI 20.6 to 11.8), with comparable estimates in the unmatched target population (ATE 15.5, 95 CI 17.5 to 13.5).

CONCLUSION

In people with HF, in silico modelling suggests that CABG is associated with fewer deaths or cardiovascular hospitalizations at 5 years vs. complex PCI. A pragmatic clinical trial is needed to test this hypothesis and this trial would be feasible.

A Transfer Learning Approach to Correct the Temporal Performance Drift of Clinical Prediction Models: Retrospective Cohort Study

Background

Clinical prediction models suffer from performance drift as the patient population shifts over time. There is a great need for model updating approaches or modeling frameworks that can effectively use the old and new data.

Objective

Based on the paradigm of transfer learning, we aimed to develop a novel modeling framework that transfers old knowledge to the new environment for prediction tasks, and contributes to performance drift correction.

Methods

The proposed predictive modeling framework maintains a logistic regression–based stacking ensemble of 2 gradient boosting machine (GBM) models representing old and new knowledge learned from old and new data, respectively (referred to as transfer learning gradient boosting machine [TransferGBM]). The ensemble learning procedure can dynamically balance the old and new knowledge. Using 2010-2017 electronic health record data on a retrospective cohort of 141,696 patients, we validated TransferGBM for hospital-acquired acute kidney injury prediction.

Results

The baseline models (ie, transported models) that were trained on 2010 and 2011 data showed significant performance drift in the temporal validation with 2012-2017 data. Refitting these models using updated samples resulted in performance gains in nearly all cases. The proposed TransferGBM model succeeded in achieving uniformly better performance than the refitted models.

Conclusions

Under the scenario of population shift, incorporating new knowledge while preserving old knowledge is essential for maintaining stable performance. Transfer learning combined with stacking ensemble learning can help achieve a balance of old and new knowledge in a flexible and adaptive way, even in the case of insufficient new data.

Performance of resistive index and semi-quantitative power doppler ultrasound score in predicting acute kidney injury: A meta-analysis of prospective studies

This study aimed to assess the predictive value of the renal resistive index (RRI) and power Doppler ultrasound (PDU) on subsequent acute kidney injury (AKI) risk using a meta-analytic approach. We searched eligible studies in PubMed, EmBase, and the Cochrane library from inception until August 2021. The parameters included the sensitivity, specificity, positive and negative likelihood ratios (PLR and NLR), diagnostic odds ratio (DOR), and area under the receiver operating characteristic curves (AUC). Twenty-three prospective studies involving 2,400 patients were selected. The pooled sensitivity and specificity of the RRI and PDU were 0.76 and 0.79, and 0.64 and 0.90, respectively. The pooled PLR and NLR were 3.64 and 0.31, and 6.58 and 0.40 for the RRI and PDU, respectively. The DORs of the RRI and PDU for predicting AKI were 11.76, and 16.32, respectively. The AUCs of the RRI and PDU for predicting AKI were 0.83, and 0.86, respectively. There were no significant differences between the RRI and PDU for predicting AKI in terms of sensitivity, PLR, NLR, DOR, and AUC. The specificity of the RRI was lower than that of the PDU for predicting AKI. This study found that the predictive performance of the RRI and PDU from the Doppler ultrasound for AKI was similar, which need to be further verified based on the direct comparison results.