Enhancing the prediction of acute kidney injury risk after percutaneous coronary intervention using machine learning techniques: A retrospective cohort study

What characteristics of clinical decision support system implementations lead to adoption for regular use? A scoping review

INTRODUCTION

Digital healthcare innovation has yielded many prototype clinical decision support (CDS) systems, however, few are fully adopted into practice, despite successful research outcomes. We aimed to explore the characteristics of implementations in clinical practice to inform future innovation.

METHODS

Web of Science, Trip Database, PubMed, NHS Digital and the BMA website were searched for examples of CDS systems in May 2022 and updated in June 2023. Papers were included if they reported on a CDS giving pathway advice to a clinician, adopted into regular clinical practice and had sufficient published information for analysis. Examples were excluded if they were only used in a research setting or intended for patients. Articles found in citation searches were assessed alongside a detailed hand search of the grey literature to gather all available information, including commercial information. Examples were excluded if there was insufficient information for analysis. The normalisation process theory (NPT) framework informed analysis.

RESULTS

22 implemented CDS projects were included, with 53 related publications or sources of information (40 peer-reviewed publications and 13 alternative sources). NPT framework analysis indicated organisational support was paramount to successful adoption of CDS. Ensuring that workflows were optimised for patient care alongside iterative, mixed-methods implementation was key to engaging clinicians.

CONCLUSION

Extensive searches revealed few examples of CDS available for analysis, highlighting the implementation gap between research and healthcare innovation. Lessons from included projects include the need for organisational support, an underpinning mixed-methods implementation strategy and an iterative approach to address clinician feedback.

Internal and External Validation of Machine Learning Models for Predicting Acute Kidney Injury Following Non-Cardiac Surgery Using Open Datasets

This study developed and validated a machine learning model to accurately predict acute kidney injury (AKI) after non-cardiac surgery, aiming to improve patient outcomes by assessing its clinical feasibility and generalizability. We conducted a retrospective cohort study using data from 76,032 adults who underwent non-cardiac surgery at a single tertiary medical center between March 2019 and February 2021, and used data from 5512 patients from the VitalDB open dataset for external model validation. The predictive variables for model training consisted of demographic, preoperative laboratory, and intraoperative data, including calculated statistical values such as the minimum, maximum, and mean intraoperative blood pressure. When predicting postoperative AKI, our gradient boosting machine model incorporating all the variables achieved the best results, with AUROC values of 0.868 and 0.757 for the internal and external validations using the VitalDB dataset, respectively. The model using intraoperative data performed best in internal validation, while the model with preoperative data excelled in external validation. In this study, we developed a predictive model for postoperative AKI in adult patients undergoing non-cardiac surgery using preoperative and intraoperative data, and external validation demonstrated the efficacy of open datasets for generalization in medical artificial modeling research.

Development and validation of a medication-based risk prediction model for acute kidney injury in older outpatients

BACKGROUND

Older adults are at an increased risk of acute kidney injury (AKI), particularly in community settings, often due to medications. Effective prevention hinges on identifying high-risk patients, yet existing models for predicting AKI risk in older outpatients are scarce, particularly those incorporating medication variables. We aimed to develop an AKI risk prediction model that included medication-related variables for older outpatients.

METHODS

We constructed a cohort of 2,272,257 outpatients aged ≥65 years using a national claims database. This cohort was split into a development (70%) and validation (30%) groups. Our primary goal was to identify newly diagnosed AKI within one month of cohort entry in an outpatient context. We screened 170 variables and developed a risk prediction model using logistic regression.

RESULTS

The final model integrated 12 variables: 2 demographic, 4 comorbid, and 6 medication-related. It showed good performance with acceptable calibration. In the validation cohort, the area under the receiver operating characteristic curve value was 0.720 (95% confidence interval, 0.692-0.748). Sensitivity and specificity were 69.9% and 61.9%, respectively. Notably, the model identified high-risk patients as having a 27-fold increased AKI risk compared with low-risk individuals.

CONCLUSION

We have developed a new AKI risk prediction model for older outpatients, incorporating critical medication-related variables with good discrimination. This tool may be useful in identifying and targeting patients who may require interventions to prevent AKI in an outpatient setting.

Predictive modeling of co-infection in lupus nephritis using multiple machine learning algorithms

This study aimed to analyze peripheral blood lymphocyte subsets in lupus nephritis (LN) patients and use machine learning (ML) methods to establish an effective algorithm for predicting co-infection in LN. This study included 111 non-infected LN patients, 72 infected LN patients, and 206 healthy controls (HCs). Patient information, infection characteristics, medication, and laboratory indexes were recorded. Eight ML methods were compared to establish a model through a training group and verify the results in a test group. We trained the ML models, including Logistic Regression, Decision Tree, K-Nearest Neighbors, Support Vector Machine, Multi-Layer Perceptron, Random Forest, Ada boost, Extreme Gradient Boosting (XGB), and further evaluated potential predictors of infection. Infected LN patients had significantly decreased levels of T, B, helper T, suppressor T, and natural killer cells compared to non-infected LN patients and HCs. The number of regulatory T cells (Tregs) in LN patients was significantly lower than in HCs, with infected patients having the lowest Tregs count. Among the ML algorithms, XGB demonstrated the highest accuracy and precision for predicting LN infections. The innate and adaptive immune systems are disrupted in LN patients, and monitoring lymphocyte subsets can help prevent and treat infections. The XGB algorithm was recommended for predicting co-infection in LN.

[Artificial intelligence and acute kidney injury]

Digitalization is increasingly finding its way into intensive care units and with it artificial intelligence (AI) for critically ill patients. One promising area for the use of AI is in the field of acute kidney injury (AKI). The use of AI is primarily focused on the prediction of AKI, but further approaches are also being used to classify existing AKI into different phenotypes. Different AI models are used for prediction. The area under the receiver operating characteristic curve values (AUROC) achieved with these models vary and are influenced by several factors, such as the prediction time and the definition of AKI. Most models have an AUROC between 0.650 and 0.900, with lower values for predictions further into the future and when applying Acute Kidney Injury Network (AKIN) instead of KDIGO criteria. Classification into phenotypes already makes it possible to categorize patients into groups with different risks of mortality or requirement of renal replacement therapy (RRT), but the etiologies or therapeutic consequences derived from this are still lacking. However, all the models suffer from AI-specific shortcomings. The use of large databases does not make it possible to promptly include recent changes in therapy and the implementation of new biomarkers in a relevant proportion. For this reason, serum creatinine and urinary output, with their known limitations, dominate current AI models for prediction impairing the performance of the current models. On the other hand, the increasingly complex models no longer allow physicians to understand the basis on which the warning of a threatening AKI is calculated and subsequent initiation of therapy should take place. The successful use of AIs in routine clinical practice will be highly determined by the trust of the physicians in the systems and overcoming the aforementioned weaknesses. However, the clinician will remain irreplaceable as the decisive authority for critically ill patients by combining measurable and nonmeasurable parameters.

Applicable Machine Learning Model for Predicting Contrast-induced Nephropathy Based on Pre-catheterization Variables

Objective Contrast agents used for radiological examinations are an important cause of acute kidney injury (AKI). We developed and validated a machine learning and clinical scoring prediction model to stratify the risk of contrast-induced nephropathy, considering the limitations of current classical and machine learning models. Methods This retrospective study included 38,481 percutaneous coronary intervention cases from 23,703 patients in a tertiary hospital. We divided the cases into development and internal test sets (8:2). Using the development set, we trained a gradient boosting machine prediction model (complex model). We then developed a simple model using seven variables based on variable importance. We validated the performance of the models using an internal test set and tested them externally in two other hospitals. Results The complex model had the best area under the receiver operating characteristic (AUROC) curve at 0.885 [95% confidence interval (CI) 0.876-0.894] in the internal test set and 0.837 (95% CI 0.819-0.854) and 0.850 (95% CI 0.781-0.918) in two different external validation sets. The simple model showed an AUROC of 0.795 (95% CI 0.781-0.808) in the internal test set and 0.766 (95% CI 0.744-0.789) and 0.782 (95% CI 0.687-0.877) in the two different external validation sets. This was higher than the value in the well-known scoring system (Mehran criteria, AUROC=0.67). The seven precatheterization variables selected for the simple model were age, known chronic kidney disease, hematocrit, troponin I, blood urea nitrogen, base excess, and N-terminal pro-brain natriuretic peptide. The simple model is available at http://52.78.230.235:8081/Conclusions We developed an AKI prediction machine learning model with reliable performance. This can aid in bedside clinical decision making.

Merging machine learning and patient preference: a novel tool for risk prediction of percutaneous coronary interventions

BACKGROUND AND AIMS

Predicting personalized risk for adverse events following percutaneous coronary intervention (PCI) remains critical in weighing treatment options, employing risk mitigation strategies, and enhancing shared decision-making. This study aimed to employ machine learning models using pre-procedural variables to accurately predict common post-PCI complications.

METHODS

A group of 66 adults underwent a semiquantitative survey assessing a preferred list of outcomes and model display. The machine learning cohort included 107 793 patients undergoing PCI procedures performed at 48 hospitals in Michigan between 1 April 2018 and 31 December 2021 in the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) registry separated into training and validation cohorts. External validation was conducted in the Cardiac Care Outcomes Assessment Program database of 56 583 procedures in 33 hospitals in Washington.

RESULTS

Overall rate of in-hospital mortality was 1.85% (n = 1999), acute kidney injury 2.51% (n = 2519), new-onset dialysis 0.44% (n = 462), stroke 0.41% (n = 447), major bleeding 0.89% (n = 942), and transfusion 2.41% (n = 2592). The model demonstrated robust discrimination and calibration for mortality {area under the receiver-operating characteristic curve [AUC]: 0.930 [95% confidence interval (CI) 0.920-0.940]}, acute kidney injury [AUC: 0.893 (95% CI 0.883-0.903)], dialysis [AUC: 0.951 (95% CI 0.939-0.964)], stroke [AUC: 0.751 (95%CI 0.714-0.787)], transfusion [AUC: 0.917 (95% CI 0.907-0.925)], and major bleeding [AUC: 0.887 (95% CI 0.870-0.905)]. Similar discrimination was noted in the external validation population. Survey subjects preferred a comprehensive list of individually reported post-procedure outcomes.

CONCLUSIONS

Using common pre-procedural risk factors, the BMC2 machine learning models accurately predict post-PCI outcomes. Utilizing patient feedback, the BMC2 models employ a patient-centred tool to clearly display risks to patients and providers (https://shiny.bmc2.org/pci-prediction/). Enhanced risk prediction prior to PCI could help inform treatment selection and shared decision-making discussions.

Predictive modeling for acute kidney injury after percutaneous coronary intervention in patients with acute coronary syndrome: a machine learning approach

BACKGROUND

Acute kidney injury (AKI) is one of the preventable complications of percutaneous coronary intervention (PCI). This study aimed to develop machine learning (ML) models to predict AKI after PCI in patients with acute coronary syndrome (ACS).

METHODS

This study was conducted at Tehran Heart Center from 2015 to 2020. Several variables were used to design five ML models: Naïve Bayes (NB), Logistic Regression (LR), CatBoost (CB), Multi-layer Perception (MLP), and Random Forest (RF). Feature importance was evaluated with the RF model, CB model, and LR coefficients while SHAP beeswarm plots based on the CB model were also used for deriving the importance of variables in the population using pre-procedural variables and all variables. Sensitivity, specificity, and the area under the receiver operating characteristics curve (ROC-AUC) were used as the evaluation measures.

RESULTS

A total of 4592 patients were included, and 646 (14.1%) experienced AKI. The train data consisted of 3672 and the test data included 920 cases. The patient population had a mean age of 65.6 ± 11.2 years and 73.1% male predominance. Notably, left ventricular ejection fraction (LVEF) and fasting plasma glucose (FPG) had the highest feature importance when training the RF model on only pre-procedural features. SHAP plots for all features demonstrated LVEF and age as the top features. With pre-procedural variables only, CB had the highest AUC for the prediction of AKI (AUC 0.755, 95% CI 0.713 to 0.797), while RF had the highest sensitivity (75.9%) and MLP had the highest specificity (64.35%). However, when considering pre-procedural, procedural, and post-procedural features, RF outperformed other models (AUC: 0.775). In this analysis, CB achieved the highest sensitivity (82.95%) and NB had the highest specificity (82.93%).

CONCLUSION

Our analyses showed that ML models can predict AKI with acceptable performance. This has potential clinical utility for assessing the individualized risk of AKI in ACS patients undergoing PCI. Additionally, the identified features in the models may aid in mitigating these risk factors.

Development and validation of a predictive models for predicting the cardiac events within one year for patients underwent percutaneous coronary intervention procedure at IJN

PURPOSE

Percutaneous coronary intervention (PCI) is a common treatment modality for coronary artery disease. Accurate prediction of patients at risk for complications and hospital readmission after PCI could improve the overall clinical management. We aimed to develop and validate predictive models to predict any cardiac event within a year post PCI procedure.

METHODS

This is a retrospective cohort study utilizing data from the National Cardiovascular Disease (NCVD)-PCI registry. The data collected (N = 28,007) were split into training set (n = 24,409) and testing set (n = 3598). Four predictive models (logistic regression [LR], random forest method, support vector machine [SVM], and artificial neural network) were developed and validated. The outcome on risk prediction were compared.

RESULTS

The demographic and clinical features of patients in the training and testing cohorts were similar. Patients had mean age ± standard deviation of 58.15 ± 10.13 years at admission with a male majority (82.66%). In over half of the procedures (50.61%), patients had chronic stable angina. Within 1 year of follow up mortality, target vessel revascularization (TVR), and composite event of mortality and TVR were 3.92%, 9.48%, and 12.98% respectively. LR was the best model in predicting mortality event within 1-year post-PCI (AUC: 0.820). SVM had the highest discrimination power for both TVR event (AUC: 0.720) and composite event of mortality and TVR (AUC: 0.720).

CONCLUSIONS

This study successfully identified optimal prediction models with the good discriminatory ability for mortality outcome and good discrimination ability for TVR and composite event of mortality and TVR with a simple machine learning framework.

Interpretable machine learning models for early prediction of acute kidney injury after cardiac surgery

OBJECTIVE

Postoperative acute kidney injury (PO-AKI) is a common complication after cardiac surgery. We aimed to evaluate whether machine learning algorithms could significantly improve the risk prediction of PO-AKI.

METHODS

The retrospective cohort study included 2310 adult patients undergoing cardiac surgery in a tertiary teaching hospital, China. Postoperative AKI and severe AKI were identified by the modified KDIGO definition. The sample was randomly divided into a derivation set and a validation set based on a ratio of 4:1. Exploiting conventional logistic regression (LR) and five ML algorithms including decision tree, random forest, gradient boosting classifier (GBC), Gaussian Naive Bayes and multilayer perceptron, we developed and validated the prediction models of PO-AKI. We implemented the interpretation of models using SHapley Additive exPlanation (SHAP) analysis.

RESULTS

Postoperative AKI and severe AKI occurred in 1020 (44.2%) and 286 (12.4%) patients, respectively. Compared with the five ML models, LR model for PO-AKI exhibited the largest AUC (0.812, 95%CI: 0.756, 0.860, all P < 0.05), sensitivity (0.774, 95%CI: 0.719, 0.813), accuracy (0.753, 95%CI: 0.719, 0.781) and Youden index (0.513, 95%CI: 0.451, 0.573). Regarding severe AKI, GBC algorithm showed a significantly higher AUC than the other four ML models (all P < 0.05). Although no significant difference (P = 0.173) was observed in AUCs between GBC (0.86, 95%CI: 0.808, 0.902) and conventional logistic regression (0.803, 95%CI: 0.746, 0.852), GBC achieved greater sensitivity, accuracy and Youden index than conventional LR. Notably, SHAP analyses showed that preoperative serum creatinine, hyperlipidemia, lipid-lowering agents and assisted ventilation time were consistently among the top five important predictors for both postoperative AKI and severe AKI.

CONCLUSION

Logistic regression and GBC algorithm demonstrated moderate to good discrimination and superior performance in predicting PO-AKI and severe AKI, respectively. Interpretation of the models identified the key contributors to the predictions, which could potentially inform clinical interventions.

Associations of dietary, sociodemographic, and anthropometric factors with anemia among the Zhuang ethnic adults: a cross-sectional study in Guangxi Zhuang Autonomous Region, China

BACKGROUND

After decades of rapid economic development, anemia remains a significant public health challenge globally. This study aimed to estimate the associations of sociodemographic, dietary, and body composition factors with anemia among the Zhuang in Guangxi Zhuang Autonomous Region, China.

METHODS

Our study population from the baseline survey of the Guangxi ethnic minority Cohort Study of Chronic Diseases consisted of 13,465 adults (6,779 women and 6,686 men) aged 24-82 years. A validated interviewer-administered laptop-based questionnaire system was used to collect information on participants' sociodemographic, lifestyle, and dietary factors. Each participant underwent a physical examination, and hematological indices were measured. Least absolute shrinkage and selection operator (LASSO) regression was used to select the variables, and logistic regression was applied to estimate the associations of independent risk factors with anemia.

RESULTS

The overall prevalences of anemia in men and women were 9.63% (95% CI: 8.94-10.36%) and 18.33% (95% CI: 17.42─19.28%), respectively. LASSO and logistic regression analyses showed that age was positively associated with anemia for both women and men. For diet in women, red meat consumption for 5-7 days/week (OR = 0.79, 95% CI: 0.65-0.98, p = 0.0290) and corn/sweet potato consumption for 5-7 days/week (OR = 0.73, 95% CI: 0.55-0.96, p = 0.0281) were negatively associated with anemia. For men, fruit consumption for 5-7 days/week (OR = 0.75, 95% CI: 0.60-0.94, p = 0.0130) and corn/sweet potato consumption for 5-7 days/week (OR = 0.66, 95% CI: 0.46-0.91, p = 0.0136) were negatively correlated with anemia. Compared with a normal body water percentage (55-65%), a body water percentage below normal (< 55%) was negatively related to anemia (OR = 0.68, 95% CI: 0.53-0.86, p = 0.0014). Conversely, a body water percentage above normal (> 65%) was positively correlated with anemia in men (OR = 1.73, 95% CI: 1.38-2.17, p < 0.0001).

CONCLUSIONS

Anemia remains a moderate public health problem for premenopausal women and the elderly population in the Guangxi Zhuang minority region. The prevention of anemia at the population level requires multifaceted intervention measures according to sex and age, with a focus on dietary factors and the control of body composition.

Development of real-time individualized risk prediction models for contrast associated acute kidney injury and 30-day dialysis after contrast enhanced computed tomography

PURPOSE

This study aimed to develop preprocedural real-time artificial intelligence (AI)-based systems for predicting individualized risks of contrast-associated acute kidney injury (CA-AKI) and dialysis requirement within 30 days following contrast-enhanced computed tomography (CECT).

METHOD

This single-center, retrospective study analyzed adult patients from emergency or in-patient departments who underwent CECT; 18,895 patients were included after excluding those who were already on dialysis, had stage V chronic kidney disease, or had missing data regarding serum creatinine levels within 7 days before and after CECT. Clinical parameters, laboratory data, medication exposure, and comorbid diseases were selected as predictive features. The patients were randomly divided into model training and testing groups at a 7:3 ratio. Logistic regression (LR) and random forest (RF) were employed to create prediction models, which were evaluated using receiver operating characteristic curves.

RESULTS

The incidence rates of CA-AKI and dialysis within 30 days post-CECT were 6.69% and 0.98%, respectively. For CA-AKI prediction, LR and RF exhibited similar performance, with areas under curve (AUCs) of 0.769 and 0.757, respectively. For 30-day dialysis prediction, LR (AUC, 0.863) and RF (AUC, 0.872) also exhibited similar performance. Relative to eGFR-alone, the LR and RF models produced significantly higher AUCs for CA-AKI prediction (LR vs. eGFR alone, 0.769 vs. 0.626, p < 0.001) and 30-day dialysis prediction (RF vs. eGFR alone, 0.872 vs. 0.738, p < 0.001).

CONCLUSIONS

The proposed AI prediction models significantly outperformed eGFR-alone for predicting the CA-AKI and 30-day dialysis risks of emergency department and hospitalized patients who underwent CECT.

Contemporary Methods for Predicting Acute Kidney Injury After Coronary Intervention

BACKGROUND

Acute kidney injury (AKI) is the most common complication after percutaneous coronary intervention (PCI). Accurately estimating patients' risks not only creates a means of benchmarking performance but can also be used prospectively to inform practice.

OBJECTIVES

The authors sought to update the 2014 National Cardiovascular Data Registry (NCDR) AKI risk model to provide contemporary estimates of AKI risk after PCI to further improve care.

METHODS

Using the NCDR CathPCI Registry, we identified all 2020 PCIs, excluding those on dialysis or lacking postprocedural creatinine. The cohort was randomly split into a 70% derivation cohort and a 30% validation cohort, and logistic regression models were built to predict AKI (an absolute increase of 0.3 mg/dL in creatinine or a 50% increase from preprocedure baseline) and AKI requiring dialysis. Bedside risk scores were created to facilitate prospective use in clinical care, along with threshold contrast doses to reduce AKI. We tested model calibration and discrimination in the validation cohort.

RESULTS

Among 455,806 PCI procedures, the median age was 67 years (IQR: 58.0-75.0 years), 68.8% were men, and 86.8% were White. The incidence of AKI and new dialysis was 7.2% and 0.7%, respectively. Baseline renal function and variables associated with clinical instability were the strongest predictors of AKI. The final AKI model included 13 variables, with a C-statistic of 0.798 and excellent calibration (intercept = -0.03 and slope = 0.97) in the validation cohort.

CONCLUSIONS

The updated NCDR AKI risk model further refines AKI prediction after PCI, facilitating enhanced clinical care, benchmarking, and quality improvement.

A novel explainable online calculator for contrast-induced AKI in diabetics: a multi-centre validation and prospective evaluation study

BACKGROUND

In patients undergoing percutaneous coronary intervention (PCI), contrast-induced acute kidney injury (CIAKI) is a frequent complication, especially in diabetics, and is connected with severe mortality and morbidity in the short and long term. Therefore, we aimed to develop a CIAKI predictive model for diabetic patients.

METHODS

3514 patients with diabetes from four hospitals were separated into three cohorts: training, internal validation, and external validation. We developed six machine learning (ML) algorithms models: random forest (RF), gradient-boosted decision trees (GBDT), logistic regression (LR), least absolute shrinkage and selection operator with LR, extreme gradient boosting trees (XGBT), and support vector machine (SVM). The area under the receiver operating characteristic curve (AUC) of ML models was compared to the prior score model, and developed a brief CIAKI prediction model for diabetes (BCPMD). We also validated BCPMD model on the prospective cohort of 172 patients from one of the hospitals. To explain the prediction model, the shapley additive explanations (SHAP) approach was used.

RESULTS

In the six ML models, XGBT performed best in the cohort of internal (AUC: 0.816 (95% CI 0.777-0.853)) and external validation (AUC: 0.816 (95% CI 0.770-0.861)), and we determined the top 15 important predictors in XGBT model as BCPMD model variables. The features of BCPMD included acute coronary syndromes (ACS), urine protein level, diuretics, left ventricular ejection fraction (LVEF) (%), hemoglobin (g/L), congestive heart failure (CHF), stable Angina, uric acid (umol/L), preoperative diastolic blood pressure (DBP) (mmHg), contrast volumes (mL), albumin (g/L), baseline creatinine (umol/L), vessels of coronary artery disease, glucose (mmol/L) and diabetes history (yrs). Then, we validated BCPMD in the cohort of internal validation (AUC: 0.819 (95% CI 0.783-0.855)), the cohort of external validation (AUC: 0.805 (95% CI 0.755-0.850)) and the cohort of prospective validation (AUC: 0.801 (95% CI 0.688-0.887)). SHAP was constructed to provide personalized interpretation for each patient. Our model also has been developed into an online web risk calculator. MissForest was used to handle the missing values of the calculator.

CONCLUSION

We developed a novel risk calculator for CIAKI in diabetes based on the ML model, which can help clinicians achieve real-time prediction and explainable clinical decisions.

An interpretable machine learning model for the prevention of contrast-induced nephropathy in patients undergoing lower extremity endovascular interventions for peripheral arterial disease

BACKGROUND

Contrast-induced nephropathy (CIN) is a postprocedural complication associated with increased morbidity and mortality. An important risk factor for development of CIN is renal impairment. Identification of patients at risk for acute renal failure will allow physicians to make appropriate decisions to minimize the incidence of CIN. We developed a machine learning model to stratify risk of acute renal failure that may assist in mitigating risk for CIN in patients with peripheral artery disease (PAD) undergoing endovascular interventions.

METHODS

We utilized the American College of Surgeons National Surgical Quality Improvement Program database to extract clinical and laboratory information associated with 14,444 patients who underwent lower extremity endovascular procedures between 2011 and 2018. Using 11,604 cases from 2011 to 2017 for training and 2840 cases from 2018 for testing, we developed a random forest model to predict risk of 30-day acute renal failure following infra-inguinal endovascular procedures.

RESULTS

Eight variables were identified as contributing optimally to model predictions, the most important being diabetes, preoperative BUN, and claudication. Using these variables, the model achieved an area under the receiver-operating characteristic (AU-ROC) curve of 0.81, accuracy of 0.83, sensitivity of 0.67, and specificity of 0.74. The model performed equally well on white and nonwhite patients (Delong p-value = 0.955) and patients age < 65 and patients age ≥ 65 (Delong p-value = 0.659).

CONCLUSIONS

We develop a model that fairly and accurately stratifies 30-day acute renal failure risk in patients undergoing lower extremity endovascular procedures for PAD. This model may assist in identifying patients who may benefit from strategies to prevent CIN.

Characterization of Risk Prediction Models for Acute Kidney Injury: A Systematic Review and Meta-analysis

Importance

Despite the expansion of published prediction models for acute kidney injury (AKI), there is little evidence of uptake of these models beyond their local derivation nor data on their association with patient outcomes.

Objective

To systematically review published AKI prediction models across all clinical subsettings.

Data Sources

MEDLINE via PubMed (January 1946 to April 2021) and Embase (January 1947 to April 2021) were searched using medical subject headings and text words related to AKI and prediction models.

Study Selection

All studies that developed a prediction model for AKI, defined as a statistical model with at least 2 predictive variables to estimate future occurrence of AKI, were eligible for inclusion. There was no limitation on study populations or methodological designs.

Data Extraction and Synthesis

Two authors independently searched the literature, screened the studies, and extracted and analyzed the data following the Preferred Reporting Items for Systematic Review and Meta-analyses guideline. The data were pooled using a random-effects model, with subgroups defined by 4 clinical settings. Between-study heterogeneity was explored using multiple methods, and funnel plot analysis was used to identify publication bias.

Main Outcomes and Measures

C statistic was used to measure the discrimination of prediction models.

Results

Of the 6955 studies initially identified through literature searching, 150 studies, with 14.4 million participants, met the inclusion criteria. The study characteristics differed widely in design, population, AKI definition, and model performance assessments. The overall pooled C statistic was 0.80 (95% CI, 0.79-0.81), with pooled C statistics in different clinical subsettings ranging from 0.78 (95% CI, 0.75-0.80) to 0.82 (95% CI, 0.78-0.86). Between-study heterogeneity was high overall and in the different clinical settings (eg, contrast medium-associated AKI: I2 = 99.9%; P < .001), and multiple methods did not identify any clear sources. A high proportion of models had a high risk of bias (126 [84.4%]) according to the Prediction Model Risk Of Bias Assessment Tool.

Conclusions and Relevance

In this study, the discrimination of the published AKI prediction models was good, reflected by high C statistics; however, the wide variation in the clinical settings, populations, and predictive variables likely drives the highly heterogenous findings that limit clinical utility. Standardized procedures for development and validation of prediction models are urgently needed.

Prediction of Prednisolone Dose Correction Using Machine Learning

Wrong dose, a common prescription error, can cause serious patient harm, especially in the case of high-risk drugs like oral corticosteroids. This study aims to build a machine learning model to predict dose-related prescription modifications for oral prednisolone tablets (i.e., highly imbalanced data with very few positive cases). Prescription data were obtained from the electronic medical records at a single institute. Cluster analysis classified the clinical departments into six clusters with similar patterns of prednisolone prescription. Two patterns of training datasets were created with/without preprocessing by the SMOTE method. Five ML models (SVM, KNN, GB, RF, and BRF) and logistic regression (LR) models were constructed by Python. The model was internally validated by five-fold stratified cross-validation and was validated with a 30% holdout test dataset. Eighty-two thousand five hundred fifty-three prescribing data for prednisolone tablets containing 135 dose-corrected positive cases were obtained. In the original dataset (without SMOTE), only the BRF model showed a good performance (in test dataset, ROC-AUC:0.917, recall: 0.951). In the training dataset preprocessed by SMOTE, performance was improved on all models. The highest performance models with SMOTE were SVM (in test dataset, ROC-AUC: 0.820, recall: 0.659) and BRF (ROC-AUC: 0.814, recall: 0.634). Although the prescribing data for dose-related collection are highly imbalanced, various techniques such as the following have allowed us to build high-performance prediction models: data preprocessing by SMOTE, stratified cross-validation, and BRF classifier corresponding to imbalanced data. ML is useful in complicated dose audits such as oral prednisolone.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41666-023-00128-3.

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.

Using machine learning for mortality prediction and risk stratification in atezolizumab-treated cancer patients: Integrative analysis of eight clinical trials

BACKGROUND

Few models exist to predict mortality in cancer patients receiving immunotherapy. Our aim was to build a machine learning-based risk stratification model for predicting mortality in atezolizumab-treated cancer patients.

METHODS

Data from 2538 patients in eight atezolizumab-treated cancer clinical trials across three cancer types (non-small-cell lung cancer, bladder transitional cell carcinoma, and renal cell carcinoma) were included. The whole cohort was randomly split into development and validation cohorts in a 7:3 ratio. Machine-learning algorithms (extreme gradient boosting, random forest, logistic regression with lasso regularization, support vector machine, and K-nearest neighbor) were applied to develop prediction models. Model performance was mainly assessed by area under the receiver operating characteristic curve (AUC) value, calibration plot, and decision curve analysis. The probability of death risk was then stratified.

RESULTS

One thousand and three hundred and seventy-nine (54.33%) patients died. The random forest (RF) model was overall the best in terms of predictive performance, with the AUC of 0.844 (95% confidence interval [CI]: 0.826-0.862) in the development cohort and 0.786 (95% CI: 0.754-0.818) in the validation cohort for predicting mortality. Twelve baseline variables contributing to mortality prediction in the RF model were C-reactive protein, PD-L1 level, cancer type, prior liver metastasis, derived neutrophil-to-lymphocyte ratio, alkaline phosphatase, albumin, hemoglobin, white blood cell count, number of metastatic sites, pulse rate, and Eastern Cooperative Oncology Group (ECOG) performance status. A total of 1782 (70.2%) patients were separated into the high-risk and 756 (29.8%) low-risk groups. Patients in the high-risk group were significantly more likely to die, experience disease progression, discontinue study, and discontinue treatment than patients in the low-risk group (all p values < 0.001). Risk groups were not associated with immune-related adverse events and grades 3-5 treatment-related adverse events (all p values > 0.05).

CONCLUSION

RF model has good performance in mortality prediction and risk stratification for cancer patients receiving atezolizumab monotherapy.

A risk prediction model for acute kidney injury in patients with pulmonary tuberculosis during anti-tuberculosis treatment

BACKGROUND

Acute kidney injury (AKI) is not a rare complication during anti-tuberculosis treatment in some patients with pulmonary tuberculosis (PTB). We aimed to develop a risk prediction model for early recognition of patients with PTB at high risk for AKI during anti-TB treatment.

METHODS

This retrospective cohort study assessed the clinical baseline, and laboratory test data of 315 inpatients with active PTB who were screened for predictive factors from January 2019 to June 2020. The elements were analyzed by logistic regression analysis. A nomogram was established by the results of the logistic regression analysis. The prediction model discrimination and calibration were evaluated by the concordance index (C-index), ROC curve, and Hosmer-Lemeshow analysis.

RESULTS

A total of 315 patients with PTB were enrolled (67 patients with AKI and 248 patients without AKI). Seven factors, including microalbuminuria, hematuria, cystatin-C (CYS-C), albumin (ALB), creatinine-based estimated glomerular filtration rates (eGFRs), body mass index (BMI), and CA-125 were acquired to develop the predictive model. According to the logistic regression, microalbuminuria (OR = 3.038, 95%CI 1.168-7.904), hematuria (OR = 3.656, 95%CI 1.325-10.083), CYS-C (OR = 4.416, 95%CI 2.296-8.491), and CA-125 (OR = 3.93, 95%CI 1.436-10.756) were risk parameter, while ALB (OR = 0.741, 95%CI 0.650-0.844) was protective parameter. The nomogram demonstrated good prediction in estimating AKI (C-index= 0.967, AUC = 0.967, 95%CI (0.941-0.984), sensitivity = 91.04%, specificity = 93.95%, Hosmer-Lemeshow analysis SD = 0.00054, and quantile of absolute error = 0.049).

CONCLUSIONS

Microalbuminuria, hematuria, ALB reduction, elevated CYS-C, and CA-125 are predictive factors for the development of AKI in patients with PTB during anti-TB treatments. The predictive nomogram based on five predictive factors is achieved good risk prediction for AKI during anti-TB treatments.

Diagnosis of hepatocellular carcinoma based on salivary protein glycopatterns and machine learning algorithms

Objectives

Hepatocellular carcinoma (HCC) is difficult to diagnose early and progresses rapidly, making it one of the most deadly malignancies worldwide. This study aimed to evaluate whether salivary glycopattern changes combined with machine learning algorithms could help in the accurate diagnosis of HCC.

Methods

Firstly, we detected the alteration of salivary glycopatterns by lectin microarrays in 118 saliva samples. Subsequently, we constructed diagnostic models for hepatic cirrhosis (HC) and HCC using three machine learning algorithms: Least Absolute Shrinkage and Selector Operation, Support Vector Machine (SVM), and Random Forest (RF). Finally, the performance of the diagnostic models was assessed in an independent validation cohort of 85 saliva samples by a series of evaluation metrics, including area under the receiver operator curve (AUC), accuracy, specificity, and sensitivity.

Results

We identified alterations in the expression levels of salivary glycopatterns in patients with HC and HCC. The results revealed that the glycopatterns recognized by 22 lectins showed significant differences in the saliva of HC and HCC patients and healthy volunteers. In addition, after Boruta feature selection, the best predictive performance was obtained with the RF algorithm for the construction of models for HC and HCC. The AUCs of the RF-HC model and RF-HCC model in the validation cohort were 0.857 (95% confidence interval [CI]: 0.780–0.935) and 0.886 (95% CI: 0.814–0.957), respectively.

Conclusions

Detecting alterations in salivary protein glycopatterns with lectin microarrays combined with machine learning algorithms could be an effective strategy for diagnosing HCC in the future.

Machine learning in predicting cardiac surgery-associated acute kidney injury: A systemic review and meta-analysis

Background

Cardiac surgery-associated acute kidney injury (CSA-AKI) is a common complication following cardiac surgery. Early prediction of CSA-AKI is of great significance for improving patients' prognoses. The aim of this study is to systematically evaluate the predictive performance of machine learning models for CSA-AKI.

Methods

Cochrane Library, PubMed, EMBASE, and Web of Science were searched from inception to 18 March 2022. Risk of bias assessment was performed using PROBAST. Rsoftware (version 4.1.1) was used to calculate the accuracy and C-index of CSA-AKI prediction. The importance of CSA-AKI prediction was defined according to the frequency of related factors in the models.

Results

There were 38 eligible studies included, with a total of 255,943 patients and 60 machine learning models. The models mainly included Logistic Regression (n = 34), Neural Net (n = 6), Support Vector Machine (n = 4), Random Forest (n = 6), Extreme Gradient Boosting (n = 3), Decision Tree (n = 3), Gradient Boosted Machine (n = 1), COX regression (n = 1), κNeural Net (n = 1), and Naïve Bayes (n = 1), of which 51 models with intact recording in the training set and 17 in the validating set. Variables with the highest predicting frequency included Logistic Regression, Neural Net, Support Vector Machine, and Random Forest. The C-index and accuracy wer 0.76 (0.740, 0.780) and 0.72 (0.70, 0.73), respectively, in the training set, and 0.79 (0.75, 0.83) and 0.73 (0.71, 0.74), respectively, in the test set.

Conclusion

The machine learning-based model is effective for the early prediction of CSA-AKI. More machine learning methods based on noninvasive or minimally invasive predictive indicators are needed to improve the predictive performance and make accurate predictions of CSA-AKI. Logistic regression remains currently the most commonly applied model in CSA-AKI prediction, although it is not the one with the best performance. There are other models that would be more effective, such as NNET and XGBoost.

Systematic review registration

https://www.crd.york.ac.uk/; review registration ID: CRD42022345259.

Development of a clinical prediction model for recurrence and mortality outcomes after Clostridioides difficile infection using a machine learning approach

OBJECTIVES

Clostridioides difficile infection (CDI) is associated with a large burden of morbidity and mortality worldwide. Previous studies have developed models for predicting recurrence and mortality following CDI, but no machine learning predictive models have been developed specifically using data from Japanese patients.

METHODS

Using a database of records from acute care hospitals in Japan, we extracted records from January 2012 to September 2016 (plus a 60-day lookback window). A total of 19,159 patients were included. We used a machine learning approach, XGBoost, and compared it to a traditional unregularized logistic regression model. The first 80% of the dataset (by patient index date) was used to optimize model hyperparameters and train the final models, and evaluation was performed on the remaining 20%. We measured model performance by the area under the receiver operator curve and assessed feature importance using Shapley additive explanations.

RESULTS

Performance was similar between the machine learning approach and the classical logistic regression model. Logistic regression performed slightly better than XGBoost for predicting mortality.

CONCLUSION

XGBoost performed slightly better than logistic regression for predicting recurrence, but it was not competitive with existing published models. Despite this, a future machine learning-based application provided in a bedside setting at low cost might be a clinically useful tool.

Development and Validation of a Personalized Model With Transfer Learning for Acute Kidney Injury Risk Estimation Using Electronic Health Records

IMPORTANCE

Acute kidney injury (AKI) is a heterogeneous syndrome prevalent among hospitalized patients. Personalized risk estimation and risk factor identification may allow effective intervention and improved outcomes.

OBJECTIVE

To develop and validate personalized AKI risk estimation models using electronic health records (EHRs), examine whether personalized models were beneficial in comparison with global and subgroup models, and assess the heterogeneity of risk factors and their outcomes in different subpopulations.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study analyzed EHR data from 1 tertiary care hospital and used machine learning and logistic regression to develop and validate global, subgroup, and personalized risk estimation models. Transfer learning was implemented to enhance the personalized model. Predictor outcomes across subpopulations were analyzed, and metaregression was used to explore predictor interactions. Adults who were hospitalized for 2 or more days from November 1, 2007, to December 31, 2016, were included in the analysis. Patients with moderate or severe kidney dysfunction at admission were excluded. Data were analyzed between August 28, 2019, and May 8, 2022.

EXPOSURES

Clinical and laboratory variables in the EHR.

MAIN OUTCOMES AND MEASURES

The main outcome was AKI of any severity, and AKI was defined using the Kidney Disease: Improving Global Outcomes serum creatinine criteria. Performance of the models was measured with area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve, and calibration.

RESULTS

The study cohort comprised 76 957 inpatient encounters. Patients had a mean (SD) age of 55.5 (17.4) years and included 42 159 men (54.8%). The personalized model with transfer learning outperformed the global model for AKI estimation in terms of AUROC among general inpatients (0.78 [95% CI, 0.77-0.79] vs 0.76 [95% CI, 0.75-0.76]; P < .001) and across the high-risk subgroups (0.79 [95% CI, 0.78-0.80] vs 0.75 [95% CI, 0.74-0.77]; P < .001) and low-risk subgroups (0.74 [95% CI, 0.73-0.75] vs 0.71 [95% CI, 0.70-0.72]; P < .001). The AUROC improvement reached 0.13 for the high-risk subgroups, such as those undergoing liver transplant and cardiac surgery. Moreover, the personalized model with transfer learning performed better than or comparably with the best published models in well-studied AKI subgroups. Predictor outcomes varied significantly between patients, and interaction analysis uncovered modifiers of the predictor outcomes.

CONCLUSIONS AND RELEVANCE

Results of this study demonstrated that a personalized modeling with transfer learning is an improved AKI risk estimation approach that can be used across diverse patient subgroups. Risk factor heterogeneity and interactions at the individual level highlighted the need for agile, personalized care.

Nonlaboratory-based risk assessment model for coronary heart disease screening: Model development and validation

BACKGROUND

Identifying groups at high risk of coronary heart disease (CHD) is important to reduce mortality due to CHD. Although machine learning methods have been introduced, many require laboratory or imaging parameters, which are not always readily available; thus, their wide applications are limited.

OBJECTIVE

The aim of this study was to develop and validate a simple, efficient, and joint machine learning model for identifying individuals at high risk of CHD using easily obtainable nonlaboratory parameters.

METHODS

This prospective study used data from the Henan Rural Cohort Study, which was conducted in rural areas of Henan Province, China, between July 2015 and September 2017. A joint machine learning model was developed by selecting and combining four base machine learning algorithms, including logistic regression (LR), artificial neural network (ANN), random forest (RF), and gradient boosting machine (GBM). We used readily accessible variables, including demographics, medical and family history, lifestyle and dietary factors, and anthropometric data, to inform the model. The model was also externally validated by a cohort of individuals from the Dongfeng-Tongji cohort study. Model discrimination was assessed by using the area under the receiver operating characteristic curve (AUC), and calibration was measured by using the Brier score (BS).

RESULTS

A total of 38 716 participants (mean [SD] age, 55.64[12.19] years; 23449[60.6%] female) from the Henan Rural Cohort Study and 17 958 subjects (mean [SD] age, 62.74 [7.59] years; 10,076 [56.1%] female) from the Dongfeng-Tongji cohort study were included in the analysis. Age, waist circumference, pulse pressure, heart rate, family history of CHD, education level, family history of type 2 diabetes mellitus (T2DM), and family history of dyslipidaemia were strongly associated with the development of CHD. In regard to internal validation, the model we built demonstrated good discrimination (AUC, 0.844 (95% CI 0.828-0.860)) and had acceptable calibration (BS, 0. 066). In regard to external validation, the model performed well with clearly useful discrimination (AUC, 0.792 (95% CI 0.774-0.810)) and robust calibration (BS, 0.069).

CONCLUSIONS

In this study, the novel and simple, machine learning-based model comprising readily accessible variables accurately identified individuals at high risk of CHD. This model has the potential to be widely applied for large-scale screening of CHD populations, especially in medical resource-constrained settings.

TRIAL REGISTRATION

The Henan Rural Cohort Study has been registered at the Chinese Clinical Trial Register. (Trial registration: ChiCTR-OOC-15006699. Registered 6 July 2015 - Retrospectively registered) http://www.chictr.org.cn/showproj.aspx?proj=11375.

A Clinically Applicable Nomogram for Predicting the Risk of Invasive Mechanical Ventilation in Pneumocystis jirovecii Pneumonia

Objective

Pneumocystis jirovecii pneumonia (PCP) is a life-threatening disease associated with a high mortality rate among immunocompromised patient populations. Invasive mechanical ventilation (IMV) is a crucial component of treatment for PCP patients with progressive hypoxemia. This study explored the risk factors for IMV and established a model for early predicting the risk of IMV among patients with PCP.

Methods

A multicenter, observational cohort study was conducted in 10 hospitals in China. Patients diagnosed with PCP were included, and their baseline clinical characteristics were collected. A Boruta analysis was performed to identify potentially important clinical features associated with the use of IMV during hospitalization. Selected variables were further analyzed using univariate and multivariable logistic regression. A logistic regression model was established based on independent risk factors for IMV and visualized using a nomogram.

Results

In total, 103 patients comprised the training cohort for model development, and 45 comprised the validation cohort to confirm the model's performance. No significant differences were observed in baseline clinical characteristics between the training and validation cohorts. Boruta analysis identified eight clinical features associated with IMV, three of which were further confirmed to be independent risk factors for IMV, including age (odds ratio [OR] 2.615 [95% confidence interval (CI) 1.110-6.159]; p = 0.028), oxygenation index (OR 0.217 [95% CI 0.078-0.604]; p = 0.003), and serum lactate dehydrogenase level (OR 1.864 [95% CI 1.040-3.341]; p = 0.037). Incorporating these three variables, the nomogram achieved good concordance indices of 0.829 (95% CI 0.752-0.906) and 0.818 (95% CI 0.686-0.950) in predicting IMV in the training and validation cohorts, respectively, and had well-fitted calibration curves.

Conclusions

The nomogram demonstrated accurate prediction of IMV in patients with PCP. Clinical application of this model enables early identification of patients with PCP who require IMV, which, in turn, may lead to rational therapeutic choices and improved clinical outcomes.

Artificial Intelligence in Cardiovascular Medicine: Historical Overview, Current Status, and Future Directions

Artificial intelligence and machine learning are rapidly gaining popularity in every aspect of our daily lives, and cardiovascular medicine is no exception. Here, we provide physicians with an overview of the past, present, and future of artificial intelligence applications in cardiovascular medicine. We describe essential and powerful examples of machine-learning applications in industry and elsewhere. Finally, we discuss the latest technologic advances, as well as the benefits and limitations of artificial intelligence and machine learning in cardiovascular medicine.

Short- and Long-Term Recovery after Moderate/Severe AKI in Patients with and without COVID-19

Background

Severe AKI is strongly associated with poor outcomes in coronavirus disease 2019 (COVID-19), but data on renal recovery are lacking.

Methods

We retrospectively analyzed these associations in 3299 hospitalized patients (1338 with COVID-19 and 1961 with acute respiratory illness but who tested negative for COVID-19). Uni- and multivariable analyses were used to study mortality and recovery after Kidney Disease Improving Global Outcomes Stages 2 and 3 AKI (AKI-2/3), and Machine Learning was used to predict AKI and recovery using admission data. Long-term renal function and other outcomes were studied in a subgroup of AKI-2/3 survivors.

Results

Among the 172 COVID-19-negative patients with AKI-2/3, 74% had partial and 44% complete renal recovery, whereas 12% died. Among 255 COVID-19 positive patients with AKI-2/3, lower recovery and higher mortality were noted (51% partial renal recovery, 25% complete renal recovery, 24% died). On multivariable analysis, intensive care unit admission and acute respiratory distress syndrome were associated with nonrecovery, and recovery was significantly associated with survival in COVID-19-positive patients. With Machine Learning, we were able to predict recovery from COVID-19-associated AKI-2/3 with an average precision of 0.62, and the strongest predictors of recovery were initial arterial partial pressure of oxygen and carbon dioxide, serum creatinine, potassium, lymphocyte count, and creatine phosphokinase. At 12-month follow-up, among 52 survivors with AKI-2/3, 26% COVID-19-positive and 24% COVID-19-negative patients had incident or progressive CKD.

Conclusions

Recovery from COVID-19-associated moderate/severe AKI can be predicted using admission data and is associated with severity of respiratory disease and in-hospital death. The risk of CKD might be similar between COVID-19-positive and -negative patients.

Machine learning prediction model of acute kidney injury after percutaneous coronary intervention

Acute kidney injury (AKI) after percutaneous coronary intervention (PCI) is associated with a significant risk of morbidity and mortality. The traditional risk model provided by the National Cardiovascular Data Registry (NCDR) is useful for predicting the preprocedural risk of AKI, although the scoring system requires a number of clinical contents. We sought to examine whether machine learning (ML) techniques could predict AKI with fewer NCDR-AKI risk model variables within a comparable PCI database in Japan. We evaluated 19,222 consecutive patients undergoing PCI between 2008 and 2019 in a Japanese multicenter registry. AKI was defined as an absolute or a relative increase in serum creatinine of 0.3 mg/dL or 50%. The data were split into training (N = 16,644; 2008-2017) and testing datasets (N = 2578; 2017-2019). The area under the curve (AUC) was calculated using the light gradient boosting model (GBM) with selected variables by Lasso and SHapley Additive exPlanations (SHAP) methods among 12 traditional variables, excluding the use of an intra-aortic balloon pump, since its use was considered operator-dependent. The incidence of AKI was 9.4% in the cohort. Lasso and SHAP methods demonstrated that seven variables (age, eGFR, preprocedural hemoglobin, ST-elevation myocardial infarction, non-ST-elevation myocardial infarction/unstable angina, heart failure symptoms, and cardiogenic shock) were pertinent. AUC calculated by the light GBM with seven variables had a performance similar to that of the conventional logistic regression prediction model that included 12 variables (light GBM, AUC [training/testing datasets]: 0.779/0.772; logistic regression, AUC [training/testing datasets]: 0.797/0.755). The AKI risk model after PCI using ML enabled adequate risk quantification with fewer variables. ML techniques may aid in enhancing the international use of validated risk models.

Machine learning approach to predict acute kidney injury after liver surgery

BACKGROUND

Acute kidney injury (AKI) after surgery appears to increase the risk of death in patients with liver cancer. In recent years, machine learning algorithms have been shown to offer higher discriminative efficiency than classical statistical analysis.

AIM

To develop prediction models for AKI after liver cancer resection using machine learning techniques.

METHODS

We screened a total of 2450 patients who had undergone primary hepatocellular carcinoma resection at Changzheng Hospital, Shanghai City, China, from January 1, 2015 to August 31, 2020. The AKI definition used was consistent with the Kidney Disease: Improving Global Outcomes. We included in our analysis preoperative data such as demographic characteristics, laboratory findings, comorbidities, and medication, as well as perioperative data such as duration of surgery. Computerized algorithms used for model development included logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGboost), and decision tree (DT). Feature importance was also ranked according to its contribution to model development.

RESULTS

AKI events occurred in 296 patients (12.1%) within 7 d after surgery. Among the original models based on machine learning techniques, the RF algorithm had optimal discrimination with an area under the curve value of 0.92, compared to 0.87 for XGBoost, 0.90 for DT, 0.90 for SVM, and 0.85 for LR. The RF algorithm also had the highest concordance-index (0.86) and the lowest Brier score (0.076). The variable that contributed the most in the RF algorithm was age, followed by cholesterol, and surgery time.

CONCLUSION

Machine learning algorithms are highly effective in discriminating patients at high risk of developing AKI. The successful application of machine learning models may help guide clinical decisions and help improve the long-term prognosis of patients.

Artificial intelligence's role in vascular surgery decision-making

Artificial intelligence (AI) is the next great advance informing medical science. Several disciplines, including vascular surgery, use AI-based decision-making tools to improve clinical performance. Although applied widely, AI functions best when confronted with voluminous, accurate data. Consistent, predictable analytic technique selection also challenges researchers. This article contextualizes AI analyses within evidence-based medicine, focusing on "big data" and health services research, as well as discussing opportunities to improve data collection and realize AI's promise.

Does Artificial Intelligence Make Clinical Decision Better? A Review of Artificial Intelligence and Machine Learning in Acute Kidney Injury Prediction

Acute kidney injury (AKI) is a common complication of hospitalization that greatly and negatively affects the short-term and long-term outcomes of patients. Current guidelines use serum creatinine level and urine output rate for defining AKI and as the staging criteria of AKI. However, because they are not sensitive or specific markers of AKI, clinicians find it difficult to predict the occurrence of AKI and prescribe timely treatment. Advances in computing technology have led to the recent use of machine learning and artificial intelligence in AKI prediction, recent research reported that by using electronic health records (EHR) the AKI prediction via machine-learning models can reach AUROC over 0.80, in some studies even reach 0.93. Our review begins with the background and history of the definition of AKI, and the evolution of AKI risk factors and prediction models is also appraised. Then, we summarize the current evidence regarding the application of e-alert systems and machine-learning models in AKI prediction.

Prediction of long-term survival after gastrectomy using random survival forests

BACKGROUND

No well validated and contemporaneous tools for personalized prognostication of gastric adenocarcinoma exist. This study aimed to derive and validate a prognostic model for overall survival after surgery for gastric adenocarcinoma using a large national dataset.

METHODS

National audit data from England and Wales were used to identify patients who underwent a potentially curative gastrectomy for adenocarcinoma of the stomach. A total of 2931 patients were included and 29 clinical and pathological variables were considered for their impact on survival. A non-linear random survival forest methodology was then trained and validated internally using bootstrapping with calibration and discrimination (time-dependent area under the receiver operator curve (tAUC)) assessed.

RESULTS

The median survival of the cohort was 69 months, with a 5-year survival of 53.2 per cent. Ten variables were found to influence survival significantly and were included in the final model, with the most important being lymph node positivity, pT stage and achieving an R0 resection. Patient characteristics including ASA grade and age were also influential. On validation the model achieved excellent performance with a 5-year tAUC of 0.80 (95 per cent c.i. 0.78 to 0.82) and good agreement between observed and predicted survival probabilities. A wide spread of predictions for 3-year (14.8-98.3 (i.q.r. 43.2-84.4) per cent) and 5-year (9.4-96.1 (i.q.r. 31.7-73.8) per cent) survival were seen.

CONCLUSIONS

A prognostic model for survival after a potentially curative resection for gastric adenocarcinoma was derived and exhibited excellent discrimination and calibration of predictions.

Importance of socioeconomic factors in predicting tooth loss among older adults in Japan: Evidence from a machine learning analysis

Prevalence of tooth loss has increased due to population aging. Tooth loss negatively affects the overall physical and social well-being of older adults. Understanding the role of socio-demographic and other predictors associated with tooth loss that are measured in non-clinical settings can be useful in community-level prevention. We used high-dimensional epidemiological data to investigate important factors in predicting tooth loss among older adults over a 6-year period of follow-up. Data was from participants of 2010 and 2016 waves of the Japan Gerontological Evaluation Study (JAGES). A total of 19,407 community-dwelling functionally independent older adults aged 65 and older were included in the analysis. Tooth loss was measured as moving from a higher number of teeth category at the baseline to a lower number of teeth category at the follow-up. Out of 119 potential predictors, age, sex, number of teeth, denture use, chewing difficulty, household income, employment, education, smoking, fruit and vegetable consumption, community participation, time since last health check-up, having a hobby, and feeling worthless were selected using Boruta algorithm. Within the 6-year follow-up, 3013 individuals (15.5%) reported incidence of tooth loss. People who experienced tooth loss were older (72.9 ± 5.2 vs 71.8 ± 4.7), and predominantly men (18.3% vs 13.1%). Extreme gradient boosting (XGBoost) machine learning prediction model had a mean accuracy of 90.5% (±0.9%). A visual analysis of machine learning predictions revealed that the prediction of tooth loss was mainly driven by demographic (older age), baseline oral health (having 10-19 teeth, wearing dentures), and socioeconomic (lower household income, manual occupations) variables. Predictors related to wide a range of determinants contribute towards tooth loss among older adults. In addition to oral health related and demographic factors, socioeconomic factors were important in predicting future tooth loss. Understanding the behaviour of these predictors can thus be useful in developing prevention strategies for tooth loss among older adults.

Performance Metrics for the Comparative Analysis of Clinical Risk Prediction Models Employing Machine Learning

BACKGROUND

New methods such as machine learning techniques have been increasingly used to enhance the performance of risk predictions for clinical decision-making. However, commonly reported performance metrics may not be sufficient to capture the advantages of these newly proposed models for their adoption by health care professionals to improve care. Machine learning models often improve risk estimation for certain subpopulations that may be missed by these metrics.

METHODS AND RESULTS

This article addresses the limitations of commonly reported metrics for performance comparison and proposes additional metrics. Our discussions cover metrics related to overall performance, discrimination, calibration, resolution, reclassification, and model implementation. Models for predicting acute kidney injury after percutaneous coronary intervention are used to illustrate the use of these metrics.

CONCLUSIONS

We demonstrate that commonly reported metrics may not have sufficient sensitivity to identify improvement of machine learning models and propose the use of a comprehensive list of performance metrics for reporting and comparing clinical risk prediction models.

Data mining in clinical big data: the frequently used databases, steps, and methodological models

Many high quality studies have emerged from public databases, such as Surveillance, Epidemiology, and End Results (SEER), National Health and Nutrition Examination Survey (NHANES), The Cancer Genome Atlas (TCGA), and Medical Information Mart for Intensive Care (MIMIC); however, these data are often characterized by a high degree of dimensional heterogeneity, timeliness, scarcity, irregularity, and other characteristics, resulting in the value of these data not being fully utilized. Data-mining technology has been a frontier field in medical research, as it demonstrates excellent performance in evaluating patient risks and assisting clinical decision-making in building disease-prediction models. Therefore, data mining has unique advantages in clinical big-data research, especially in large-scale medical public databases. This article introduced the main medical public database and described the steps, tasks, and models of data mining in simple language. Additionally, we described data-mining methods along with their practical applications. The goal of this work was to aid clinical researchers in gaining a clear and intuitive understanding of the application of data-mining technology on clinical big-data in order to promote the production of research results that are beneficial to doctors and patients.

Prognostically relevant periprocedural myocardial injury and infarction associated with percutaneous coronary interventions: a Consensus Document of the ESC Working Group on Cellular Biology of the Heart and European Association of Percutaneous Cardiovascular Interventions (EAPCI)

A substantial number of chronic coronary syndrome (CCS) patients undergoing percutaneous coronary intervention (PCI) experience periprocedural myocardial injury or infarction. Accurate diagnosis of these PCI-related complications is required to guide further management given that their occurrence may be associated with increased risk of major adverse cardiac events (MACE). Due to lack of scientific data, the cut-off thresholds of post-PCI cardiac troponin (cTn) elevation used for defining periprocedural myocardial injury and infarction, have been selected based on expert consensus opinions, and their prognostic relevance remains unclear. In this Consensus Document from the ESC Working Group on Cellular Biology of the Heart and European Association of Percutaneous Cardiovascular Interventions (EAPCI), we recommend, whenever possible, the measurement of baseline (pre-PCI) cTn and post-PCI cTn values in all CCS patients undergoing PCI. We confirm the prognostic relevance of the post-PCI cTn elevation >5× 99th percentile URL threshold used to define type 4a myocardial infarction (MI). In the absence of periprocedural angiographic flow-limiting complications or electrocardiogram (ECG) and imaging evidence of new myocardial ischaemia, we propose the same post-PCI cTn cut-off threshold (>5× 99th percentile URL) be used to define prognostically relevant ‘major’ periprocedural myocardial injury. As both type 4a MI and major periprocedural myocardial injury are strong independent predictors of all-cause mortality at 1 year post-PCI, they may be used as quality metrics and surrogate endpoints for clinical trials. Further research is needed to evaluate treatment strategies for reducing the risk of major periprocedural myocardial injury, type 4a MI, and MACE in CCS patients undergoing PCI.

Toward Dynamic Risk Prediction of Outcomes After Coronary Artery Bypass Graft: Improving Risk Prediction With Intraoperative Events Using Gradient Boosting

BACKGROUND

Intraoperative data may improve models predicting postoperative events. We evaluated the effect of incorporating intraoperative variables to the existing preoperative model on the predictive performance of the model for coronary artery bypass graft.

METHODS

We analyzed 378 572 isolated coronary artery bypass graft cases performed across 1083 centers, using the national Society of Thoracic Surgeons Adult Cardiac Surgery Database between 2014 and 2016. Outcomes were operative mortality, 5 postoperative complications, and composite representation of all events. We fitted models by logistic regression or extreme gradient boosting (XGBoost). For each modeling approach, we used preoperative only, intraoperative only, or pre+intraoperative variables. We developed 84 models with unique combinations of the 3 variable sets, 2 variable selection methods, 2 modeling approaches, and 7 outcomes. Each model was tested in 20 iterations of 70:30 stratified random splitting into development/testing samples. Model performances were evaluated on the testing dataset using the C statistic, area under the precision-recall curve, and calibration metrics, including the Brier score.

RESULTS

The mean patient age was 65.3 years, and 24.7% were women. Operative mortality, excluding intraoperative death, occurred in 1.9%. In all outcomes, models that considered pre+intraoperative variables demonstrated significantly improved Brier score and area under the precision-recall curve compared with models considering pre or intraoperative variables alone. XGBoost without external variable selection had the best C statistics, Brier score, and area under the precision-recall curve values in 4 of the 7 outcomes (mortality, renal failure, prolonged ventilation, and composite) compared with logistic regression models with or without variable selection. Based on the calibration plots, risk restratification for mortality showed that the logistic regression model underestimated the risk in 11 114 patients (9.8%) and overestimated in 12 005 patients (10.6%). In contrast, the XGBoost model underestimated the risk in 7218 patients (6.4%) and overestimated in 0 patients (0%).

CONCLUSIONS

In isolated coronary artery bypass graft, adding intraoperative variables to preoperative variables resulted in improved predictions of all 7 outcomes. Risk models based on XGBoost may provide a better prediction of adverse events to guide clinical care.

Application of machine learning models for predicting acute kidney injury following donation after cardiac death liver transplantation

BACKGROUND

Acute kidney injury (AKI) is a common complication after liver transplantation (LT) and is an indicator of poor prognosis. The establishment of a more accurate preoperative prediction model of AKI could help to improve the prognosis of LT. Machine learning algorithms provide a potentially effective approach.

METHODS

A total of 493 patients with donation after cardiac death LT (DCDLT) were enrolled. AKI was defined according to the clinical practice guidelines of kidney disease: improving global outcomes (KDIGO). The clinical data of patients with AKI (AKI group) and without AKI (non-AKI group) were compared. With logistic regression analysis as a conventional model, four predictive machine learning models were developed using the following algorithms: random forest, support vector machine, classical decision tree, and conditional inference tree. The predictive power of these models was then evaluated using the area under the receiver operating characteristic curve (AUC).

RESULTS

The incidence of AKI was 35.7% (176/493) during the follow-up period. Compared with the non-AKI group, the AKI group showed a remarkably lower survival rate (P < 0.001). The random forest model demonstrated the highest prediction accuracy of 0.79 with AUC of 0.850 [95% confidence interval (CI): 0.794-0.905], which was significantly higher than the AUCs of the other machine learning algorithms and logistic regression models (P < 0.001).

CONCLUSIONS

The random forest model based on machine learning algorithms for predicting AKI occurring after DCDLT demonstrated stronger predictive power than other models in our study. This suggests that machine learning methods may provide feasible tools for forecasting AKI after DCDLT.

Comparison of machine learning and logistic regression models in predicting acute kidney injury: A systematic review and meta-analysis

INTRODUCTION

We aimed to assess whether machine learning models are superior at predicting acute kidney injury (AKI) compared to logistic regression (LR), a conventional prediction model.

METHODS

Eligible studies were identified using PubMed and Embase. A total of 24 studies consisting of 84 prediction models met inclusion criteria. Independent samples t-test was performed to detect mean differences in area under the curve (AUC) between ML and LR models. One-way ANOVA and post-hoc t-tests were performed to assess mean differences in AUC between ML methods.

RESULTS

AUC data were similar between ML (0.736 ± 0.116) and LR (0.748 ± 0.057) models (p = 0.538). However, specific ML models, such as gradient boosting (0.838 ± 0.077), exhibited superior performance at predicting AKI as compared to other ML models in the literature (p < 0.05). Creatinine and urine output, standard variables assessed for AKI staging, were classified as significant predictors across multiple ML models, although the majority of significant predictors were unique and study specific.

CONCLUSIONS

These data suggest that ML models perform equally to that of LR, however ML models exhibit variable performance with some ML models displaying exceptional performance. The variability in ML prediction of AKI can be attributed, in part, to the specific ML model utilized, variable selection and processing, study and subject characteristics, and the steps associated with model training, validation, testing, and calibration.

Incidence, Predictors, and Outcomes of Acute Kidney Injury in Patients Undergoing Transcatheter Aortic Valve Replacement: Insights From the Society of Thoracic Surgeons/American College of Cardiology National Cardiovascular Data Registry-Transcatheter Valve Therapy Registry

[Figure: see text].

Artificial Intelligence in Hypertension: Seeing Through a Glass Darkly

Hypertension remains the largest modifiable cause of mortality worldwide despite the availability of effective medications and sustained research efforts over the past 100 years. Hypertension requires transformative solutions that can help reduce the global burden of the disease. Artificial intelligence and machine learning, which have made a substantial impact on our everyday lives over the last decade may be the route to this transformation. However, artificial intelligence in health care is still in its nascent stages and realizing its potential requires numerous challenges to be overcome. In this review, we provide a clinician-centric perspective on artificial intelligence and machine learning as applied to medicine and hypertension. We focus on the main roadblocks impeding implementation of this technology in clinical care and describe efforts driving potential solutions. At the juncture, there is a critical requirement for clinical and scientific expertise to work in tandem with algorithmic innovation followed by rigorous validation and scrutiny to realize the promise of artificial intelligence-enabled health care for hypertension and other chronic diseases.

Identification of variation in nutritional practice in neonatal units in England and association with clinical outcomes using agnostic machine learning

We used agnostic, unsupervised machine learning to cluster a large clinical database of information on infants admitted to neonatal units in England. Our aim was to obtain insights into nutritional practice, an area of central importance in newborn care, utilising the UK National Neonatal Research Database (NNRD). We performed clustering on time-series data of daily nutritional intakes for very preterm infants born at a gestational age less than 32 weeks (n = 45,679) over a six-year period. This revealed 46 nutritional clusters heterogeneous in size, showing common interpretable clinical practices alongside rarer approaches. Nutritional clusters with similar admission profiles revealed associations between nutritional practice, geographical location and outcomes. We show how nutritional subgroups may be regarded as distinct interventions and tested for associations with measurable outcomes. We illustrate the potential for identifying relationships between nutritional practice and outcomes with two examples, discharge weight and bronchopulmonary dysplasia (BPD). We identify the well-known effect of formula milk on greater discharge weight as well as support for the plausible, but insufficiently evidenced view that human milk is protective against BPD. Our framework highlights the potential of agnostic machine learning approaches to deliver clinical practice insights and generate hypotheses using routine data.