AKI Prediction Models in ICU: A Comparative Study (Preprint)

2020 ◽  
Author(s):  
Qing Qian ◽  
Haixia Sun ◽  
Jinming Wu ◽  
Jiayang Wang ◽  
Lei Yang
Keyword(s):  
Kidney Disease ◽  
Comparative Study ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Gradient Boosting ◽  
Support Vector ◽  
Early Prediction ◽  
Icu Admission ◽  
Extreme Gradient Boosting ◽  
Mimic Iii

BACKGROUND Acute kidney injury (AKI) is highly prevalent in critically ill patients and associated with significant morbidity and mortality as well as high financial costs. Early prediction of AKI provides an opportunity to develop strategies for early diagnosis, effective prevention, and timely treatment. Machine learning models have been developed for early prediction of AKI on critically ill patients by different researchers under different scenario. OBJECTIVE This comparative study aims to assess the performances of existing models for early prediction of AKI in the Intensive Care Unit (ICU) setting. METHODS The data was collected from the MIMIC-III database for all patients above 18 years old who had valid creatinine measured for 72 hours following ICU admission. Those with existing condition of kidney disease on admission were excluded. 17 predictor variables including patient demographics and physiological indicators were selected on the basis of the Kidney Disease Improving Global Outcomes (KDIGO) and medical literatures. Six models from three different types of methods were tested including Logistic Regression (LR), Support Vector Machines (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Decision (LightGBM), and Convolutional Neural Network (CNN). The area under ROC curve (AUC), accuracy, precision, recall and F1 value were calculated for each model to evaluate the performance. RESULTS We extracted 17205patient ICU records from MIMIC-III dataset. LightGBM had the best performance, with all the evaluation indicators achieved the highest (with average AUC 0.905, F1 0.897, Recall 0.836, P<.001). XGBoost had the second best performance (P<.001) and LR, RF, SVM performed similarly (P=0.082, 0.158, 0.710) on AUC. CNN got the lowest score on accuracy, precision, F1 and AUC. SVM and LR had relatively low recall than others. Creatinine were found to have the most significant effect on the early prediction of AKI onset in LR, RF, SVM and LightGBM. CONCLUSIONS LightGBM demonstrated the best predictive capability in predicting AKI present at the first 72 hours of ICU admission. LightGBM and XGBoost showed great potential for clinical application owing to their high recall. This study can provide references for AI-powered clinical decision support system for early AKI prediction in ICU setting.

scholarly journals Comparison of Approaches for Prediction of Renal Replacement Therapy-Free Survival in Patients with Acute Kidney Injury

2021 ◽  
pp. 1-7
Author(s):  
Pattharawin Pattharanitima ◽  
Akhil Vaid ◽  
Suraj K. Jaladanki ◽  
Ishan Paranjpe ◽  
Ross O’Hagan ◽  
...  
Keyword(s):  
Acute Kidney Injury ◽  
Renal Replacement Therapy ◽  
Replacement Therapy ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Kidney Injury ◽  
Support Vector ◽  
Adaptive Boosting ◽  
Free Survival ◽  
Extreme Gradient Boosting

Background/Aims: Acute kidney injury (AKI) in critically ill patients is common, and continuous renal replacement therapy (CRRT) is a preferred mode of renal replacement therapy (RRT) in hemodynamically unstable patients. Prediction of clinical outcomes in patients on CRRT is challenging. We utilized several approaches to predict RRT-free survival (RRTFS) in critically ill patients with AKI requiring CRRT. Methods: We used the Medical Information Mart for Intensive Care (MIMIC-III) database to identify patients ≥18 years old with AKI on CRRT, after excluding patients who had ESRD on chronic dialysis, and kidney transplantation. We defined RRTFS as patients who were discharged alive and did not require RRT ≥7 days prior to hospital discharge. We utilized all available biomedical data up to CRRT initiation. We evaluated 7 approaches, including logistic regression (LR), random forest (RF), support vector machine (SVM), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), multilayer perceptron (MLP), and MLP with long short-term memory (MLP + LSTM). We evaluated model performance by using area under the receiver operating characteristic (AUROC) curves. Results: Out of 684 patients with AKI on CRRT, 205 (30%) patients had RRTFS. The median age of patients was 63 years and their median Simplified Acute Physiology Score (SAPS) II was 67 (interquartile range 52–84). The MLP + LSTM showed the highest AUROC (95% CI) of 0.70 (0.67–0.73), followed by MLP 0.59 (0.54–0.64), LR 0.57 (0.52–0.62), SVM 0.51 (0.46–0.56), AdaBoost 0.51 (0.46–0.55), RF 0.44 (0.39–0.48), and XGBoost 0.43 (CI 0.38–0.47). Conclusions: A MLP + LSTM model outperformed other approaches for predicting RRTFS. Performance could be further improved by incorporating other data types.

scholarly journals KDClassifier: A urinary proteomic spectra analysis tool based on machine learning for the classification of kidney diseases

2021 ◽  
Vol 3 (3) ◽  
pp. 63-72
Author(s):  
Wanjun Zhao ◽  
Keyword(s):  
Kidney Disease ◽  
Kidney Diseases ◽  
Confusion Matrix ◽  
Gradient Boosting ◽  
Support Vector ◽  
Diagnostic Model ◽  
Analysis Tool ◽  
Data Set ◽  
Extreme Gradient Boosting

Background: We aimed to establish a novel diagnostic model for kidney diseases by combining artificial intelligence with complete mass spectrum information from urinary proteomics. Methods: We enrolled 134 patients (IgA nephropathy, membranous nephropathy, and diabetic kidney disease) and 68 healthy participants as controls, with a total of 610,102 mass spectra from their urinary proteomic profiles. The training data set (80%) was used to create a diagnostic model using XGBoost, random forest (RF), a support vector machine (SVM), and artificial neural networks (ANNs). The diagnostic accuracy was evaluated using a confusion matrix with a test dataset (20%). We also constructed receiver operating-characteristic, Lorenz, and gain curves to evaluate the diagnostic model. Results: Compared with the RF, SVM, and ANNs, the modified XGBoost model, called Kidney Disease Classifier (KDClassifier), showed the best performance. The accuracy of the XGBoost diagnostic model was 96.03%. The area under the curve of the extreme gradient boosting (XGBoost) model was 0.952 (95% confidence interval, 0.9307–0.9733). The Kolmogorov-Smirnov (KS) value of the Lorenz curve was 0.8514. The Lorenz and gain curves showed the strong robustness of the developed model. Conclusions: The KDClassifier achieved high accuracy and robustness and thus provides a potential tool for the classification of kidney diseases

scholarly journals Predictive Analytics of Chronic Kidney Disease using Machine Learning Algorithm

2019 ◽  
Vol 8 (2) ◽  
pp. 940-947
Keyword(s):  
Chronic Kidney Disease ◽  
Kidney Disease ◽  
Predictive Analytics ◽  
Research Work ◽  
Classification And Regression Tree ◽  
Gradient Boosting ◽  
Support Vector ◽  
Linear Discriminant ◽  
Extreme Gradient Boosting ◽  
Data Points

According to the health statistics of India on Chronic Kidney Disease (CKD) a total of 63538 cases has been registered. Average age of men and women prone to kidney disease lies in the range of 48 to 70 years. CKD is more prevalent among male than among female. India ranks 17th position in CKD during 2015[1]. This paper focus on the predictive analytics architecture to analyse CKD dataset using feature engineering and classification algorithm. The proposed model incorporates techniques to validate the feasibility of the data points used for analysis. The main focus of this research work is to analyze the dataset of chronic kidney failure and perform the classification of CKD and Non CKD cases. The feasibility of the proposed dataset is determined through the Learning curve performance. The features which play a vital role in classification are determined using sequential forward selection algorithm. The training dataset with the selected features is fed into various classifier to determine which classifier plays a vital and accurate role in detection of CKD. The proposed dataset is classified using various Classification algorithms like Linear Regression(LR), Linear Discriminant Analysis(LDA), K-Nearest Neighbour(KNN), Classification and Regression Tree(CART), Naive Bayes(NB), Support Vector Machine(SVM), Random Forest(RF), eXtreme Gradient Boosting(XGBoost) and Ada Boost Regressor (ABR). It was found that for the given CKD dataset with 25 attributes of 11 Numeric and 14 Nominal the following classifier like LR, LDA, CART,NB,RF,XGB and ABR provides an accuracy ranging from 98% to 100% . The proposed architecture validates the dataset against the thumb rule when working with less number of data points used for classification and the classifier is validated against under fit, over fit conditions. The performance of the classifier is evaluated using accuracy and F-Score. The proposed architecture indicates that LR, RF and ABR provides a very high accuracy and F-Score

scholarly journals Performance Analysis of Conventional Machine Learning Algorithms for Identification of Chronic Kidney Disease in Type 1 Diabetes Mellitus Patients

Diagnostics ◽  
2021 ◽  
Vol 11 (12) ◽  
pp. 2267
Author(s):  
Nakib Hayat Chowdhury ◽  
Mamun Bin Ibne Reaz ◽  
Fahmida Haque ◽  
Shamim Ahmad ◽  
Sawal Hamid Md Ali ◽  
...  
Keyword(s):  
Diabetes Mellitus ◽  
Machine Learning ◽  
Chronic Kidney Disease ◽  
Type 1 Diabetes ◽  
Kidney Disease ◽  
Type 1 Diabetes Mellitus ◽  
Gradient Boosting ◽  
Support Vector ◽  
Extreme Gradient Boosting

Chronic kidney disease (CKD) is one of the severe side effects of type 1 diabetes mellitus (T1DM). However, the detection and diagnosis of CKD are often delayed because of its asymptomatic nature. In addition, patients often tend to bypass the traditional urine protein (urinary albumin)-based CKD detection test. Even though disease detection using machine learning (ML) is a well-established field of study, it is rarely used to diagnose CKD in T1DM patients. This research aimed to employ and evaluate several ML algorithms to develop models to quickly predict CKD in patients with T1DM using easily available routine checkup data. This study analyzed 16 years of data of 1375 T1DM patients, obtained from the Epidemiology of Diabetes Interventions and Complications (EDIC) clinical trials directed by the National Institute of Diabetes, Digestive, and Kidney Diseases, USA. Three data imputation techniques (RF, KNN, and MICE) and the SMOTETomek resampling technique were used to preprocess the primary dataset. Ten ML algorithms including logistic regression (LR), k-nearest neighbor (KNN), Gaussian naïve Bayes (GNB), support vector machine (SVM), stochastic gradient descent (SGD), decision tree (DT), gradient boosting (GB), random forest (RF), extreme gradient boosting (XGB), and light gradient-boosted machine (LightGBM) were applied to developed prediction models. Each model included 19 demographic, medical history, behavioral, and biochemical features, and every feature’s effect was ranked using three feature ranking techniques (XGB, RF, and Extra Tree). Lastly, each model’s ROC, sensitivity (recall), specificity, accuracy, precision, and F-1 score were estimated to find the best-performing model. The RF classifier model exhibited the best performance with 0.96 (±0.01) accuracy, 0.98 (±0.01) sensitivity, and 0.93 (±0.02) specificity. LightGBM performed second best and was quite close to RF with 0.95 (±0.06) accuracy. In addition to these two models, KNN, SVM, DT, GB, and XGB models also achieved more than 90% accuracy.

A Comparative Study of Machine Learning Techniques for Predicting Sepsis for MIMIC-III Patients

2021 ◽  
Author(s):  
Xuze Zhao ◽  
Bo Qu
Keyword(s):  
Machine Learning ◽  
Intensive Care ◽  
Intensive Care Units ◽  
Medical Information ◽  
Machine Learning Techniques ◽  
Gradient Boosting ◽  
Support Vector ◽  
Mortality And Morbidity ◽  
Extreme Gradient Boosting ◽  
Mimic Iii

Abstract Background: Sepsis is one of the dominating causes of mortality and morbidity in-hospital especially in intensive care units (ICU) patients. Therefore, a reliable decision-making model for predicting sepsis is of great importance. The purpose of this study was to develop an eXtreme Gradient Boosting (XGBoost) based model and explore whether it performs better in predicting sepsis from the time of admission in intensive care units (ICU) than other machine learning (ML) methods. Methods: The source data used for model establishment in this study were from a retrospective medical information mart for intensive care (MIMIC) III dataset, restricted to intensive care units (ICUs) patients aged between 18 and 89. Model performance of the XGBoost model was compared to logistic regression (LR), recursive neural network (RNN), and support vector machine (SVM). Then, the performances of the models were evaluated and compared by the area under the curve (AUC) of the receiver operating characteristic (ROC) curves.Results: A total of 6430 MIMIC-III cases are included in this article, in which, 3021 cases have encountered sepsis while 3409 cases have not, respectively. As for the AUC (0.808 (95% CI): 0.767-0.848,DT), 0.802 (95%CI: 0.762-0.842，RNN), 0.790 (95%CI: 0.751-0.830,SVM), 0.775 (95%CI: 0.736-0.813,LR) , results of the models, XGBoost performs best in predicting sepsis.Conclusions: By using the DT algorithm, a more accurate prediction model can be established. Amongst other ML methods, the XGBoost model demonstrated the best ability in detecting the sepsis of the patients in ICU.

scholarly journals Effect of timing of intubation on clinical outcomes of critically ill patients with COVID-19: a systematic review and meta-analysis of non-randomized cohort studies

Critical Care ◽  
2021 ◽  
Vol 25 (1) ◽  
Author(s):  
Eleni Papoutsi ◽  
Vassilis G. Giannakoulis ◽  
Eleni Xourgia ◽  
Christina Routsi ◽  
Anastasia Kotanidou ◽  
...  
Keyword(s):  
Systematic Review ◽  
Mechanical Ventilation ◽  
Clinical Outcomes ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Meta Analysis ◽  
Mortality And Morbidity ◽  
Icu Admission ◽  
All Cause Mortality ◽  
Detectable Difference

Abstract Background Although several international guidelines recommend early over late intubation of patients with severe coronavirus disease 2019 (COVID-19), this issue is still controversial. We aimed to investigate the effect (if any) of timing of intubation on clinical outcomes of critically ill patients with COVID-19 by carrying out a systematic review and meta-analysis. Methods PubMed and Scopus were systematically searched, while references and preprint servers were explored, for relevant articles up to December 26, 2020, to identify studies which reported on mortality and/or morbidity of patients with COVID-19 undergoing early versus late intubation. “Early” was defined as intubation within 24 h from intensive care unit (ICU) admission, while “late” as intubation at any time after 24 h of ICU admission. All-cause mortality and duration of mechanical ventilation (MV) were the primary outcomes of the meta-analysis. Pooled risk ratio (RR), pooled mean difference (MD) and 95% confidence intervals (CI) were calculated using a random effects model. The meta-analysis was registered with PROSPERO (CRD42020222147). Results A total of 12 studies, involving 8944 critically ill patients with COVID-19, were included. There was no statistically detectable difference on all-cause mortality between patients undergoing early versus late intubation (3981 deaths; 45.4% versus 39.1%; RR 1.07, 95% CI 0.99–1.15, p = 0.08). This was also the case for duration of MV (1892 patients; MD − 0.58 days, 95% CI − 3.06 to 1.89 days, p = 0.65). In a sensitivity analysis using an alternate definition of early/late intubation, intubation without versus with a prior trial of high-flow nasal cannula or noninvasive mechanical ventilation was still not associated with a statistically detectable difference on all-cause mortality (1128 deaths; 48.9% versus 42.5%; RR 1.11, 95% CI 0.99–1.25, p = 0.08). Conclusions The synthesized evidence suggests that timing of intubation may have no effect on mortality and morbidity of critically ill patients with COVID-19. These results might justify a wait-and-see approach, which may lead to fewer intubations. Relevant guidelines may therefore need to be updated.

scholarly journals Prognostic impact of elevated lactate levels on mortality in critically ill patients with and without preadmission metformin treatment: a Danish registry-based cohort study

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Rene A. Posma ◽  
Trine Frøslev ◽  
Bente Jespersen ◽  
Iwan C. C. van der Horst ◽  
Daan J. Touw ◽  
...  
Keyword(s):  
Cohort Study ◽  
Mortality Rate ◽  
Critically Ill ◽  
Critically Ill Patients ◽  
Cox Regression ◽  
Lower Mortality ◽  
Prognostic Impact ◽  
Icu Admission ◽  
Lactate Levels ◽  
Restricted Cubic Splines

Abstract Background Lactate is a robust prognostic marker for the outcome of critically ill patients. Several small studies reported that metformin users have higher lactate levels at ICU admission without a concomitant increase in mortality. However, this has not been investigated in a larger cohort. We aimed to determine whether the association between lactate levels around ICU admission and mortality is different in metformin users compared to metformin nonusers. Methods This cohort study included patients admitted to ICUs in northern Denmark between January 2010 and August 2017 with any circulating lactate measured around ICU admission, which was defined as 12 h before until 6 h after admission. The association between the mean of the lactate levels measured during this period and 30-day mortality was determined for metformin users and nonusers by modelling restricted cubic splines obtained from a Cox regression model. Results Of 37,293 included patients, 3183 (9%) used metformin. The median (interquartile range) lactate level was 1.8 (1.2–3.2) in metformin users and 1.6 (1.0–2.7) mmol/L in metformin nonusers. Lactate levels were strongly associated with mortality for both metformin users and nonusers. However, the association of lactate with mortality was different for metformin users, with a lower mortality rate in metformin users than in nonusers when admitted with similar lactate levels. This was observed over the whole range of lactate levels, and consequently, the relation of lactate with mortality was shifted rightwards for metformin users. Conclusion In this large observational cohort of critically ill patients, early lactate levels were strongly associated with mortality. Irrespective of the degree of hyperlactataemia, similar lactate levels were associated with a lower mortality rate in metformin users compared with metformin nonusers. Therefore, lactate levels around ICU admission should be interpreted according to metformin use.

scholarly journals Machine learning models to identify low adherence to influenza vaccination among Korean adults with cardiovascular disease

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Moojung Kim ◽  
Young Jae Kim ◽  
Sung Jin Park ◽  
Kwang Gi Kim ◽  
Pyung Chun Oh ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cardiovascular Disease ◽  
Influenza Vaccination ◽  
Machine Learning Techniques ◽  
Gradient Boosting ◽  
Support Vector ◽  
Age Group ◽  
Learning Models ◽  
Extreme Gradient Boosting ◽  
Machine Learning Models

Abstract Background Annual influenza vaccination is an important public health measure to prevent influenza infections and is strongly recommended for cardiovascular disease (CVD) patients, especially in the current coronavirus disease 2019 (COVID-19) pandemic. The aim of this study is to develop a machine learning model to identify Korean adult CVD patients with low adherence to influenza vaccination Methods Adults with CVD (n = 815) from a nationally representative dataset of the Fifth Korea National Health and Nutrition Examination Survey (KNHANES V) were analyzed. Among these adults, 500 (61.4%) had answered "yes" to whether they had received seasonal influenza vaccinations in the past 12 months. The classification process was performed using the logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGB) machine learning techniques. Because the Ministry of Health and Welfare in Korea offers free influenza immunization for the elderly, separate models were developed for the < 65 and ≥ 65 age groups. Results The accuracy of machine learning models using 16 variables as predictors of low influenza vaccination adherence was compared; for the ≥ 65 age group, XGB (84.7%) and RF (84.7%) have the best accuracies, followed by LR (82.7%) and SVM (77.6%). For the < 65 age group, SVM has the best accuracy (68.4%), followed by RF (64.9%), LR (63.2%), and XGB (61.4%). Conclusions The machine leaning models show comparable performance in classifying adult CVD patients with low adherence to influenza vaccination.

scholarly journals Explainable machine learning can outperform Cox regression predictions and provide insights in breast cancer survival

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Arturo Moncada-Torres ◽  
Marissa C. van Maaren ◽  
Mathijs P. Hendriks ◽  
Sabine Siesling ◽  
Gijs Geleijnse
Keyword(s):  
Breast Cancer ◽  
Machine Learning ◽  
Explicit Knowledge ◽  
Cox Regression ◽  
Metastatic Breast ◽  
Gradient Boosting ◽  
Support Vector ◽  
Netherlands Cancer Registry ◽  
Extreme Gradient Boosting ◽  
The Impact

AbstractCox Proportional Hazards (CPH) analysis is the standard for survival analysis in oncology. Recently, several machine learning (ML) techniques have been adapted for this task. Although they have shown to yield results at least as good as classical methods, they are often disregarded because of their lack of transparency and little to no explainability, which are key for their adoption in clinical settings. In this paper, we used data from the Netherlands Cancer Registry of 36,658 non-metastatic breast cancer patients to compare the performance of CPH with ML techniques (Random Survival Forests, Survival Support Vector Machines, and Extreme Gradient Boosting [XGB]) in predicting survival using the $$c$$ c -index. We demonstrated that in our dataset, ML-based models can perform at least as good as the classical CPH regression ($$c$$ c -index $$\sim \,0.63$$ ∼ 0.63 ), and in the case of XGB even better ($$c$$ c -index $$\sim 0.73$$ ∼ 0.73 ). Furthermore, we used Shapley Additive Explanation (SHAP) values to explain the models’ predictions. We concluded that the difference in performance can be attributed to XGB’s ability to model nonlinearities and complex interactions. We also investigated the impact of specific features on the models’ predictions as well as their corresponding insights. Lastly, we showed that explainable ML can generate explicit knowledge of how models make their predictions, which is crucial in increasing the trust and adoption of innovative ML techniques in oncology and healthcare overall.

Sign in / Sign up

Export Citation Format

Share Document