scholarly journals Toward Optimal Heparin Dosing by Comparing Multiple Machine Learning Methods: Retrospective Study (Preprint)

2019 ◽  
Author(s):  
Longxiang Su ◽  
Chun Liu ◽  
Dongkai Li ◽  
Jie He ◽  
Fanglan Zheng ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Data Set ◽  
Machine Learning Methods ◽  
Heparin Treatment ◽  
Neural Network Algorithm ◽  
Mimic Iii ◽  
Machine Learning Models

BACKGROUND Heparin is one of the most commonly used medications in intensive care units. In clinical practice, the use of a weight-based heparin dosing nomogram is standard practice for the treatment of thrombosis. Recently, machine learning techniques have dramatically improved the ability of computers to provide clinical decision support and have allowed for the possibility of computer generated, algorithm-based heparin dosing recommendations. OBJECTIVE The objective of this study was to predict the effects of heparin treatment using machine learning methods to optimize heparin dosing in intensive care units based on the predictions. Patient state predictions were based upon activated partial thromboplastin time in 3 different ranges: subtherapeutic, normal therapeutic, and supratherapeutic, respectively. METHODS Retrospective data from 2 intensive care unit research databases (Multiparameter Intelligent Monitoring in Intensive Care III, MIMIC-III; e–Intensive Care Unit Collaborative Research Database, eICU) were used for the analysis. Candidate machine learning models (random forest, support vector machine, adaptive boosting, extreme gradient boosting, and shallow neural network) were compared in 3 patient groups to evaluate the classification performance for predicting the subtherapeutic, normal therapeutic, and supratherapeutic patient states. The model results were evaluated using precision, recall, F1 score, and accuracy. RESULTS Data from the MIMIC-III database (n=2789 patients) and from the eICU database (n=575 patients) were used. In 3-class classification, the shallow neural network algorithm performed the best (F1 scores of 87.26%, 85.98%, and 87.55% for data set 1, 2, and 3, respectively). The shallow neural network algorithm achieved the highest F1 scores within the patient therapeutic state groups: subtherapeutic (data set 1: 79.35%; data set 2: 83.67%; data set 3: 83.33%), normal therapeutic (data set 1: 93.15%; data set 2: 87.76%; data set 3: 84.62%), and supratherapeutic (data set 1: 88.00%; data set 2: 86.54%; data set 3: 95.45%) therapeutic ranges, respectively. CONCLUSIONS The most appropriate model for predicting the effects of heparin treatment was found by comparing multiple machine learning models and can be used to further guide optimal heparin dosing. Using multicenter intensive care unit data, our study demonstrates the feasibility of predicting the outcomes of heparin treatment using data-driven methods, and thus, how machine learning–based models can be used to optimize and personalize heparin dosing to improve patient safety. Manual analysis and validation suggested that the model outperformed standard practice heparin treatment dosing.

scholarly journals Toward Optimal Heparin Dosing by Comparing Multiple Machine Learning Methods: Retrospective Study

10.2196/17648 ◽  
2020 ◽  
Vol 8 (6) ◽  
pp. e17648
Author(s):  
Longxiang Su ◽  
Chun Liu ◽  
Dongkai Li ◽  
Jie He ◽  
Fanglan Zheng ◽  
...  
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Data Set ◽  
Machine Learning Methods ◽  
Heparin Treatment ◽  
Neural Network Algorithm ◽  
Mimic Iii ◽  
Machine Learning Models

Background Heparin is one of the most commonly used medications in intensive care units. In clinical practice, the use of a weight-based heparin dosing nomogram is standard practice for the treatment of thrombosis. Recently, machine learning techniques have dramatically improved the ability of computers to provide clinical decision support and have allowed for the possibility of computer generated, algorithm-based heparin dosing recommendations. Objective The objective of this study was to predict the effects of heparin treatment using machine learning methods to optimize heparin dosing in intensive care units based on the predictions. Patient state predictions were based upon activated partial thromboplastin time in 3 different ranges: subtherapeutic, normal therapeutic, and supratherapeutic, respectively. Methods Retrospective data from 2 intensive care unit research databases (Multiparameter Intelligent Monitoring in Intensive Care III, MIMIC-III; e–Intensive Care Unit Collaborative Research Database, eICU) were used for the analysis. Candidate machine learning models (random forest, support vector machine, adaptive boosting, extreme gradient boosting, and shallow neural network) were compared in 3 patient groups to evaluate the classification performance for predicting the subtherapeutic, normal therapeutic, and supratherapeutic patient states. The model results were evaluated using precision, recall, F1 score, and accuracy. Results Data from the MIMIC-III database (n=2789 patients) and from the eICU database (n=575 patients) were used. In 3-class classification, the shallow neural network algorithm performed the best (F1 scores of 87.26%, 85.98%, and 87.55% for data set 1, 2, and 3, respectively). The shallow neural network algorithm achieved the highest F1 scores within the patient therapeutic state groups: subtherapeutic (data set 1: 79.35%; data set 2: 83.67%; data set 3: 83.33%), normal therapeutic (data set 1: 93.15%; data set 2: 87.76%; data set 3: 84.62%), and supratherapeutic (data set 1: 88.00%; data set 2: 86.54%; data set 3: 95.45%) therapeutic ranges, respectively. Conclusions The most appropriate model for predicting the effects of heparin treatment was found by comparing multiple machine learning models and can be used to further guide optimal heparin dosing. Using multicenter intensive care unit data, our study demonstrates the feasibility of predicting the outcomes of heparin treatment using data-driven methods, and thus, how machine learning–based models can be used to optimize and personalize heparin dosing to improve patient safety. Manual analysis and validation suggested that the model outperformed standard practice heparin treatment dosing.

A recurrent neural network-based predictive model for pressure ulcers based on an intensive care unit dataset (MIMIC-III) (Preprint)

2021 ◽  
Author(s):  
Minseop Park ◽  
Hyeok Choi ◽  
Hee-Sung Ahn ◽  
Hee-Ju Kang ◽  
Saehoon Kim ◽  
...  
Keyword(s):  
Neural Network ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Recurrent Neural Network ◽  
Short Term Memory ◽  
Machine Learning Algorithms ◽  
Dynamic Features ◽  
Braden Scale ◽  
Mlp Model ◽  
Mimic Iii

BACKGROUND A pressure ulcer (PU) is a localized cutaneous injury caused by pressure or shear, which usually occurs in the region of a bony prominence. PUs are common in hospitalized patients and cause complications including infection. OBJECTIVE This study aimed to build a recurrent neural network-based algorithm to predict PUs 24 hours before their occurrence. METHODS This study analyzed a freely accessible intensive care unit (ICU) dataset, MIMIC- III. Deep learning and machine learning algorithms including long short-term memory (LSTM), multilayer perceptron (MLP), and XGBoost were applied to 37 dynamic features (including the Braden scale, vital signs and laboratory results, and interventions to reduce the risk of PUs) and 35 static features (including the length of time spent in the ICU, demographics, and comorbidities). Their outcomes were compared in terms of the area under the receiver operating characteristic (AUROC) and the area under the precision-recall curve (AUPRC). RESULTS A total of 1,048 cases of PUs (10.0%) and 9,402 controls (90.0%) without PUs satisfied the inclusion criteria for analysis. The LSTM + MLP model (AUROC: 0.7929 ± 0.0095, AUPRC: 0.4819 ± 0.0109) outperformed the other models, namely: MLP model (AUROC: 0.7777 ± 0.0083, AUPRC: 0.4527 ± 0.0195) and XGBoost (AUROC: 0.7465 ± 0.0087, AUPRC: 0.4052 ± 0.0087). Various features, including the length of time spent in the ICU, Glasgow coma scale, and the Braden scale, contributed to the prediction model. CONCLUSIONS This study suggests that recurrent neural network-based algorithms such as LSTM can be applied to evaluate the risk of PUs in ICU patients.

scholarly journals Machine learning can accurately predict pre-admission baseline hemoglobin and creatinine in intensive care patients

2019 ◽  
Vol 2 (1) ◽  
Author(s):  
Antonin Dauvin ◽  
Carolina Donado ◽  
Patrik Bachtiger ◽  
Ke-Chun Huang ◽  
Christopher Martin Sauer ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Impaired Renal Function ◽  
Kidney Injury ◽  
Learning Models ◽  
Intensive Care Patients ◽  
Using Data ◽  
Machine Learning Models ◽  
Baseline Hemoglobin

AbstractPatients admitted to the intensive care unit frequently have anemia and impaired renal function, but often lack historical blood results to contextualize the acuteness of these findings. Using data available within two hours of ICU admission, we developed machine learning models that accurately (AUC 0.86–0.89) classify an individual patient’s baseline hemoglobin and creatinine levels. Compared to assuming the baseline to be the same as the admission lab value, machine learning performed significantly better at classifying acute kidney injury regardless of initial creatinine value, and significantly better at predicting baseline hemoglobin value in patients with admission hemoglobin of <10 g/dl.

scholarly journals Machine Learning Methods Applied to Predict Ventilator-Associated Pneumonia with Pseudomonas aeruginosa Infection via Sensor Array of Electronic Nose in Intensive Care Unit

Sensors ◽  
2019 ◽  
Vol 19 (8) ◽  
pp. 1866 ◽  
Author(s):  
Liao ◽  
Wang ◽  
Zhang ◽  
Abbod ◽  
Shih ◽  
...  
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Pseudomonas Aeruginosa ◽  
Intensive Care ◽  
Prediction Models ◽  
Support Vector ◽  
Ventilator Associated Pneumonia ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Pseudomonas Aeruginosa Infection

One concern to the patients is the off-line detection of pneumonia infection status after using the ventilator in the intensive care unit. Hence, machine learning methods for ventilator-associated pneumonia (VAP) rapid diagnose are proposed. A popular device, Cyranose 320 e-nose, is usually used in research on lung disease, which is a highly integrated system and sensor comprising 32 array using polymer and carbon black materials. In this study, a total of 24 subjects were involved, including 12 subjects who are infected with pneumonia, and the rest are non-infected. Three layers of back propagation artificial neural network and support vector machine (SVM) methods were applied to patients’ data to predict whether they are infected with VAP with Pseudomonas aeruginosa infection. Furthermore, in order to improve the accuracy and the generalization of the prediction models, the ensemble neural networks (ENN) method was applied. In this study, ENN and SVM prediction models were trained and tested. In order to evaluate the models’ performance, a fivefold cross-validation method was applied. The results showed that both ENN and SVM models have high recognition rates of VAP with Pseudomonas aeruginosa infection, with 0.9479 ± 0.0135 and 0.8686 ± 0.0422 accuracies, 0.9714 ± 0.0131, 0.9250 ± 0.0423 sensitivities, and 0.9288 ± 0.0306, 0.8639 ± 0.0276 positive predictive values, respectively. The ENN model showed better performance compared to SVM in the recognition of VAP with Pseudomonas aeruginosa infection. The areas under the receiver operating characteristic curve of the two models were 0.9842 ± 0.0058 and 0.9410 ± 0.0301, respectively, showing that both models are very stable and accurate classifiers. This study aims to assist the physician in providing a scientific and effective reference for performing early detection in Pseudomonas aeruginosa infection or other diseases.

scholarly journals Novel Method of Atrial Fibrillation Case Identification and Burden Estimation Using the MIMIC-III Electronic Health Data Set

2019 ◽  
Vol 34 (10) ◽  
pp. 851-857 ◽  
Author(s):  
Eric Y. Ding ◽  
Daniella Albuquerque ◽  
Michael Winter ◽  
Sophia Binici ◽  
Jaclyn Piche ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Gold Standard ◽  
Onset Time ◽  
Data Set ◽  
Case Identification ◽  
Novel Method ◽  
Electronic Health ◽  
Mimic Iii

Background: Atrial fibrillation (AF) portends poor prognoses in intensive care unit patients with sepsis. However, AF research is challenging: Previous studies demonstrate that International Classification of Disease ( ICD) codes may underestimate the incidence of AF, but chart review is expensive and often not feasible. We aim to examine the accuracy of nurse-charted AF and its temporal precision in critical care patients with sepsis. Methods: Patients with sepsis with continuous electrocardiogram (ECG) waveforms were identified from the Medical Information Mart for Intensive Care (MIMIC-III) database, a de-identified, single-center intensive care unit electronic health record (EHR) source. We selected a random sample of ECGs of 6 to 50 hours’ duration for manual review. Nurse-charted AF occurrence and onset time and ICD-9-coded AF were compared to gold-standard ECG adjudication by a board-certified cardiac electrophysiologist blinded to AF status. Descriptive statistics were calculated for all variables in patients diagnosed with AF by nurse charting, ICD-9 code, or both. Results: From 142 ECG waveforms (58 AF and 84 sinus rhythm), nurse charting identified AF events with 93% sensitivity (95% confidence interval [CI]: 87%-100%) and 87% specificity (95% CI: 80%-94%) compared to the gold standard manual ECG review. Furthermore, nurse-charted AF onset time was within 1 hour of expert reader onset time for 85% of the reviewed tracings. The ICD-9 codes were 97% sensitive (95% CI: 88-100%) and 82% specific (95% CI: 74-90%) for incident AF during admission but unable to identify AF time of onset. Conclusion: Nurse documentation of AF in EHR is accurate and has high precision for determining AF onset to within 1 hour. Our study suggests that nurse-charted AF in the EHR represents a potentially novel method for AF case identification, timing, and burden estimation.

scholarly journals Cyberattacks Detection and Analysis in a Network Log System Using XGBoost with ELK Stack

2021 ◽  
Author(s):  
Chao-Tung Yang ◽  
Yu-Wei Chan ◽  
Jung-Chun Liu Liu ◽  
Endah Kristiani ◽  
Cing-Han Lai
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Abnormal Behavior ◽  
Attack Behavior ◽  
Learning Methods ◽  
Data Set ◽  
Log Data ◽  
Network Attack ◽  
Machine Learning Methods ◽  
Gradient Enhancement

Abstract The usage of artificial intelligence and machine learning methods on cyberattacks increasing significantly recently. For the defense method of cyberattacks, it is possible to detect and identify the attack event by observing the log data and analyzing whether it has abnormal behavior or not. This paper implemented the ELK Stack network log system (NetFlow Log) to visually analyze log data and present several network attack behavior characteristics for further analysis. Additionally, this system evaluated the extreme gradient enhancement (XGBoost), Recurrent Neural Network (RNN), and Deep Neural Network (DNN) model for machine learning methods. Keras was used as a deep learning framework for building a model to detect the attack event. From the experiments, it can be confirmed that the XGBoost model has an accuracy rate of 96.01% for potential threats. The full attack data set can achieve 96.26% accuracy, which is better than RNN and DNN models.

scholarly journals Accurate and reproducible prediction of ICU readmissions

2019 ◽  
Author(s):  
Dinh-Phong Nguyen ◽  
Nicolas Paris ◽  
Adrien Parrot
Keyword(s):  
Machine Learning ◽  
Intensive Care Unit ◽  
Intensive Care ◽  
Clinical Outcomes ◽  
Scoring Methods ◽  
Time Points ◽  
Boosting Method ◽  
Mimic Iii ◽  
The Way

Readmission in the intensive care unit (ICU) is associated with poor clinical outcomes and high costs. Traditional scoring methods to help clinicians deciding whether a patient is ready for discharge have failed to meet expectations, paving the way for machine learning based approaches. Freely available datasets such as MIMIC-III have served as benchmarking media to compare such tools. We used the OMOP-CDM version of MIMIC-III (MIMIC-OMOP) to train and evaluate a lightweight tree boosting method to predict readmission in ICU at different time points after discharge (3, 7, 15 and 30 days), outperforming existing solutions with an AUROC of 0.802 (SD=0.011) and a recall of 0.837 (SD=0.016) for 3-days readmission.

scholarly journals Using Item Response Theory for Explainable Machine Learning in Predicting Mortality in the Intensive Care Unit: Case-Based Approach

10.2196/20268 ◽  
2020 ◽  
Vol 22 (9) ◽  
pp. e20268
Author(s):  
Adrienne Kline ◽  
Theresa Kline ◽  
Zahra Shakeri Hossein Abad ◽  
Joon Lee
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Intensive Care Unit ◽  
Logistic Regression ◽  
Intensive Care ◽  
Item Response ◽  
Nearest Neighbors ◽  
K Nearest Neighbors ◽  
Linear Discriminant ◽  
Case Based

Background Supervised machine learning (ML) is being featured in the health care literature with study results frequently reported using metrics such as accuracy, sensitivity, specificity, recall, or F1 score. Although each metric provides a different perspective on the performance, they remain to be overall measures for the whole sample, discounting the uniqueness of each case or patient. Intuitively, we know that all cases are not equal, but the present evaluative approaches do not take case difficulty into account. Objective A more case-based, comprehensive approach is warranted to assess supervised ML outcomes and forms the rationale for this study. This study aims to demonstrate how the item response theory (IRT) can be used to stratify the data based on how difficult each case is to classify, independent of the outcome measure of interest (eg, accuracy). This stratification allows the evaluation of ML classifiers to take the form of a distribution rather than a single scalar value. Methods Two large, public intensive care unit data sets, Medical Information Mart for Intensive Care III and electronic intensive care unit, were used to showcase this method in predicting mortality. For each data set, a balanced sample (n=8078 and n=21,940, respectively) and an imbalanced sample (n=12,117 and n=32,910, respectively) were drawn. A 2-parameter logistic model was used to provide scores for each case. Several ML algorithms were used in the demonstration to classify cases based on their health-related features: logistic regression, linear discriminant analysis, K-nearest neighbors, decision tree, naive Bayes, and a neural network. Generalized linear mixed model analyses were used to assess the effects of case difficulty strata, ML algorithm, and the interaction between them in predicting accuracy. Results The results showed significant effects (P<.001) for case difficulty strata, ML algorithm, and their interaction in predicting accuracy and illustrated that all classifiers performed better with easier-to-classify cases and that overall the neural network performed best. Significant interactions suggest that cases that fall in the most arduous strata should be handled by logistic regression, linear discriminant analysis, decision tree, or neural network but not by naive Bayes or K-nearest neighbors. Conventional metrics for ML classification have been reported for methodological comparison. Conclusions This demonstration shows that using the IRT is a viable method for understanding the data that are provided to ML algorithms, independent of outcome measures, and highlights how well classifiers differentiate cases of varying difficulty. This method explains which features are indicative of healthy states and why. It enables end users to tailor the classifier that is appropriate to the difficulty level of the patient for personalized medicine.

Sign in / Sign up

Export Citation Format

Share Document