Dataset Information

Validation of a Machine Learning Model to Predict Childhood Lead Poisoning.

ABSTRACT: Importance:Childhood lead poisoning causes irreversible neurobehavioral deficits, but current practice is secondary prevention. Objective:To validate a machine learning (random forest) prediction model of elevated blood lead levels (EBLLs) by comparison with a parsimonious logistic regression. Design, Setting, and Participants:This prognostic study for temporal validation of multivariable prediction models used data from the Women, Infants, and Children (WIC) program of the Chicago Department of Public Health. Participants included a development cohort of children born from January 1, 2007, to December 31, 2012, and a validation WIC cohort born from January 1 to December 31, 2013. Blood lead levels were measured until December 31, 2018. Data were analyzed from January 1 to October 31, 2019. Exposures:Blood lead level test results; lead investigation findings; housing characteristics, permits, and violations; and demographic variables. Main Outcomes and Measures:Incident EBLL (?6 ?g/dL). Models were assessed using the area under the receiver operating characteristic curve (AUC) and confusion matrix metrics (positive predictive value, sensitivity, and specificity) at various thresholds. Results:Among 6812 children in the WIC validation cohort, 3451 (50.7%) were female, 3057 (44.9%) were Hispanic, 2804 (41.2%) were non-Hispanic Black, 458 (6.7%) were non-Hispanic White, and 442 (6.5%) were Asian (mean [SD] age, 5.5 [0.3] years). The median year of housing construction was 1919 (interquartile range, 1903-1948). Random forest AUC was 0.69 compared with 0.64 for logistic regression (difference, 0.05; 95% CI, 0.02-0.08). When predicting the 5% of children at highest risk to have EBLLs, random forest and logistic regression models had positive predictive values of 15.5% and 7.8%, respectively (difference, 7.7%; 95% CI, 3.7%-11.3%), sensitivity of 16.2% and 8.1%, respectively (difference, 8.1%; 95% CI, 3.9%-11.7%), and specificity of 95.5% and 95.1% (difference, 0.4%; 95% CI, 0.0%-0.7%). Conclusions and Relevance:The machine learning model outperformed regression in predicting childhood lead poisoning, especially in identifying children at highest risk. Such a model could be used to target the allocation of lead poisoning prevention resources to these children.

SUBMITTER: Potash E

PROVIDER: S-EPMC7495240 | biostudies-literature | 2020 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Validation of a Machine Learning Model to Predict Childhood Lead Poisoning.

Potash Eric E Ghani Rayid R Walsh Joe J Jorgensen Emile E Lohff Cortland C Prachand Nik N Mansour Raed R

JAMA network open 20200901 9

<h4>Importance</h4>Childhood lead poisoning causes irreversible neurobehavioral deficits, but current practice is secondary prevention.<h4>Objective</h4>To validate a machine learning (random forest) prediction model of elevated blood lead levels (EBLLs) by comparison with a parsimonious logistic regression.<h4>Design, setting, and participants</h4>This prognostic study for temporal validation of multivariable prediction models used data from the Women, Infants, and Children (WIC) program of the ...[more]

PMID: 32936296

Similar Datasets

Project description:ImportanceMachine-learning algorithms offer better predictive accuracy than traditional prognostic models but are too complex and opaque for clinical use.ObjectiveTo compare different machine learning methods in predicting overall mortality in cirrhosis and to use machine learning to select easily scored clinical variables for a novel cirrhosis prognostic model.Design, setting, and participantsThis prognostic study used a retrospective cohort of adult patients with cirrhosis or its complications seen in 130 hospitals and affiliated ambulatory clinics in the integrated, national Veterans Affairs health care system from October 1, 2011, to September 30, 2015. Patients were followed up through December 31, 2018. Data were analyzed from October 1, 2017, to May 31, 2020.ExposuresPotential predictors included demographic characteristics; liver disease etiology, severity, and complications; use of health care resources; comorbid conditions; and comprehensive laboratory and medication data. Patients were randomly selected for model development (66.7%) and validation (33.3%). Three different statistical and machine learning methods were evaluated: gradient descent boosting, logistic regression with least absolute shrinkage and selection operator (LASSO) regularization, and logistic regression with LASSO constrained to select no more than 10 predictors (partial pathway model). Predictor inclusion and model performance were evaluated in a 5-fold cross-validation. Last, the predictors identified in the most parsimonious (the partial path) model were refit using maximum-likelihood estimation (Cirrhosis Mortality Model [CiMM]), and its predictive performance was compared with that of the widely used Model for End Stage Liver Disease with sodium (MELD-Na) score.Main outcomes and measuresAll-cause mortality.ResultsOf the 107 939 patients with cirrhosis (mean [SD] age, 62.7 [9.6] years; 96.6% male; 66.3% white, 18.4% African American), the annual mortality rate ranged from 8.8% to 15.3%. In total, 32.7% of patients died within 3 years, and 46.2% died within 5 years after the index date. Models predicting 1-year mortality had good discrimination for the gradient descent boosting (area under the receiver operating characteristics curve [AUC], 0.81; 95% CI, 0.80-0.82), logistic regression with LASSO regularization (AUC, 0.78; 95% CI, 0.77-0.79), and the partial path logistic model (AUC, 0.78; 95% CI, 0.76-0.78). All models showed good calibration. The final CiMM model with machine learning-derived clinical variables offered significantly better discrimination than the MELD-Na score, with AUCs of 0.78 (95% CI, 0.77-0.79) vs 0.67 (95% CI, 0.66-0.68) for 1-year mortality, respectively (DeLong z = 17.00; P < .001).Conclusions and relevanceIn this study, simple machine learning techniques performed as well as the more advanced ensemble gradient boosting. Using the clinical variables identified from simple machine learning in a cirrhosis mortality model produced a new score more transparent than machine learning and more predictive than the MELD-Na score.

Project description:BackgroundEarly prediction model of hemodynamic instability has the potential to improve the critical care, whereas limited external validation on the generalizability. We aimed to independently validate the Hemodynamic Stability Index (HSI), a multi-parameter machine learning model, in predicting hemodynamic instability in Asian patients.MethodHemodynamic instability was marked by using inotropic, vasopressor, significant fluid therapy, and/or blood transfusions. This retrospective study included among 15,967 ICU patients who aged 20 years or older (not included 20 years) and stayed in ICU for more than 6 h admitted to Taipei Veteran General Hospital (TPEVGH) between January 1, 2010, and March 31, 2020, of whom hemodynamic instability occurred in 3053 patients (prevalence = 19%). These patients in unstable group received at least one intervention during their ICU stays, and the HSI score of both stable and unstable group was calculated in every hour before intervention. The model performance was assessed using the area under the receiver operating characteristic curve (AUROC) and was compared to single indicators like systolic blood pressure (SBP) and shock index. The hemodynamic instability alarm was set by selecting optimal threshold with high sensitivity, acceptable specificity, and lead time before intervention was calculated to indicate when patients were firstly identified as high risk of hemodynamic instability.ResultsThe AUROC of HSI was 0.76 (95% CI, 0.75-0.77), which performed significantly better than shock Index (0.7; 95% CI, 0.69-0.71) and SBP (0.69; 95% CI, 0.68-0.70). By selecting 0.7 as a threshold, HSI predicted 72% of all 3053 patients who received hemodynamic interventions with 67% in specificity. Time-varying results also showed that HSI score significantly outperformed single indicators even up to 24 h before intervention. And 95% unstable patients can be identified more than 5 h in advance.ConclusionsThe HSI has acceptable discrimination but underestimates the risk of stable patients in predicting the onset of hemodynamic instability in an external cohort.

Project description:BackgroundAlthough there is a growing interest in prediction models based on electronic medical records (EMRs) to identify patients at risk of adverse cardiac events following invasive coronary treatment, robust models fully utilizing EMR data are limited.ObjectiveWe aimed to develop and validate machine learning (ML) models by using diverse fields of EMR to predict the risk of 30-day adverse cardiac events after percutaneous intervention or bypass surgery.MethodsEMR data of 5,184,565 records of 16,793 patients at a quaternary hospital between 2006 and 2016 were categorized into static basic (eg, demographics), dynamic time-series (eg, laboratory values), and cardiac-specific data (eg, coronary angiography). The data were randomly split into training, tuning, and testing sets in a ratio of 3:1:1. Each model was evaluated with 5-fold cross-validation and with an external EMR-based cohort at a tertiary hospital. Logistic regression (LR), random forest (RF), gradient boosting machine (GBM), and feedforward neural network (FNN) algorithms were applied. The primary outcome was 30-day mortality following invasive treatment.ResultsGBM showed the best performance with area under the receiver operating characteristic curve (AUROC) of 0.99; RF had a similar AUROC of 0.98. AUROCs of FNN and LR were 0.96 and 0.93, respectively. GBM had the highest area under the precision-recall curve (AUPRC) of 0.80, and the AUPRCs of RF, LR, and FNN were 0.73, 0.68, and 0.63, respectively. All models showed low Brier scores of <0.1 as well as highly fitted calibration plots, indicating a good fit of the ML-based models. On external validation, the GBM model demonstrated maximal performance with an AUROC of 0.90, while FNN had an AUROC of 0.85. The AUROCs of LR and RF were slightly lower at 0.80 and 0.79, respectively. The AUPRCs of GBM, LR, and FNN were similar at 0.47, 0.43, and 0.41, respectively, while that of RF was lower at 0.33. Among the categories in the GBM model, time-series dynamic data demonstrated a high AUROC of >0.95, contributing majorly to the excellent results.ConclusionsExploiting the diverse fields of the EMR data set, the ML-based 30-day adverse cardiac event prediction models demonstrated outstanding results, and the applied framework could be generalized for various health care prediction models.

Project description:ImportanceAccurate clinical decision support tools are needed to identify patients at risk for iatrogenic hypoglycemia, a potentially serious adverse event, throughout hospitalization.ObjectiveTo predict the risk of iatrogenic hypoglycemia within 24 hours after each blood glucose (BG) measurement during hospitalization using a machine learning model.Design, setting, and participantsThis retrospective cohort study, conducted at 5 hospitals within the Johns Hopkins Health System, included 54?978 admissions of 35?147 inpatients who had at least 4 BG measurements and received at least 1 U of insulin during hospitalization between December 1, 2014, and July 31, 2018. Data from the largest hospital were split into a 70% training set and 30% test set. A stochastic gradient boosting machine learning model was developed using the training set and validated on internal and external validation.ExposuresA total of 43 clinical predictors of iatrogenic hypoglycemia were extracted from the electronic medical record, including demographic characteristics, diagnoses, procedures, laboratory data, medications, orders, anthropomorphometric data, and vital signs.Main outcomes and measuresIatrogenic hypoglycemia was defined as a BG measurement less than or equal to 70 mg/dL occurring within the pharmacologic duration of action of administered insulin, sulfonylurea, or meglitinide.ResultsThis cohort study included 54?978 admissions (35?147 inpatients; median [interquartile range] age, 66.0 [56.0-75.0] years; 27?781 [50.5%] male; 30?429 [55.3%] White) from 5 hospitals. Of 1?612?425 index BG measurements, 50?354 (3.1%) were followed by iatrogenic hypoglycemia in the subsequent 24 hours. On internal validation, the model achieved a C statistic of 0.90 (95% CI, 0.89-0.90), a positive predictive value of 0.09 (95% CI, 0.08-0.09), a positive likelihood ratio of 4.67 (95% CI, 4.59-4.74), a negative predictive value of 1.00 (95% CI, 1.00-1.00), and a negative likelihood ratio of 0.22 (95% CI, 0.21-0.23). On external validation, the model achieved C statistics ranging from 0.86 to 0.88, positive predictive values ranging from 0.12 to 0.13, negative predictive values of 0.99, positive likelihood ratios ranging from 3.09 to 3.89, and negative likelihood ratios ranging from 0.23 to 0.25. Basal insulin dose, coefficient of variation of BG, and previous hypoglycemic episodes were the strongest predictors.Conclusions and relevanceThese findings suggest that iatrogenic hypoglycemia can be predicted in a short-term prediction horizon after each BG measurement during hospitalization. Further studies are needed to translate this model into a real-time informatics alert and evaluate its effectiveness in reducing the incidence of inpatient iatrogenic hypoglycemia.

Project description:ImportanceSystems-level barriers to diabetes care could be improved with population health planning tools that accurately discriminate between high- and low-risk groups to guide investments and targeted interventions.ObjectiveTo develop and validate a population-level machine learning model for predicting type 2 diabetes 5 years before diabetes onset using administrative health data.Design, setting, and participantsThis decision analytical model study used linked administrative health data from the diverse, single-payer health system in Ontario, Canada, between January 1, 2006, and December 31, 2016. A gradient boosting decision tree model was trained on data from 1 657 395 patients, validated on 243 442 patients, and tested on 236 506 patients. Costs associated with each patient were estimated using a validated costing algorithm. Data were analyzed from January 1, 2006, to December 31, 2016.ExposuresA random sample of 2 137 343 residents of Ontario without type 2 diabetes was obtained at study start time. More than 300 features from data sets capturing demographic information, laboratory measurements, drug benefits, health care system interactions, social determinants of health, and ambulatory care and hospitalization records were compiled over 2-year patient medical histories to generate quarterly predictions.Main outcomes and measuresDiscrimination was assessed using the area under the receiver operating characteristic curve statistic, and calibration was assessed visually using calibration plots. Feature contribution was assessed with Shapley values. Costs were estimated in 2020 US dollars.ResultsThis study trained a gradient boosting decision tree model on data from 1 657 395 patients (12 900 257 instances; 6 666 662 women [51.7%]). The developed model achieved a test area under the curve of 80.26 (range, 80.21-80.29), demonstrated good calibration, and was robust to sex, immigration status, area-level marginalization with regard to material deprivation and race/ethnicity, and low contact with the health care system. The top 5% of patients predicted as high risk by the model represented 26% of the total annual diabetes cost in Ontario.Conclusions and relevanceIn this decision analytical model study, a machine learning model approach accurately predicted the incidence of diabetes in the population using routinely collected health administrative data. These results suggest that the model could be used to inform decision-making for population health planning and diabetes prevention.

Dataset Information

Validation of a Machine Learning Model to Predict Childhood Lead Poisoning.

Publications

Validation of a Machine Learning Model to Predict Childhood Lead Poisoning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure