Machine Learning Prediction Models for Chronic Kidney Disease Using National Health Insurance Claim Data in Taiwan.
Ontology highlight
ABSTRACT:
SUBMITTER: Krishnamurthy S
PROVIDER: S-EPMC8151834 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Gene expression profiles were generated from 199 primary breast cancer patients. Samples 1-176 were used in another study, GEO Series GSE22820, and form the training data set in this study. Sample numbers 200-222 form a validation set. This data is used to model a machine learning classifier for Estrogen Receptor Status. RNA was isolated from 199 primary breast cancer patients. A machine learning classifier was built to predict ER status using only three gene features.
2013-01-01 | E-GEOD-29210 | biostudies-arrayexpress
Project description:BackgroundThere are many Machine Learning (ML) models which predict acute kidney injury (AKI) for hospitalised patients. While a primary goal of these models is to support clinical decision-making, the adoption of inconsistent methods of estimating baseline serum creatinine (sCr) may result in a poor understanding of these models' effectiveness in clinical practice. Until now, the performance of such models with different baselines has not been compared on a single dataset. Additionally, AKI prediction models are known to have a high rate of false positive (FP) events regardless of baseline methods. This warrants further exploration of FP events to provide insight into potential underlying reasons.ObjectiveThe first aim of this study was to assess the variance in performance of ML models using three methods of baseline sCr on a retrospective dataset. The second aim was to conduct an error analysis to gain insight into the underlying factors contributing to FP events.Materials and methodsThe Intensive Care Unit (ICU) patients of the Medical Information Mart for Intensive Care (MIMIC)-IV dataset was used with the KDIGO (Kidney Disease Improving Global Outcome) definition to identify AKI episodes. Three different methods of estimating baseline sCr were defined as (1) the minimum sCr, (2) the Modification of Diet in Renal Disease (MDRD) equation and the minimum sCr and (3) the MDRD equation and the mean of preadmission sCr. For the first aim of this study, a suite of ML models was developed for each baseline and the performance of the models was assessed. An analysis of variance was performed to assess the significant difference between eXtreme Gradient Boosting (XGB) models across all baselines. To address the second aim, Explainable AI (XAI) methods were used to analyse the XGB errors with Baseline 3.ResultsRegarding the first aim, we observed variances in discriminative metrics and calibration errors of ML models when different baseline methods were adopted. Using Baseline 1 resulted in a 14% reduction in the f1 score for both Baseline 2 and Baseline 3. There was no significant difference observed in the results between Baseline 2 and Baseline 3. For the second aim, the FP cohort was analysed using the XAI methods which led to relabelling data with the mean of sCr in 180 to 0 days pre-ICU as the preferred sCr baseline method. The XGB model using this relabelled data achieved an AUC of 0.85, recall of 0.63, precision of 0.54 and f1 score of 0.58. The cohort size was 31,586 admissions, of which 5,473 (17.32%) had AKI.ConclusionIn the absence of a widely accepted method of baseline sCr, AKI prediction studies need to consider the impact of different baseline methods on the effectiveness of ML models and their potential implications in real-world implementations. The utilisation of XAI methods can be effective in providing insight into the occurrence of prediction errors. This can potentially augment the success rate of ML implementation in routine care.
| S-EPMC10563357 | biostudies-literature
Project description:This study proposes a cardiovascular diseases (CVD) prediction model using machine learning (ML) algorithms based on the National Health Insurance Service-Health Screening datasets. We extracted 4699 patients aged over 45 as the CVD group, diagnosed according to the international classification of diseases system (I20-I25). In addition, 4699 random subjects without CVD diagnosis were enrolled as a non-CVD group. Both groups were matched by age and gender. Various ML algorithms were applied to perform CVD prediction; then, the performances of all the prediction models were compared. The extreme gradient boosting, gradient boosting, and random forest algorithms exhibited the best average prediction accuracy (area under receiver operating characteristic curve (AUROC): 0.812, 0.812, and 0.811, respectively) among all algorithms validated in this study. Based on AUROC, the ML algorithms improved the CVD prediction performance, compared to previously proposed prediction models. Preexisting CVD history was the most important factor contributing to the accuracy of the prediction model, followed by total cholesterol, low-density lipoprotein cholesterol, waist-height ratio, and body mass index. Our results indicate that the proposed health screening dataset-based CVD prediction model using ML algorithms is readily applicable, produces validated results and outperforms the previous CVD prediction models.
| S-EPMC8229422 | biostudies-literature
Project description:Nitrogen is the most limiting nutrient for turfgrass growth. Instead of pursuing the maximum yield, most turfgrass managers use nitrogen (N) to maintain a sub-maximal growth rate. Few tools or soil tests exist to help managers guide N fertilizer decisions. Turf growth prediction models have the potential to be useful, but the currently existing turf growth prediction model only takes temperature into account, limiting its accuracy. This study developed machine-learning-based turf growth models using the random forest (RF) algorithm to estimate short-term turfgrass clipping yield. To build the RF model, a large set of variables were extracted as predictors including the 7-day weather, traffic intensity, soil moisture content, N fertilization rate, and the normalized difference red edge (NDRE) vegetation index. In this study, the data were collected from two putting greens where the turfgrass received 0 to 1,800 round/week traffic rates, various irrigation rates to maintain the soil moisture content between 9 and 29%, and N fertilization rates of 0 to 17.5 kg ha-1 applied biweekly. The RF model agreed with the actual clipping yield collected from the experimental results. The temperature and relative humidity were the most important weather factors. Including NDRE improved the prediction accuracy of the model. The highest coefficient of determination (R2) of the RF model was 0.64 for the training dataset and was 0.47 for the testing data set upon the evaluation of the model. This represented a large improvement over the existing growth prediction model (R 2 = 0.01). However, the machine-learning models created were not able to accurately predict the clipping production at other locations. Individual golf courses can create customized growth prediction models using clipping volume to eliminate the deviation caused by temporal and spatial variability. Overall, this study demonstrated the feasibility of creating machine-learning-based yield prediction models that may be able to guide N fertilization decisions on golf course putting greens and presumably other turfgrass areas.
| S-EPMC8600360 | biostudies-literature
Project description:BackgroundAccurate and reliable predictions of infectious disease can be valuable to public health organizations that plan interventions to decrease or prevent disease transmission. A great variety of models have been developed for this task. However, for different data series, the performance of these models varies. Hepatitis E, as an acute liver disease, has been a major public health problem. Which model is more appropriate for predicting the incidence of hepatitis E? In this paper, three different methods are used and the performance of the three methods is compared.MethodsAutoregressive integrated moving average(ARIMA), support vector machine(SVM) and long short-term memory(LSTM) recurrent neural network were adopted and compared. ARIMA was implemented by python with the help of statsmodels. SVM was accomplished by matlab with libSVM library. LSTM was designed by ourselves with Keras, a deep learning library. To tackle the problem of overfitting caused by limited training samples, we adopted dropout and regularization strategies in our LSTM model. Experimental data were obtained from the monthly incidence and cases number of hepatitis E from January 2005 to December 2017 in Shandong province, China. We selected data from July 2015 to December 2017 to validate the models, and the rest was taken as training set. Three metrics were applied to compare the performance of models, including root mean square error(RMSE), mean absolute percentage error(MAPE) and mean absolute error(MAE).ResultsBy analyzing data, we took ARIMA(1, 1, 1), ARIMA(3, 1, 2) as monthly incidence prediction model and cases number prediction model, respectively. Cross-validation and grid search were used to optimize parameters of SVM. Penalty coefficient C and kernel function parameter g were set 8, 0.125 for incidence prediction, and 22, 0.01 for cases number prediction. LSTM has 4 nodes. Dropout and L2 regularization parameters were set 0.15, 0.001, respectively. By the metrics of RMSE, we obtained 0.022, 0.0204, 0.01 for incidence prediction, using ARIMA, SVM and LSTM. And we obtained 22.25, 20.0368, 11.75 for cases number prediction, using three models. For MAPE metrics, the results were 23.5%, 21.7%, 15.08%, and 23.6%, 21.44%, 13.6%, for incidence prediction and cases number prediction, respectively. For MAE metrics, the results were 0.018, 0.0167, 0.011 and 18.003, 16.5815, 9.984, for incidence prediction and cases number prediction, respectively.ConclusionsComparing ARIMA, SVM and LSTM, we found that nonlinear models(SVM, LSTM) outperform linear models(ARIMA). LSTM obtained the best performance in all three metrics of RSME, MAPE, MAE. Hence, LSTM is the most suitable for predicting hepatitis E monthly incidence and cases number.
| S-EPMC7497991 | biostudies-literature
Project description:BackgroundUrinary protein quantification is critical for assessing the severity of chronic kidney disease (CKD). However, the current procedure for determining the severity of CKD is completed through evaluating 24-h urinary protein, which is inconvenient during follow-up.ObjectiveTo quickly predict the severity of CKD using more easily available demographic and blood biochemical features during follow-up, we developed and compared several predictive models using statistical, machine learning and neural network approaches.MethodsThe clinical and blood biochemical results from 551 patients with proteinuria were collected. Thirteen blood-derived tests and 5 demographic features were used as non-urinary clinical variables to predict the 24-h urinary protein outcome response. Nine predictive models were established and compared, including logistic regression, Elastic Net, lasso regression, ridge regression, support vector machine, random forest, XGBoost, neural network and k-nearest neighbor. The AU-ROC, sensitivity (recall), specificity, accuracy, log-loss and precision of each of the models were evaluated. The effect sizes of each variable were analysed and ranked.ResultsThe linear models including Elastic Net, lasso regression, ridge regression and logistic regression showed the highest overall predictive power, with an average AUC and a precision above 0.87 and 0.8, respectively. Logistic regression ranked first, reaching an AUC of 0.873, with a sensitivity and specificity of 0.83 and 0.82, respectively. The model with the highest sensitivity was Elastic Net (0.85), while XGBoost showed the highest specificity (0.83). In the effect size analyses, we identified that ALB, Scr, TG, LDL and EGFR had important impacts on the predictability of the models, while other predictors such as CRP, HDL and SNA were less important.ConclusionsBlood-derived tests could be applied as non-urinary predictors during outpatient follow-up. Features in routine blood tests, including ALB, Scr, TG, LDL and EGFR levels, showed predictive ability for CKD severity. The developed online tool can facilitate the prediction of proteinuria progress during follow-up in clinical practice.
| S-EPMC6458616 | biostudies-literature
Project description:Chronic kidney disease (CKD) patients have high risks of end-stage kidney disease (ESKD) and pre-ESKD death. Therefore, accurately predicting these outcomes is useful among CKD patients, especially in those who are at high risk. Thus, we evaluated whether a machine-learning system can predict accurately these risks in CKD patients and attempted its application by developing a Web-based risk-prediction system. We developed 16 risk-prediction machine-learning models using Random Forest (RF), Gradient Boosting Decision Tree, and eXtreme Gradient Boosting with 22 variables or selected variables for the prediction of the primary outcome (ESKD or death) on the basis of repeatedly measured data of CKD patients (n = 3,714; repeatedly measured data, n = 66,981) in their electronic-medical records. The performances of the models were evaluated using data from a cohort study of CKD patients carried out over 3 years (n = 26,906). One RF model with 22 variables and another RF model with 8 variables of time-series data showed high accuracies of the prediction of the outcomes and were selected for use in a risk-prediction system. In the validation, the 22- and 8-variable RF models showed high C-statistics for the prediction of the outcomes: 0.932 (95% CI 0.916, 0.948) and 0.93 (0.915, 0.945), respectively. Cox proportional hazards models using splines showed a highly significant relationship between the high probability and high risk of an outcome (p<0.0001). Moreover, the risks of patients with high probabilities were higher than those with low probabilities: 22-variable model, hazard ratio of 104.9 (95% CI 70.81, 155.3); 8-variable model, 90.9 (95% CI 62.29, 132.7). Then, a Web-based risk-prediction system was actually developed for the implementation of the models in clinical practice. This study showed that a machine-learning-based Web system is a useful tool for the risk prediction and treatment of CKD patients. Author summary Chronic kidney disease (CKD) patients have high risks of end-stage kidney disease (ESKD) and pre-ESKD death. Although the development of a new artificial intelligence (AI) model is expected to be useful for screening CKD patients at high risk, none of the models developed so far have yet been put into practical use. There are many reasons for the difficulties in developing and applying AI models to clinical settings: (1) AI models often have too many variables, some of which are uncommon, to input in busy clinical settings. (2) The accuracy of AI models to predict risks is not always high when evaluating nontarget patients’ prognoses. (3) The codes of AI models do not always match electronic medical records of different hospitals. Therefore, in this study, AI models with a small number of commonly used variables were developed. Then, their performances were rigorously evaluated on subgroups of CKD patients classified on the basis of kidney function, age, and diabetes mellitus. Moreover, we developed and opened a Web-based risk-prediction system using the selected AI models that showed the highest accuracies for clinical practice: http://160.16.88.112:8000/. Our study fills the gap between AI development in research and AI in clinical practice.
| S-EPMC9931312 | biostudies-literature
Project description:Periodontitis is a widespread chronic inflammatory disease caused by interactions between periodontal bacteria and homeostasis in the host. We aimed to investigate the performance and reliability of machine learning models in predicting the severity of chronic periodontitis. Mouthwash samples from 692 subjects (144 healthy controls and 548 generalized chronic periodontitis patients) were collected, the genomic DNA was isolated, and the copy numbers of nine pathogens were measured using multiplex qPCR. The nine pathogens are as follows: Porphyromonas gingivalis (Pg), Tannerella forsythia (Tf), Treponema denticola (Td), Prevotella intermedia (Pi), Fusobacterium nucleatum (Fn), Campylobacter rectus (Cr), Aggregatibacter actinomycetemcomitans (Aa), Peptostreptococcus anaerobius (Pa), and Eikenella corrodens (Ec). By adding the species one by one in order of high accuracy to find the optimal combination of input features, we developed an algorithm that predicts the severity of periodontitis using four machine learning techniques. The accuracy was the highest when the models classified "healthy" and "moderate or severe" periodontitis (H vs. M-S, average accuracy of four models: 0.93, AUC = 0.96, sensitivity of 0.96, specificity of 0.81, and diagnostic odds ratio = 112.75). One or two red complex pathogens were used in three models to distinguish slight chronic periodontitis patients from healthy controls (average accuracy of 0.78, AUC = 0.82, sensitivity of 0.71, and specificity of 0.84, diagnostic odds ratio = 12.85). Although the overall accuracy was slightly reduced, the models showed reliability in predicting the severity of chronic periodontitis from 45 newly obtained samples. Our results suggest that a well-designed combination of salivary bacteria can be used as a biomarker for classifying between a periodontally healthy group and a chronic periodontitis group.
| S-EPMC7701273 | biostudies-literature
Project description:Background and purposeThe claims data of the Korean National Health Insurance (NHI) system can be useful in stroke research. The aim of this study was to validate the accuracy of hospital discharge data used for NHI claims in identifying acute stroke and use of thrombolytic therapy.MethodsThe hospital discharge data of 1,811 patients with stroke-related diagnosis codes were obtained from Jeju National University Hospital (JNUH) and Seoul Medical Center (SMC). Three algorithms were tested to identify discharges with acute stroke [ischemic stroke (IS), intracranial hemorrhage (ICH), or subarachnoid hemorrhage (SAH)]: 1) all diagnosis codes up to nine positions, 2) one primary diagnosis and one secondary diagnosis, and 3) only one primary diagnosis code. Reviews of medical records were considered the gold standards.ResultsOverall, the degree of agreement (?) was higher for algorithms 1 and 2 than for algorithm 3, and the sensitivity and specificity of the first two algorithms for IS and SAH were both >90%, with almost perfect agreement (?=0.83-0.84) in the JNUH data set. Regarding ICH, only algorithm 1 yielded an almost perfect agreement (?=0.82). In the SMC data set, almost perfect agreement was found for both ICH and SAH in all three algorithms. In contrast, the three algorithms yielded a range of agreement levels, though all substantial, for IS. Almost perfect agreement was obtained for use of thrombolytic therapy in both data sets (?=0.91-0.99).ConclusionsDischarge with hemorrhagic stroke and use of thrombolytic therapy were identified with high reliability in administrative discharge data. A substantial level of agreement was also obtained for IS, despite variation between the algorithms and data sets.
| S-EPMC4712285 | biostudies-literature
Project description:Uncertainty quantification in artificial intelligence (AI)-based predictions of material properties is of immense importance for the success and reliability of AI applications in materials science. While confidence intervals are commonly reported for machine learning (ML) models, prediction intervals, i.e., the evaluation of the uncertainty on each prediction, are not as frequently available. In this work, we compare three different approaches to obtain such individual uncertainty, testing them on 12 ML-physical properties. Specifically, we investigated using the quantile loss function, machine learning the prediction intervals directly, and using Gaussian processes. We identify each approach's advantages and disadvantages and end up slightly favoring the modeling of the individual uncertainties directly, as it is the easiest to fit and, in most of the cases, minimizes over- and underestimation of the predicted errors. All data for training and testing were taken from the publicly available JARVIS-DFT database, and the codes developed for computing the prediction intervals are available through the JARVIS-tools package.
| S-EPMC8655759 | biostudies-literature