Using machine learning to predict rapid decline of kidney function in sickle cell anemia.
Ontology highlight
ABSTRACT:
SUBMITTER: Gunturkun F
PROVIDER: S-EPMC9176130 | biostudies-literature | 2021 May
REPOSITORIES: biostudies-literature
Publications
Using machine learning to predict rapid decline of kidney function in sickle cell anemia.
Güntürkün Fatma F Chen Daiqing D Akbilgic Oguz O Davis Robert L RL Karabayir Ibrahim I Strome Maxwell M Dai Yang Y Saraf Santosh L SL Ataga Kenneth I KI
EJHaem 20210210 2
Similar Datasets
Project description:BackgroundTo develop a reliable model to predict rapid kidney function decline (RKFD) among population at risk of cardiovascular disease.MethodsIn this retrospective study, key monitoring residents including the elderly, and patients with hypertension or diabetes of China National Basic Public Health Service who underwent community annual physical examinations from January 2015 to December 2020 were included. Healthy records were extracted from regional chronic disease management platform. RKFD was defined as the reduction of estimated glomerular filtration rate (eGFR) ≥ 40% during follow-up period. The entire cohort were randomly assigned to a development cohort and a validation cohort in a 2:1 ratio. Cox regression analysis was used to identify the independent predictors. A nomogram was established based on the development cohort. The concordance index (C-index) and calibration plots were calculated. Decision curve analysis was applied to evaluate the clinical utility.ResultsA total of 8455 subjects were included. During the median follow-up period of 3.72 years, the incidence of RKFD was 11.96% (n = 1011), 11.98% (n = 676) and 11.92% (n = 335) in the entire cohort, development cohort and validation cohort, respectively. Age, eGFR, hemoglobin, systolic blood pressure, and diabetes were identified as predictors for RKFD. Good discriminating performance was observed in both the development (C-index, 0.73) and the validation (C-index, 0.71) cohorts, and the AUCs for predicting 5-years RKFD was 0.763 and 0.740 in the development and the validation cohort, respectively. Decision curve analysis further confirmed the clinical utility of the nomogram.ConclusionsOur nomogram based on five readily accessible variables (age, eGFR, hemoglobin, systolic blood pressure, and diabetes) is a useful tool to identify high risk patients for RKFD among population at risk of cardiovascular disease in primary care. Whereas, further external validations are needed before clinical generalization.
| S-EPMC8830119 | biostudies-literature
Project description:Due to the increasing prevalence of chronic kidney disease and its high mortality rate, study of risk factors affecting the progression of the disease is of great importance. Here in this work, we aim to develop a framework for using machine learning methods to identify factors affecting kidney function. To this end classification methods are trained to predict the serum creatinine level based on numerical values of other blood test parameters in one of the three classes representing different ranges of the variable values. Models are trained using the data from blood test results of healthy and patient subjects including 46 different blood test parameters. The best developed models are random forest and LightGBM. Interpretation of the resulting model reveals a direct relationship between vitamin D and blood creatinine level. The detected analogy between these two parameters is reliable, regarding the relatively high predictive accuracy of the random forest model reaching the AUC of 0.90 and the accuracy of 0.74. Moreover, in this paper we develop a Bayesian network to infer the direct relationships between blood test parameters which have consistent results with the classification models. The proposed framework uses an inclusive set of advanced imputation methods to deal with the main challenge of working with electronic health data, missing values. Hence it can be applied to similar clinical studies to investigate and discover the relationships between the factors under study.
| S-EPMC9758148 | biostudies-literature
Project description:We used an ensemble of statistical methods to build a model that predicts kidney transplant survival and identifies important predictive variables. The proposed model achieved better performance, measured by Harrell's concordance index, than the Estimated Post Transplant Survival model used in the kidney allocation system in the U.S., and other models published recently in the literature. The model has a five-year concordance index of 0.724 (in comparison, the concordance index is 0.697 for the Estimated Post Transplant Survival model, the state of the art currently in use). It combines predictions from random survival forests with a Cox proportional hazards model. The rankings of importance for the model's variables differ by transplant recipient age. Better survival predictions could eventually lead to more efficient allocation of kidneys and improve patient outcomes.
| S-EPMC6326487 | biostudies-literature
Project description:ObjectivesGenetic risks for cognitive decline are not modifiable; however their relative importance compared to modifiable factors is unclear. We used machine learning to evaluate modifiable and genetic risk factors for Alzheimer's disease (AD), to predict cognitive decline.MethodsHealth and Retirement Study participants, aged 65-90 years, with DNA and >2 cognitive evaluations, were included (n = 7,142). Predictors included age, body mass index, gender, education, APOE ε4, cardiovascular, hypertension, diabetes, stroke, neighborhood socioeconomic status (NSES), and AD risk genes. Latent class trajectory analyses of cognitive scores determined the form and number of classes. Random Forests (RF) classification investigated predictors of cognitive trajectories. Performance metrics (accuracy, sensitivity, and specificity) were reported.ResultsThree classes were identified. Discriminating highest from lowest classes produced the best RF performance: accuracy = 78% (1.0%), sensitivity = 75% (1.0%), and specificity = 81% (1.0%). Top ranked predictors were education, age, gender, stroke, NSES, and diabetes, APOE ε4 carrier status, and body mass index (BMI). When discriminating high from medium classes, top predictors were education, age, gender, stroke, diabetes, NSES, and BMI. When discriminating medium from the low classes, education, NSES, age, diabetes, and stroke were top predictors.DiscussionThe combination of latent trajectories and RF classification techniques suggested that nongenetic factors contribute more to cognitive decline than genetic factors. Education was the most relevant predictor for discrimination.
| S-EPMC7205421 | biostudies-literature
Project description:The purpose of this study was to assess the feasibility of machine learning (ML) in predicting the risk of end-stage kidney disease (ESKD) from patients with chronic kidney disease (CKD). Data were obtained from a longitudinal CKD cohort. Predictor variables included patients' baseline characteristics and routine blood test results. The outcome of interest was the presence or absence of ESKD by the end of 5 years. Missing data were imputed using multiple imputation. Five ML algorithms, including logistic regression, naïve Bayes, random forest, decision tree, and K-nearest neighbors were trained and tested using fivefold cross-validation. The performance of each model was compared to that of the Kidney Failure Risk Equation (KFRE). The dataset contained 748 CKD patients recruited between April 2006 and March 2008, with the follow-up time of 6.3 ± 2.3 years. ESKD was observed in 70 patients (9.4%). Three ML models, including the logistic regression, naïve Bayes and random forest, showed equivalent predictability and greater sensitivity compared to the KFRE. The KFRE had the highest accuracy, specificity, and precision. This study showed the feasibility of ML in evaluating the prognosis of CKD based on easily accessible features. Three ML models with adequate performance and sensitivity scores suggest a potential use for patient screenings. Future studies include external validation and improving the models with additional predictor variables.
| S-EPMC9120106 | biostudies-literature
Project description:The African American Study of Kidney Disease and Hypertension (AASK), a randomized double-blinded treatment trial, was motivated by the high rate of hypertension-related renal disease in the African-American population and the scarcity of effective therapies. This study describes a pattern-based classification approach to predict the rate of decline of kidney function using surface-enhanced laser desorption ionization/time of flight proteomic data from rapid and slow progressors classified by rate of change in glomerular filtration rate. An accurate classification model consisting of 7 out of 5,751 serum proteomic features is constructed by applying the logical analysis of data (LAD) methodology. On cross-validation by 10-folding, the model was shown to have an accuracy of 80.6 ± 0.11%, sensitivity of 78.4 ± 0.17%, and specificity of 78.5 ± 0.16%. The LAD discriminant is used to identify the patients in different risk groups. The LAD risk scores assigned to 116 AASK patients generated a receiver operating curves curve with AUC 0.899 (CI 0.845-0.953) and outperforms the risk scores assigned by proteinuria, one of the best predictors of chronic kidney disease progression.
| S-EPMC5516355 | biostudies-other
Project description:BACKGROUND:While elevated serum uric acid level (SUA) is a recognized risk factor for chronic kidney disease, it remains unclear whether change in SUA is independently associated with change in estimated glomerular filtration rate (eGFR) over time. Accordingly, we examined the longitudinal associations between change in SUA and change in eGFR over 5 years in a general Japanese population. METHODS:This was a large, single-center, retrospective 5-year cohort study at St. Luke's International Hospital, Tokyo, Japan, between 2004 and 2009. We included 13,070 subjects (30-85 years) in our analyses whose data were available between 2004 and 2009. Of those, we excluded 492 subjects with eGFR <60 mL/min/1.73 m2 at baseline. In addition to examining the entire cohort (n = 12,578), we stratified our analyses by baseline eGFR groups: 60-90, 90-120, and ?120 mL/min/1.73 m2. Linear and logistic regressions models were applied to examine the relationships between baseline and change in SUA, change in eGFR, and rapid eGFR decline (defined as the highest quartile of change in eGFR), adjusted for age, gender, body mass index, abdominal circumference, hypertension, dyslipidemia, and diabetes mellitus. RESULTS:After multivariable adjustments including baseline eGFR, 1 mg/dL increase in baseline SUA was associated with greater odds of developing rapid eGFR decline (OR 1.27, 95% CI 1.17-1.38), and 1 mg/dL increase in SUA over 5 years was associated with 3.77-fold greater odds of rapid eGFR decline (OR 3.77, 95% CI 3.35-4.26). CONCLUSIONS:Elevated baseline SUA and increasing SUA over time were independent risk factors for rapid eGFR decline over 5 years.
| S-EPMC5894505 | biostudies-literature
Project description:Chronic kidney disease (CKD), defined at a specific time point, is an important risk factor for cardiovascular disease. Whether the rate of kidney function decline contributes additional cardiovascular risk is unknown. In the Cardiovascular Health Study, we compared the associations of changes in kidney function during the first 7 yr with the incidence of heart failure (HF), myocardial infarction (MI), stroke, and peripheral arterial disease (PAD) during the subsequent 8 yr. We defined a rapid decline in cystatin C-based estimated GFR as >3 ml/min per 1.73 m(2)/yr, on the basis of determination at baseline, year 3, and year 7. Among eligible participants, 1083 (24%) had rapid kidney decline. The incidence of each type of cardiovascular event was significantly higher among patients with rapid decline (all P < 0.001). After multivariate adjustment for demographics, cardiovascular disease risk factors, and baseline kidney function, rapid kidney function decline was significantly associated with HF (adjusted hazard ratio [HR] 1.32; 95% confidence interval [CI] 1.13 to 1.53), MI (HR 1.48; 95% CI 1.21 to 1.83), and PAD (HR 1.67; 95% CI 1.02 to 2.75) but not with stroke (HR 1.19; 95% CI 0.97 to 1.45). The association of rapid decline with each outcome did not differ by the presence or absence of CKD. In conclusion, declining kidney function associates with higher risk for HF, MI, and PAD among patients with or without CKD.
| S-EPMC2794238 | biostudies-literature
Project description:Rationale & objectiveThe direct and indirect effects of the coronavirus disease 2019 (COVID-19) pandemic on kidney function in the chronic kidney disease (CKD) population are not well understood.Study designCohort study.Setting & participantsRetrospective study of kidney function trajectories using deidentified administrative claims and laboratory data for Medicare Advantage and commercially insured enrollees with CKD stages G3-4 between 2018 and 2021.PredictorsCOVID-19 infection.OutcomeRapid kidney function decline defined as annual estimated glomerular filtration rate (eGFR) decline of ≥40%.Analytical approachPropensity score matching was used to identify individuals without COVID-19 infection matched 1:1 to a COVID-19 infected cohort and indexed on the date of diagnosing COVID-19 infection, age, sex, race or ethnicity, and Charlson comorbidity index score. Outpatient kidney function was compared during the prepandemic period (January 1, 2018, to February 29, 2020) with the pandemic period (March 1, 2020, to August 31, 2021). Two creatinine measurements, after the infection date and ≥60 days apart, were required to reduce correlation with acute infection.ResultsOf 97,203 enrollees with CKD G3-4, 9% experienced a COVID-19 infection. Characteristics of 8,901 propensity matched enrollees include mean age 74 years, 58% women, 67% White, and 63% CKD G3a, 28% CKD G3b, and 9% CKD G4. Median overall annual eGFR change was -2.65 ml/min/1.73m2, with 76% of the cohort experiencing worsened eGFR in the pandemic period. Rapid kidney function decline was observed in 1.9% and 2.0% of enrollees in the prepandemic and pandemic periods, respectively. Rapid kidney function decline was observed in 2.5% of those with COVID-19 infection and 1.5% of those without COVID-19 infection (P < 0.05). Factors associated with increased odds of rapid kidney function decline during pandemic included Asian race, higher Charlson comorbidity index, advancing CKD stage, prepandemic rapid kidney function decline, and COVID-19 infection.LimitationsRetrospective study design with potential bias.ConclusionsCOVID-19 infection increased odds of rapid kidney function decline during the pandemic. The downstream impact of pandemic-related eGFR decline on health outcomes, such as kidney failure or mortality, requires further study.Plain-language summaryWe used a cohort of insured individuals with moderate-to-severe chronic kidney disease (CKD) to compare the rates of rapid kidney function decline in prepandemic and pandemic periods and to evaluate the impact of the coronavirus disease 19 (COVID-19) on kidney function decline. We found that overall rates of rapid kidney function decline did not change during the prepandemic and pandemic periods but were significantly higher in both periods among individuals with a COVID-19 infection. As CKD severity increased, rates of both rapid kidney function decline and COVID-19 increased. Advancing CKD, higher comorbid condition, Asian race, prepandemic rapid kidney function decline, and COVID-19 were all associated with higher odds of rapid kidney function decline in the pandemic. These findings suggest close monitoring is warranted for individuals with CKD and COVID-19.
| S-EPMC10462879 | biostudies-literature
Project description:BackgroundFibroblast growth factor-23 (FGF-23) associates with decreased kidney function in patients with chronic kidney disease (CKD). However, the correlation between circulating FGF-23 levels and the rate of renal function decline in healthy individuals is largely unknown. We aimed to evaluate the predictive performance of FGF-23 for rapid kidney function decline (RKFD) in a community-based study.MethodsA total of 2963 people residing in northern Taiwan were enrolled from August 2013 to May 2018 for an annual assessment of kidney function for five years. The baseline estimated glomerular filtration rates (eGFR) were calculated using the 2009 and 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation, which aggregates the values of serum creatinine and cystatin C (eGFRcr-cys). The outcome was RKFD-a 15% decrease in estimated glomerular filtration rate (eGFR) within the first four years, and a reduction in eGFR without improvement in the 5th year. A generalized additive model (GAM) was used to determine the cut-off value of FGF-23 to predict RKFD.ResultsThe incidence of RKFD was 18.0% (114/634). After matching for age and sex at a 1:1 ratio, a total of 220 subjects were analyzed. eGFRcr-cys was negatively correlated with total vitamin D level but seemed irrelevant to FGF-23. Multivariable logistic regression analysis showed that FGF-23, eGFRcr-cys, and urine albumin-to-creatinine ratio (UACR) were independent predictors of the possibility of RKFD. FGF-23 showed the best predictive performance for RKFD (AUROC: 0.803), followed by baseline eGFRcr-cys (AUROC: 0.639) and UACR (AUROC: 0.591). From the GAM, 32 pg/mL was the most appropriate cut-off value of FGF-23 with which to predict RKFD. The subgroup and sensitivity analyses showed consistent results that high-FGF-23 subjects had higher risks of RKFD.ConclusionsCirculating FGF-23 level could be a helpful predictor for RKFD in this community-based population.
| S-EPMC9856057 | biostudies-literature